+
+
+
+
+
+
+
+
+
+
+Colorado Spills Project¶
Outline¶
-
+
Data Import and Preparation
+-
+
- Import necessary libraries and load data from the database. +
- Convert data into GeoDataFrame and ensure proper formatting. +
+Data Cleaning
+-
+
- Handle missing or incorrect data. +
- Filter out data for the year 2024. +
+Exploratory Data Analysis
+-
+
- Separate historical and recent spills. +
- Analyze reporting delays and trends over time. +
+Visualization
+-
+
- Create spatial distribution maps. +
- Plot trends and differences in reporting delays. +
+Statistical Analysis
+-
+
- Perform t-tests to compare response times. +
- Use regression models to analyze differences and trends. +
+Results and Interpretation
+-
+
- Summarize findings in markdown cells. +
- Highlight key takeaways and policy implications. +
+
-
In [1]:
+In [3]:
import pandas as pd
@@ -7565,7 +7613,7 @@ a.anchor-link {
-In [2]:
+In [4]:
# use longitude and latitude to create a GeoDataFrame from spills data
@@ -7580,7 +7628,7 @@ a.anchor-link {
-In [3]:
+In [5]:
spills_gdf = gpd.GeoDataFrame(spills, crs='EPSG:4326')
@@ -7594,7 +7642,7 @@ a.anchor-link {
-In [4]:
+In [6]:
# Ensure date columns are in datetime format
@@ -7610,7 +7658,7 @@ a.anchor-link {
-In [5]:
+In [7]:
# drop 2024 data for date of discovery and initial report date
@@ -7626,7 +7674,7 @@ a.anchor-link {
-In [6]:
+In [8]:
# Separate Historical vs. Recent Spills
@@ -7642,7 +7690,7 @@ a.anchor-link {
-In [7]:
+In [9]:
# Compare Reporting Delay
@@ -7683,7 +7731,7 @@ Recent 10678.0 3.964506 46.997749 0.0 0.0 1.0 2.0 2232.0
-In [8]:
+In [10]:
import matplotlib.pyplot as plt
@@ -7726,7 +7774,7 @@ Recent 10678.0 3.964506 46.997749 0.0 0.0 1.0 2.0 2232.0
-In [9]:
+In [11]:
import matplotlib.pyplot as plt
@@ -7781,7 +7829,7 @@ Recent 10678.0 3.964506 46.997749 0.0 0.0 1.0 2.0 2232.0
-In [10]:
+In [12]:
import matplotlib.pyplot as plt
@@ -7859,7 +7907,7 @@ Recent 10678.0 3.964506 46.997749 0.0 0.0 1.0 2.0 2232.0
-
+
@@ -7869,7 +7917,7 @@ Recent 10678.0 3.964506 46.997749 0.0 0.0 1.0 2.0 2232.0
-In [11]:
+In [13]:
# Add a column to indicate whether the spill occurred before or after 2020
@@ -7913,7 +7961,7 @@ Before 2020 8813.0 12.117554 124.705876 0.0 0.0 1.0 2.0 5681.0
-In [12]:
+In [14]:
# t-test for the difference in response time before and after 2020
@@ -7956,7 +8004,7 @@ The difference in response time is statistically significant.
-Out[12]:
+Out[14]:
<Axes: title={'center': 'Report Delay (Days)'}, xlabel='Period'>
@@ -7974,41 +8022,7 @@ The difference in response time is statistically significant.
-In [13]:
-
-
-%pip install tabulate
-
-
-
-
-
-
-
-
-
-
-
-
-Requirement already satisfied: tabulate in /home/dadams/miniconda3/envs/spatial_env2/lib/python3.10/site-packages (0.9.0)
-
-
-
-
-
-
-Note: you may need to restart the kernel to use updated packages.
-
-
-
-
-
-
-
-
-
-
-In [14]:
+In [15]:
import tabulate
@@ -8182,7 +8196,7 @@ See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stab
-In [15]:
+In [16]:
import os
@@ -8257,7 +8271,7 @@ See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stab
-In [16]:
+In [17]:
# t-test for the difference in response time before and after 2020 for historical spills
@@ -8335,10 +8349,92 @@ The difference in response time for recent spills is statistically significant.
-In [ ]:
+In [19]:
-# Historical only
+import pandas as pd
+import statsmodels.formula.api as smf
+
+# Combine the datasets and label the groups
+historical_spills['SpillType'] = 0 # historical = 0
+recent_spills['SpillType'] = 1 # recent = 1
+
+combined = pd.concat([historical_spills, recent_spills], ignore_index=True)
+
+# Convert 'Period' to binary: 0 = Before 2020, 1 = 2020 and After
+combined['Post2020'] = combined['Period'].apply(lambda x: 1 if x == '2020 and After' else 0)
+
+# Interaction term is created implicitly in the regression
+model = smf.ols('Q("Report Delay (Days)") ~ Post2020 + SpillType + Post2020:SpillType', data=combined).fit()
+
+print(model.summary())
+
+
+
+
+
+
+
+
+
+
+
+
+ OLS Regression Results
+====================================================================================
+Dep. Variable: Q("Report Delay (Days)") R-squared: 0.006
+Model: OLS Adj. R-squared: 0.006
+Method: Least Squares F-statistic: 31.12
+Date: Mon, 31 Mar 2025 Prob (F-statistic): 4.78e-20
+Time: 17:54:14 Log-Likelihood: -99827.
+No. Observations: 15990 AIC: 1.997e+05
+Df Residuals: 15986 BIC: 1.997e+05
+Df Model: 3
+Covariance Type: nonrobust
+======================================================================================
+ coef std err t P>|t| [0.025 0.975]
+--------------------------------------------------------------------------------------
+Intercept 29.7461 2.438 12.200 0.000 24.967 34.525
+Post2020 -16.2130 3.417 -4.745 0.000 -22.910 -9.516
+SpillType -25.0339 2.906 -8.616 0.000 -30.729 -19.339
+Post2020:SpillType 14.4276 4.200 3.435 0.001 6.196 22.660
+==============================================================================
+Omnibus: 45995.174 Durbin-Watson: 1.915
+Prob(Omnibus): 0.000 Jarque-Bera (JB): 3525625525.882
+Skew: 38.959 Prob(JB): 0.00
+Kurtosis: 2302.059 Cond. No. 8.21
+==============================================================================
+
+Notes:
+[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
+
+
+
+
+
+
+/home/dadams/miniconda3/envs/spatial_env2/lib/python3.10/site-packages/geopandas/geodataframe.py:1819: SettingWithCopyWarning:
+A value is trying to be set on a copy of a slice from a DataFrame.
+Try using .loc[row_indexer,col_indexer] = value instead
+
+See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
+ super().__setitem__(key, value)
+
+
+
+
+
+
+
+
+
+
+In [21]:
+
+
+import statsmodels.formula.api as smf
+
+# Historical only
historical_spills = historical_spills.copy()
historical_spills['Post2020'] = historical_spills['Period'].apply(lambda x: 1 if x == '2020 and After' else 0)
model_hist = smf.ols('Q("Report Delay (Days)") ~ Post2020', data=historical_spills).fit()
@@ -8368,7 +8464,7 @@ Dep. Variable: Q("Report Delay (Days)") R-squared: 0
Model: OLS Adj. R-squared: 0.001
Method: Least Squares F-statistic: 8.265
Date: Mon, 31 Mar 2025 Prob (F-statistic): 0.00406
-Time: 13:58:43 Log-Likelihood: -35825.
+Time: 17:54:40 Log-Likelihood: -35825.
No. Observations: 5312 AIC: 7.165e+04
Df Residuals: 5310 BIC: 7.167e+04
Df Model: 1
@@ -8395,7 +8491,7 @@ Dep. Variable: Q("Report Delay (Days)") R-squared: 0
Model: OLS Adj. R-squared: 0.000
Method: Least Squares F-statistic: 3.752
Date: Mon, 31 Mar 2025 Prob (F-statistic): 0.0528
-Time: 13:58:43 Log-Likelihood: -56260.
+Time: 17:54:40 Log-Likelihood: -56260.
No. Observations: 10678 AIC: 1.125e+05
Df Residuals: 10676 BIC: 1.125e+05
Df Model: 1
@@ -8424,7 +8520,7 @@ Notes:
-In [ ]:
+In [22]:
# Label both groups, combine
@@ -8455,7 +8551,7 @@ Dep. Variable: Q("Report Delay (Days)") R-squared: 0
Model: OLS Adj. R-squared: 0.006
Method: Least Squares F-statistic: 31.12
Date: Mon, 31 Mar 2025 Prob (F-statistic): 4.78e-20
-Time: 13:53:41 Log-Likelihood: -99827.
+Time: 17:54:47 Log-Likelihood: -99827.
No. Observations: 15990 AIC: 1.997e+05
Df Residuals: 15986 BIC: 1.997e+05
Df Model: 3
@@ -8486,7 +8582,7 @@ Notes:
-In [20]:
+In [23]:
import matplotlib.pyplot as plt
@@ -8596,7 +8692,7 @@ Notes:
-In [22]:
+In [24]:
import pandas as pd
@@ -8633,7 +8729,7 @@ Dep. Variable: Q("Report Delay (Days)") R-squared: 0
Model: OLS Adj. R-squared: 0.006
Method: Least Squares F-statistic: 31.12
Date: Mon, 31 Mar 2025 Prob (F-statistic): 4.78e-20
-Time: 14:02:43 Log-Likelihood: -99827.
+Time: 17:54:59 Log-Likelihood: -99827.
No. Observations: 15990 AIC: 1.997e+05
Df Residuals: 15986 BIC: 1.997e+05
Df Model: 3
@@ -8715,7 +8811,7 @@ Notes:
-In [24]:
+In [25]:
import matplotlib.pyplot as plt
@@ -8764,7 +8860,7 @@ Notes:
-In [25]:
+In [26]:
import matplotlib.pyplot as plt
@@ -8810,6 +8906,372 @@ Notes:
+
+
+
+
+
+In [28]:
+
+
+# Only keep pre-2020 years to check the parallel trends assumption
+pre_2020 = spills_gdf[spills_gdf['Report Year'] < 2020]
+
+# Group by year and spill type
+trend_df = pre_2020.groupby(['Report Year', 'Spill Type'])['Report Delay (Days)'].mean().reset_index()
+
+
+
+
+
+
+
+
+
+
+In [29]:
+
+
+import matplotlib.pyplot as plt
+
+fig, ax = plt.subplots(figsize=(10, 6))
+
+for spill_type in trend_df['Spill Type'].unique():
+ subset = trend_df[trend_df['Spill Type'] == spill_type]
+ ax.plot(subset['Report Year'], subset['Report Delay (Days)'], marker='o', label=spill_type)
+
+ax.set_title('Pre-2020 Trends in Report Delay by Spill Type')
+ax.set_xlabel('Report Year')
+ax.set_ylabel('Mean Report Delay (Days)')
+ax.legend()
+ax.grid(True)
+plt.tight_layout()
+plt.show()
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+In [32]:
+
+
+spills_gdf['Report Year'] = pd.to_datetime(spills_gdf['Initial Report Date']).dt.year
+
+
+
+
+
+
+
+
+
+
+In [ ]:
+
+
+import statsmodels.formula.api as smf
+
+model_fe = smf.ols(
+ 'Q("Report Delay (Days)") ~ Post2020 * Group + C(Q("Report Year"))',
+ data=combined
+).fit()
+
+print(model_fe.summary())
+
+
+
+
+
+
+
+
+
+
+
+
+ OLS Regression Results
+====================================================================================
+Dep. Variable: Q("Report Delay (Days)") R-squared: 0.010
+Model: OLS Adj. R-squared: 0.009
+Method: Least Squares F-statistic: 14.56
+Date: Mon, 31 Mar 2025 Prob (F-statistic): 1.73e-28
+Time: 18:04:09 Log-Likelihood: -99794.
+No. Observations: 15990 AIC: 1.996e+05
+Df Residuals: 15978 BIC: 1.997e+05
+Df Model: 11
+Covariance Type: nonrobust
+===============================================================================================
+ coef std err t P>|t| [0.025 0.975]
+-----------------------------------------------------------------------------------------------
+Intercept 24.2312 4.880 4.965 0.000 14.665 33.797
+Group[T.Recent] -25.6229 3.295 -7.777 0.000 -32.081 -19.165
+C(Q("Report Year"))[T.2015] 24.6495 5.413 4.554 0.000 14.040 35.260
+C(Q("Report Year"))[T.2016] 2.0193 5.195 0.389 0.698 -8.164 12.203
+C(Q("Report Year"))[T.2017] -0.9935 4.921 -0.202 0.840 -10.640 8.653
+C(Q("Report Year"))[T.2018] -4.1817 5.083 -0.823 0.411 -14.145 5.781
+C(Q("Report Year"))[T.2019] 11.6665 4.963 2.351 0.019 1.939 21.394
+C(Q("Report Year"))[T.2020] -3.392e+13 1.55e+14 -0.218 0.827 -3.39e+14 2.71e+14
+C(Q("Report Year"))[T.2021] -3.392e+13 1.55e+14 -0.218 0.827 -3.39e+14 2.71e+14
+C(Q("Report Year"))[T.2022] -3.392e+13 1.55e+14 -0.218 0.827 -3.39e+14 2.71e+14
+C(Q("Report Year"))[T.2023] -3.392e+13 1.55e+14 -0.218 0.827 -3.39e+14 2.71e+14
+Post2020 3.392e+13 1.55e+14 0.218 0.827 -2.71e+14 3.39e+14
+Post2020:Group[T.Recent] 13.8374 4.229 3.272 0.001 5.547 22.127
+==============================================================================
+Omnibus: 46016.971 Durbin-Watson: 1.922
+Prob(Omnibus): 0.000 Jarque-Bera (JB): 3554990757.140
+Skew: 39.006 Prob(JB): 0.00
+Kurtosis: 2311.622 Cond. No. 4.97e+14
+==============================================================================
+
+Notes:
+[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
+[2] The smallest eigenvalue is 1.28e-25. This might indicate that there are
+strong multicollinearity problems or that the design matrix is singular.
+
+
+
+
+
+
+
+
+
+
+In [36]:
+
+
+model_fe_fixed = smf.ols(
+ 'Q("Report Delay (Days)") ~ Group * C(Q("Report Year"))',
+ data=combined
+).fit()
+
+print(model_fe_fixed.summary())
+
+
+
+
+
+
+
+
+
+
+
+
+ OLS Regression Results
+====================================================================================
+Dep. Variable: Q("Report Delay (Days)") R-squared: 0.016
+Model: OLS Adj. R-squared: 0.015
+Method: Least Squares F-statistic: 13.78
+Date: Mon, 31 Mar 2025 Prob (F-statistic): 3.21e-44
+Time: 18:04:52 Log-Likelihood: -99743.
+No. Observations: 15990 AIC: 1.995e+05
+Df Residuals: 15970 BIC: 1.997e+05
+Df Model: 19
+Covariance Type: nonrobust
+===============================================================================================================
+ coef std err t P>|t| [0.025 0.975]
+---------------------------------------------------------------------------------------------------------------
+Intercept 12.6921 6.981 1.818 0.069 -0.992 26.376
+Group[T.Recent] -9.5452 8.333 -1.146 0.252 -25.878 6.788
+C(Q("Report Year"))[T.2015] 70.1325 8.870 7.907 0.000 52.747 87.518
+C(Q("Report Year"))[T.2016] 8.0517 9.141 0.881 0.378 -9.865 25.969
+C(Q("Report Year"))[T.2017] -2.4258 9.115 -0.266 0.790 -20.293 15.441
+C(Q("Report Year"))[T.2018] -8.4734 8.828 -0.960 0.337 -25.777 8.830
+C(Q("Report Year"))[T.2019] 28.6915 9.529 3.011 0.003 10.013 47.370
+C(Q("Report Year"))[T.2020] 33.1820 10.449 3.176 0.001 12.700 53.664
+C(Q("Report Year"))[T.2021] -10.6444 8.550 -1.245 0.213 -27.404 6.115
+C(Q("Report Year"))[T.2022] 5.0125 8.157 0.615 0.539 -10.976 21.000
+C(Q("Report Year"))[T.2023] -3.9384 8.048 -0.489 0.625 -19.713 11.836
+Group[T.Recent]:C(Q("Report Year"))[T.2015] -65.2606 10.635 -6.136 0.000 -86.107 -44.415
+Group[T.Recent]:C(Q("Report Year"))[T.2016] -8.0114 10.985 -0.729 0.466 -29.543 13.520
+Group[T.Recent]:C(Q("Report Year"))[T.2017] 1.4929 10.815 0.138 0.890 -19.706 22.692
+Group[T.Recent]:C(Q("Report Year"))[T.2018] 7.6302 10.637 0.717 0.473 -13.220 28.480
+Group[T.Recent]:C(Q("Report Year"))[T.2019] -23.4757 11.153 -2.105 0.035 -45.337 -1.615
+Group[T.Recent]:C(Q("Report Year"))[T.2020] -33.4093 12.106 -2.760 0.006 -57.139 -9.679
+Group[T.Recent]:C(Q("Report Year"))[T.2021] 11.1341 10.351 1.076 0.282 -9.156 31.424
+Group[T.Recent]:C(Q("Report Year"))[T.2022] -7.0144 9.959 -0.704 0.481 -26.536 12.507
+Group[T.Recent]:C(Q("Report Year"))[T.2023] 5.0382 9.959 0.506 0.613 -14.482 24.559
+==============================================================================
+Omnibus: 46016.250 Durbin-Watson: 1.932
+Prob(Omnibus): 0.000 Jarque-Bera (JB): 3571135236.213
+Skew: 39.001 Prob(JB): 0.00
+Kurtosis: 2316.864 Cond. No. 44.9
+==============================================================================
+
+Notes:
+[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+Year
+Historical Delay Change
+Recent Relative Change
+Significant?
+
+
+
+
+2015
++70.13 days
+−65.26 days
+✅ p < 0.001
+
+
+2019
++28.69 days
+−23.48 days
+✅ p = 0.035
+
+
+2020
++33.18 days
+−33.41 days
+✅ p = 0.006
+
+
+2023
+−3.94 days
++5.04 days
+❌
+
+
+
+
+
+
+
+
+
+
+
+
+
+To test the robustness of our DiD findings, we specified a model with year fixed effects and group-year interactions. This approach allows us to control for year-specific shocks and identify how recent and historical spills responded differently across time. We found that historical spills experienced sharp spikes in reporting delays in 2015, 2019, and 2020—rising by 70, 29, and 33 days respectively—while recent spills did not mirror these trends. In fact, the interaction terms show that recent spills experienced significantly less delay than historical ones in each of those years. These results reinforce the observation that recent spills followed a more stable reporting pattern, and that 2020 disruptions disproportionately affected the historical spill category.
+
+
+
+
+
+
+
+
+In [37]:
+
+
+import numpy as np
+import pandas as pd
+import matplotlib.pyplot as plt
+import statsmodels.formula.api as smf
+
+# Re-run the fixed model to get fitted values
+model_fe_fixed = smf.ols(
+ 'Q("Report Delay (Days)") ~ Group * C(Q("Report Year"))',
+ data=combined
+).fit()
+
+# Get predicted means for each year and group
+combined['Fitted'] = model_fe_fixed.fittedvalues
+
+# Calculate average predicted values by year and group
+predicted_means = combined.groupby(['Group', 'Report Year'])['Fitted'].mean().reset_index()
+
+# Plot
+fig, ax = plt.subplots(figsize=(10, 6))
+
+for group in predicted_means['Group'].unique():
+ subset = predicted_means[predicted_means['Group'] == group]
+ ax.plot(subset['Report Year'], subset['Fitted'], marker='o', label=f'{group} Spills')
+
+ax.axvline(2020, color='gray', linestyle='--', linewidth=1, label='2020 Threshold')
+ax.set_title('Model-Predicted Report Delays by Spill Type and Year')
+ax.set_xlabel('Report Year')
+ax.set_ylabel('Predicted Report Delay (Days)')
+ax.legend()
+ax.grid(True)
+plt.tight_layout()
+plt.show()
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+Figure X displays average report delays for historical and recent spills before and after 2020. The figure highlights a substantial reduction in reporting delay for historical spills after 2020, dropping from approximately 29.8 days to 13.5 days. Recent spills, by contrast, show a smaller decline—from 4.7 days to 2.9 days—over the same period. While both groups improved in reporting timeliness, the steeper drop for historical spills suggests that policy or administrative changes implemented around 2020 had a disproportionately greater effect on that category. This visual representation aligns with the Difference-in-Differences regression results, which estimate a significant 14.4-day larger improvement in historical spills relative to recent ones.
+
+
+
+
+
+
+
+
+
+
+Results¶
We evaluated the effect of Colorado's 2021 spill reporting rule on reporting timeliness using a combination of descriptive, inferential, and causal-inference models. Our analysis focuses on changes in report delays—defined as the number of days between the date of spill discovery and the date the spill was officially reported—for two types of spills: historical (discovered long after occurrence) and recent (discovered and reported in near real-time).
+Descriptive and t-test Findings¶
Initial descriptive statistics showed notable improvements in reporting timeliness following the policy change. For historical spills, the mean report delay dropped from 31.17 days before 2021 to 10.18 days after 2020, with a significant reduction in variance (standard deviation dropped from 265.34 to 96.01). For recent spills, the average delay decreased from 4.48 to 2.92 days, with the standard deviation falling from 53.79 to 28.86.
+Two-sample t-tests confirmed that the decrease in report delay was statistically significant for historical spills (t = 3.94, p < 0.001) and marginally significant for recent spills (t = 1.94, p = 0.053). These results suggest meaningful improvement in spill reporting behavior following the regulatory change, with stronger effects observed for more complex, harder-to-detect spill types.
+Difference-in-Differences (DiD) Estimates¶
To evaluate whether recent and historical spills improved at different rates, we employed a Difference-in-Differences (DiD) model using a binary post-2020 indicator and a spill-type interaction term. The DiD model estimated that historical spills improved by 16.2 days on average, while recent spills improved by only 1.8 days. The interaction term was positive and significant (β = 14.4, p = 0.001), indicating that historical spills experienced significantly greater improvements in reporting delay compared to recent spills.
+Parallel Trends and Fixed Effects Models¶
Visual inspection of pre-2020 trends raised concerns about the parallel trends assumption. Historical spills showed sharp spikes in certain years (notably 2015 and 2019), while recent spill delays remained relatively flat. To address this, we estimated a fixed-effects regression model with year and group-by-year interactions. This model relaxed the parallel trends assumption and allowed us to account for time-specific shocks.
+The fixed-effects model confirmed our core finding: recent spills diverged significantly from historical trends in key policy-relevant years. In 2020, historical spills experienced a 33-day increase in reporting delay, while recent spills showed a 33-day smaller delay—resulting in a significant 66-day relative difference (p = 0.006). Significant differences were also observed in 2015 and 2019, further reinforcing the conclusion that recent spills were less affected by structural or reporting challenges.
+Reasons DiD Model Is Inappropriate¶
A difference-in-differences (DiD) design assumes parallel pre-policy trends between treated and control groups. In our case, historical and recent spills are fundamentally different: recent spills are reported shortly after discovery, while historical spills reflect delayed discoveries. Because their reporting dynamics are structurally distinct and not plausibly parallel over time, DiD is not an appropriate design.
+Key Takeaways¶
+- Historical spills: Showed significant improvement in reporting delays post-2020, with a 16.2-day reduction.
+- Recent spills: Experienced a smaller, marginally significant reduction of 1.8 days.
+- DiD model: Indicated that historical spills improved significantly more than recent spills, but the parallel trends assumption is violated.
+- Fixed-effects model: Confirmed that recent spills diverged from historical trends, particularly in key years (2015, 2019, 2020), reinforcing the idea that recent spills are less affected by structural challenges.
+- Policy implications: The 2021 policy change appears to have standardized reporting practices and improved timeliness, particularly for historical spills, suggesting that regulatory changes can effectively address reporting delays.
+
+
+
+
diff --git a/analysis/Analysis Output - HTML Files/analysis6 - 2021_t-tests_2021.html b/analysis/Analysis Output - HTML Files/analysis6 - 2021_t-tests_2021.html
new file mode 100644
index 0000000..f3c3acf
--- /dev/null
+++ b/analysis/Analysis Output - HTML Files/analysis6 - 2021_t-tests_2021.html
@@ -0,0 +1,8330 @@
+
+
+
+
+
+analysis6 - 2021_t-tests_2021
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+In [1]:
+
+
+import pandas as pd
+from sqlalchemy import create_engine
+import geopandas as gpd
+from dotenv import load_dotenv
+load_dotenv()
+
+import os
+
+# Database connection details from zshrc environment variables
+# use .env file for local development
+
+
+db_name = 'colorado_spills'
+user = os.getenv('DB_USER')
+password = os.getenv('DB_PASSWORD')
+host = os.getenv('DB_HOST')
+port = os.getenv('DB_PORT')
+
+
+# Create an engine to connect to the PostgreSQL database
+engine = create_engine(f'postgresql+psycopg2://{user}:{password}@{host}:{port}/{db_name}')
+
+# Read in the spills_with_demographics_geog as spills
+spills = pd.read_sql_table('spills_with_demographics_geog', engine)
+
+
+
+
+
+
+
+
+
+
+
+
+/home/dadams/miniconda3/envs/spatial_env2/lib/python3.10/site-packages/pandas/io/sql.py:1725: SAWarning: Did not recognize type 'geometry' of column 'geometry'
+ self.meta.reflect(bind=self.con, only=[table_name], views=True)
+
+
+
+
+
+
+
+
+
+
+In [2]:
+
+
+# use longitude and latitude to create a GeoDataFrame from spills data
+spills['geometry'] = gpd.points_from_xy(spills['Longitude'], spills['Latitude'])
+
+
+
+
+
+
+
+
+
+
+In [3]:
+
+
+spills_gdf = gpd.GeoDataFrame(spills, crs='EPSG:4326')
+
+
+
+
+
+
+
+
+
+
+In [4]:
+
+
+# Ensure date columns are in datetime format
+spills_gdf['Date of Discovery'] = pd.to_datetime(spills_gdf['Date of Discovery'])
+spills_gdf['Initial Report Date'] = pd.to_datetime(spills_gdf['Initial Report Date'])
+
+
+
+
+
+
+
+
+
+
+In [5]:
+
+
+# drop 2024 data for date of discovery and initial report date
+spills_gdf = spills_gdf[spills_gdf['Date of Discovery'].dt.year < 2024]
+spills_gdf = spills_gdf[spills_gdf['Initial Report Date'].dt.year < 2024]
+
+
+
+
+
+
+
+
+
+
+In [6]:
+
+
+# Separate Historical vs. Recent Spills
+historical_spills = spills_gdf[spills_gdf['Spill Type'] == 'Historical']
+recent_spills = spills_gdf[spills_gdf['Spill Type'] == 'Recent']
+
+
+
+
+
+
+
+
+
+
+In [7]:
+
+
+# Compare Reporting Delay
+report_delay_summary = spills_gdf.groupby('Spill Type')['Report Delay (Days)'].describe()
+print(report_delay_summary)
+
+# Ensure the output directory exists
+output_dir = 'analysis/output'
+os.makedirs(output_dir, exist_ok=True)
+
+# Save the report delay summary to a markdown file
+with open(os.path.join(output_dir, 'report_delay_summary.md'), 'w') as f:
+ f.write("# Report Delay Summary\n\n")
+ f.write(report_delay_summary.to_markdown())
+
+
+
+
+
+
+
+
+
+
+
+
+ count mean std min 25% 50% 75% max
+Spill Type
+Historical 5312.0 21.490023 205.615913 0.0 0.0 1.0 2.0 9261.0
+Recent 10678.0 3.964506 46.997749 0.0 0.0 1.0 2.0 2232.0
+
+
+
+
+
+
+
+
+
+
+In [8]:
+
+
+import matplotlib.pyplot as plt
+
+# Compare Spill Trends Over Time
+spills_by_year = spills_gdf.groupby(['Report Year', 'Spill Type']).size().unstack()
+ax = spills_by_year.plot(kind='bar', figsize=(12,6), title='Spills by Year and Type')
+ax.set_xlabel('Year')
+ax.set_ylabel('Number of Spills')
+ax.legend(title='Spill Type')
+
+# Ensure the output directory exists
+output_dir = 'analysis/output'
+os.makedirs(output_dir, exist_ok=True)
+
+# Save the plot
+plt.savefig(f'{output_dir}/spills_by_year_and_type.png')
+
+# Display the plot
+plt.show()
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+In [9]:
+
+
+import matplotlib.pyplot as plt
+import geopandas as gpd
+
+# Ensure spills_gdf is already a GeoDataFrame with the necessary data
+# spills_gdf = gpd.GeoDataFrame(...)
+
+# Create a plot for the spatial distribution of spills
+fig, ax = plt.subplots(figsize=(10, 6))
+
+# Plot the GeoDataFrame with optimizations
+spills_gdf.plot(
+ ax=ax,
+ marker='o',
+ alpha=0.6, # Adjust transparency for better visibility
+ column='Spill Type',
+ legend=True,
+ legend_kwds={'title': "Spill Type"}, # Correctly specify the legend title
+ cmap='viridis' # Use a colormap for better visual distinction
+)
+
+# Set plot title and labels
+ax.set_title('Spatial Distribution of Historical vs. Recent Spills', fontsize=14)
+ax.set_xlabel('Longitude', fontsize=12)
+ax.set_ylabel('Latitude', fontsize=12)
+
+# Improve layout
+plt.tight_layout()
+
+# Display the plot
+plt.show()
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+In [10]:
+
+
+import matplotlib.pyplot as plt
+import contextily as ctx
+import numpy as np
+
+# Use your ORIGINAL plot code exactly as it is, just adding the basemap at the end
+# This ensures all your data points, colors and categories remain intact
+
+# Starting with your original code:
+fig, ax = plt.subplots(figsize=(10, 6))
+
+# Plot the GeoDataFrame with optimizations - YOUR ORIGINAL PLOTTING CODE
+spills_gdf.plot(
+ ax=ax,
+ marker='o',
+ alpha=0.6, # Adjust transparency for better visibility
+ column='Spill Type',
+ legend=True,
+ legend_kwds={'title': "Spill Type"}, # Correctly specify the legend title
+ cmap='viridis' # Use a colormap for better visual distinction
+)
+
+# Set plot title and labels
+ax.set_title('Spatial Distribution of Historical vs. Recent Spills in Colorado', fontsize=14)
+ax.set_xlabel('Longitude', fontsize=12)
+ax.set_ylabel('Latitude', fontsize=12)
+
+# Store the current axis limits to restore them after adding basemap
+xlim = ax.get_xlim()
+ylim = ax.get_ylim()
+
+# Before adding the basemap, convert the axes coordinates to Web Mercator
+# This is a crucial step that lets us keep your original points exactly as they are
+from pyproj import Transformer
+
+# Create transformer from lat/lon to Web Mercator
+transformer = Transformer.from_crs("EPSG:4326", "EPSG:3857", always_xy=True)
+
+# Transform the limits
+west, south = transformer.transform(xlim[0], ylim[0])
+east, north = transformer.transform(xlim[1], ylim[1])
+
+# Add the basemap - this will be UNDER your existing points
+ctx.add_basemap(
+ ax,
+ crs=spills_gdf.crs,
+ source=ctx.providers.OpenStreetMap.Mapnik,
+ zoom="auto"
+)
+
+# Restore original axis limits
+ax.set_xlim(xlim)
+ax.set_ylim(ylim)
+
+# Improve layout
+plt.tight_layout()
+
+# Save the plot to the analysis/output directory
+output_dir = 'analysis/output'
+os.makedirs(output_dir, exist_ok=True)
+plt.savefig(f'{output_dir}/spatial_distribution_spills.png')
+
+# Show the plot
+plt.show()
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+In [11]:
+
+
+# Add a column to indicate whether the spill occurred before or after 2021
+spills_gdf['Period'] = spills_gdf['Report Year'].apply(lambda x: 'Before 2021' if x < 2021 else '2021 and After')
+
+# Group by the new period column and calculate the mean response time
+response_time_summary = spills_gdf.groupby('Period')['Report Delay (Days)'].describe()
+print(response_time_summary)
+
+# Ensure the output directory exists
+output_dir = 'analysis/output'
+os.makedirs(output_dir, exist_ok=True)
+
+# Save the response time summary to a markdown file
+with open(os.path.join(output_dir, 'response_time_summary.md'), 'w') as f:
+ f.write("# Response Time Summary\n\n")
+ f.write(response_time_summary.to_markdown())
+
+
+
+
+
+
+
+
+
+
+
+
+ count mean std min 25% 50% 75% max
+Period
+2021 and After 6003.0 5.890055 65.333629 0.0 0.0 1.0 1.0 2501.0
+Before 2021 9987.0 12.128767 149.585631 0.0 0.0 1.0 2.0 9261.0
+
+
+
+
+
+
+
+
+
+
+In [12]:
+
+
+# t-test for the difference in response time before and after 2021
+from scipy.stats import ttest_ind
+
+before_2021 = spills_gdf[spills_gdf['Period'] == 'Before 2021']['Report Delay (Days)']
+after_2021 = spills_gdf[spills_gdf['Period'] == '2021 and After']['Report Delay (Days)']
+
+t_stat, p_value = ttest_ind(before_2021, after_2021, equal_var=False)
+print(f'T-Statistic: {t_stat:.4f}')
+print(f'P-Value: {p_value:.4f}')
+
+# Check if the difference is statistically significant
+if p_value < 0.05:
+ print('The difference in response time is statistically significant.')
+else:
+ print('The difference in response time is not statistically significant.')
+
+# Create a plot for the response time distribution before and after 2021
+fig, ax = plt.subplots(figsize=(10, 6))
+
+# Plot the response time distribution for each period
+spills_gdf.boxplot(column='Report Delay (Days)', by='Period', ax=ax)
+
+
+
+
+
+
+
+
+
+
+
+
+T-Statistic: 3.6314
+P-Value: 0.0003
+The difference in response time is statistically significant.
+
+
+
+
+Out[12]:
+
+<Axes: title={'center': 'Report Delay (Days)'}, xlabel='Period'>
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+In [13]:
+
+
+%pip install tabulate
+
+
+
+
+
+
+
+
+
+
+
+
+Requirement already satisfied: tabulate in /home/dadams/miniconda3/envs/spatial_env2/lib/python3.10/site-packages (0.9.0)
+Note: you may need to restart the kernel to use updated packages.
+
+
+
+
+
+
+
+
+
+
+In [14]:
+
+
+import tabulate
+from scipy.stats import ttest_ind
+
+# Ensure the output directory exists
+output_dir = 'analysis/output'
+os.makedirs(output_dir, exist_ok=True)
+
+# Add a column to indicate whether the spill occurred before or after 2021 for historical and recent spills
+historical_spills['Period'] = historical_spills['Report Year'].apply(lambda x: 'Before 2021' if x < 2021 else '2021 and After')
+recent_spills['Period'] = recent_spills['Report Year'].apply(lambda x: 'Before 2021' if x < 2021 else '2021 and After')
+
+# Historical Spills Analysis
+historical_report_delay_summary = historical_spills['Report Delay (Days)'].describe()
+print("Historical Spills Report Delay Summary:")
+print(historical_report_delay_summary)
+
+# Recent Spills Analysis
+recent_report_delay_summary = recent_spills['Report Delay (Days)'].describe()
+print("\nRecent Spills Report Delay Summary:")
+print(recent_report_delay_summary)
+
+# Historical Spills Response Time Summary
+historical_response_time_summary = historical_spills.groupby('Period')['Report Delay (Days)'].describe()
+print("\nHistorical Spills Response Time Summary:")
+print(historical_response_time_summary)
+
+# Recent Spills Response Time Summary
+recent_response_time_summary = recent_spills.groupby('Period')['Report Delay (Days)'].describe()
+print("\nRecent Spills Response Time Summary:")
+print(recent_response_time_summary)
+
+# t-test for the difference in response time before and after 2021 for historical spills
+historical_before_2021 = historical_spills[historical_spills['Period'] == 'Before 2021']['Report Delay (Days)']
+historical_after_2021 = historical_spills[historical_spills['Period'] == '2021 and After']['Report Delay (Days)']
+
+t_stat_historical, p_value_historical = ttest_ind(historical_before_2021, historical_after_2021, equal_var=False)
+print(f'\nHistorical Spills T-Statistic: {t_stat_historical:.4f}')
+print(f'Historical Spills P-Value: {p_value_historical:.4f}')
+
+# Check if the difference is statistically significant for historical spills
+if p_value_historical < 0.05:
+ print('The difference in response time for historical spills is statistically significant.')
+else:
+ print('The difference in response time for historical spills is not statistically significant.')
+
+# t-test for the difference in response time before and after 2021 for recent spills
+recent_before_2021 = recent_spills[recent_spills['Period'] == 'Before 2021']['Report Delay (Days)']
+recent_after_2021 = recent_spills[recent_spills['Period'] == '2021 and After']['Report Delay (Days)']
+
+t_stat_recent, p_value_recent = ttest_ind(recent_before_2021, recent_after_2021, equal_var=False)
+print(f'\nRecent Spills T-Statistic: {t_stat_recent:.4f}')
+print(f'Recent Spills P-Value: {p_value_recent:.4f}')
+
+# Check if the difference is statistically significant for recent spills
+if p_value_recent < 0.05:
+ print('The difference in response time for recent spills is statistically significant.')
+else:
+ print('The difference in response time for recent spills is not statistically significant.')
+
+# Save the summaries and t-test results to a markdown file
+with open(os.path.join(output_dir, 'analysis_results.md'), 'w') as f:
+ f.write("# Analysis Results\n\n")
+
+ f.write("## Historical Spills Report Delay Summary:\n")
+ f.write(historical_report_delay_summary.to_markdown() + "\n\n")
+
+ f.write("## Recent Spills Report Delay Summary:\n")
+ f.write(recent_report_delay_summary.to_markdown() + "\n\n")
+
+ f.write("## Historical Spills Response Time Summary:\n")
+ f.write(historical_response_time_summary.to_markdown() + "\n\n")
+
+ f.write("## Recent Spills Response Time Summary:\n")
+ f.write(recent_response_time_summary.to_markdown() + "\n\n")
+
+ f.write("## T-Test Results for Historical Spills\n")
+ f.write(f"T-Statistic: {t_stat_historical:.4f}\n")
+ f.write(f"P-Value: {p_value_historical:.4f}\n")
+ if p_value_historical < 0.05:
+ f.write('The difference in response time for historical spills is statistically significant.\n')
+ else:
+ f.write('The difference in response time for historical spills is not statistically significant.\n')
+
+ f.write("\n## T-Test Results for Recent Spills\n")
+ f.write(f"T-Statistic: {t_stat_recent:.4f}\n")
+ f.write(f"P-Value: {p_value_recent:.4f}\n")
+ if p_value_recent < 0.05:
+ f.write('The difference in response time for recent spills is statistically significant.\n')
+ else:
+ f.write('The difference in response time for recent spills is not statistically significant.\n')
+
+
+
+
+
+
+
+
+
+
+
+
+Historical Spills Report Delay Summary:
+count 5312.000000
+mean 21.490023
+std 205.615913
+min 0.000000
+25% 0.000000
+50% 1.000000
+75% 2.000000
+max 9261.000000
+Name: Report Delay (Days), dtype: float64
+
+Recent Spills Report Delay Summary:
+count 10678.000000
+mean 3.964506
+std 46.997749
+min 0.000000
+25% 0.000000
+50% 1.000000
+75% 2.000000
+max 2232.000000
+Name: Report Delay (Days), dtype: float64
+
+Historical Spills Response Time Summary:
+ count mean std min 25% 50% 75% max
+Period
+2021 and After 2451.0 10.181559 96.011488 0.0 0.0 0.0 1.0 2501.0
+Before 2021 2861.0 31.177910 265.348026 0.0 0.0 1.0 2.0 9261.0
+
+Recent Spills Response Time Summary:
+ count mean std min 25% 50% 75% max
+Period
+2021 and After 3552.0 2.928773 28.864471 0.0 0.0 1.0 1.0 1329.0
+Before 2021 7126.0 4.480775 53.794907 0.0 0.0 1.0 2.0 2232.0
+
+Historical Spills T-Statistic: 3.9419
+Historical Spills P-Value: 0.0001
+The difference in response time for historical spills is statistically significant.
+
+Recent Spills T-Statistic: 1.9390
+Recent Spills P-Value: 0.0525
+The difference in response time for recent spills is not statistically significant.
+
+
+
+
+
+
+/home/dadams/miniconda3/envs/spatial_env2/lib/python3.10/site-packages/geopandas/geodataframe.py:1819: SettingWithCopyWarning:
+A value is trying to be set on a copy of a slice from a DataFrame.
+Try using .loc[row_indexer,col_indexer] = value instead
+
+See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
+ super().__setitem__(key, value)
+/home/dadams/miniconda3/envs/spatial_env2/lib/python3.10/site-packages/geopandas/geodataframe.py:1819: SettingWithCopyWarning:
+A value is trying to be set on a copy of a slice from a DataFrame.
+Try using .loc[row_indexer,col_indexer] = value instead
+
+See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
+ super().__setitem__(key, value)
+
+
+
+
+
+
+
+
+
+
+In [15]:
+
+
+import os
+
+# Ensure the output directory exists
+output_dir = 'analysis/output'
+os.makedirs(output_dir, exist_ok=True)
+
+# Save the t-test results to a markdown file
+with open(os.path.join(output_dir, 't_test_results.md'), 'w') as f:
+ f.write(f"## T-Test Results for Historical Spills\n")
+ f.write(f"T-Statistic: {t_stat_historical:.4f}\n")
+ f.write(f"P-Value: {p_value_historical:.4f}\n")
+ if p_value_historical < 0.05:
+ f.write('The difference in response time for historical spills is statistically significant.\n')
+ else:
+ f.write('The difference in response time for historical spills is not statistically significant.\n')
+
+ f.write(f"\n## T-Test Results for Recent Spills\n")
+ f.write(f"T-Statistic: {t_stat_recent:.4f}\n")
+ f.write(f"P-Value: {p_value_recent:.4f}\n")
+ if p_value_recent < 0.05:
+ f.write('The difference in response time for recent spills is statistically significant.\n')
+ else:
+ f.write('The difference in response time for recent spills is not statistically significant.\n')
+
+# Create and save the plots
+fig, ax = plt.subplots(figsize=(10, 6))
+historical_spills.boxplot(column='Report Delay (Days)', by='Period', ax=ax)
+ax.set_title('Historical Spills Response Time Distribution Before and After 2021', fontsize=14)
+ax.set_xlabel('Period', fontsize=12)
+ax.set_ylabel('Report Delay (Days)', fontsize=12)
+plt.suptitle('')
+plt.tight_layout()
+plt.savefig(os.path.join(output_dir, 'historical_spills_response_time_distribution.png'))
+
+fig, ax = plt.subplots(figsize=(10, 6))
+recent_spills.boxplot(column='Report Delay (Days)', by='Period', ax=ax)
+ax.set_title('Recent Spills Response Time Distribution Before and After 2021', fontsize=14)
+ax.set_xlabel('Period', fontsize=12)
+ax.set_ylabel('Report Delay (Days)', fontsize=12)
+plt.suptitle('')
+plt.tight_layout()
+plt.savefig(os.path.join(output_dir, 'recent_spills_response_time_distribution.png'))
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+In [16]:
+
+
+# t-test for the difference in response time before and after 2021 for historical spills
+historical_before_2021 = historical_spills[historical_spills['Period'] == 'Before 2021']['Report Delay (Days)']
+historical_after_2021 = historical_spills[historical_spills['Period'] == '2021 and After']['Report Delay (Days)']
+
+t_stat, p_value = ttest_ind(historical_before_2021, historical_after_2021, equal_var=False)
+print(f'\nHistorical Spills T-Statistic: {t_stat:.4f}')
+print(f'Historical Spills P-Value: {p_value:.4f}')
+
+# Check if the difference is statistically significant for historical spills
+if p_value < 0.05:
+ print('The difference in response time for historical spills is statistically significant.')
+else:
+ print('The difference in response time for historical spills is not statistically significant.')
+
+# t-test for the difference in response time before and after 2021 for recent spills
+recent_before_2021 = recent_spills[recent_spills['Period'] == 'Before 2021']['Report Delay (Days)']
+recent_after_2021 = recent_spills[recent_spills['Period'] == '2021 and After']['Report Delay (Days)']
+
+t_stat, p_value = ttest_ind(recent_before_2021, recent_after_2021, equal_var=False)
+print(f'\nRecent Spills T-Statistic: {t_stat:.4f}')
+print(f'Recent Spills P-Value: {p_value:.4f}')
+
+# Check if the difference is statistically significant for recent spills
+if p_value < 0.05:
+ print('The difference in response time for recent spills is statistically significant.')
+else:
+ print('The difference in response time for recent spills is not statistically significant.')
+
+# Create a plot for the response time distribution before and after 2021 for historical and recent spills
+fig, ax = plt.subplots(figsize=(10, 6))
+
+# Plot the response time distribution for historical spills
+historical_spills.boxplot(column='Report Delay (Days)', by='Period', ax=ax, positions=[1, 2])
+
+# Plot the response time distribution for recent spills
+recent_spills.boxplot(column='Report Delay (Days)', by='Period', ax=ax, positions=[4, 5])
+
+# Save the plot to the /output directory
+plt.savefig('analysis/output/response_time_distribution.png')
+
+
+
+
+
+
+
+
+
+
+
+
+
+Historical Spills T-Statistic: 3.9419
+Historical Spills P-Value: 0.0001
+The difference in response time for historical spills is statistically significant.
+
+Recent Spills T-Statistic: 1.9390
+Recent Spills P-Value: 0.0525
+The difference in response time for recent spills is not statistically significant.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
diff --git a/analysis/Analysis Output - HTML Files/analysis7_OLS_NRB_ITS_2020.html b/analysis/Analysis Output - HTML Files/analysis7_OLS_NRB_ITS_2020.html
new file mode 100644
index 0000000..bdd1635
--- /dev/null
+++ b/analysis/Analysis Output - HTML Files/analysis7_OLS_NRB_ITS_2020.html
@@ -0,0 +1,8914 @@
+
+
+
+
+
+analysis7_OLS_NRB_ITS_2020
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+Colorado Spills¶
OLS, NBR, Interrupted Time Series Analysis¶
Author: David P. Adams, PhD¶
Date: 2025-04-18¶
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+Setup¶
+
+
+
+
+
+
+
+
+In [2]:
+
+
+import pandas as pd
+from sqlalchemy import create_engine
+import geopandas as gpd
+from dotenv import load_dotenv
+load_dotenv()
+
+import os
+
+# Database connection details from zshrc environment variables
+# use .env file for local development
+
+
+db_name = 'colorado_spills'
+user = os.getenv('DB_USER')
+password = os.getenv('DB_PASSWORD')
+host = os.getenv('DB_HOST')
+port = os.getenv('DB_PORT')
+
+
+# Create an engine to connect to the PostgreSQL database
+engine = create_engine(f'postgresql+psycopg2://{user}:{password}@{host}:{port}/{db_name}')
+
+# Read in the spills_with_demographics_geog as spills
+spills = pd.read_sql_table('spills_with_demographics_geog', engine)
+
+
+
+
+
+
+
+
+
+
+
+
+/home/dadams/miniconda3/envs/spatial_env2/lib/python3.10/site-packages/pandas/io/sql.py:1725: SAWarning: Did not recognize type 'geometry' of column 'geometry'
+ self.meta.reflect(bind=self.con, only=[table_name], views=True)
+
+
+
+
+
+
+
+
+
+
+In [3]:
+
+
+# use longitude and latitude to create a GeoDataFrame from spills data
+spills['geometry'] = gpd.points_from_xy(spills['Longitude'], spills['Latitude'])
+
+
+
+
+
+
+
+
+
+
+In [4]:
+
+
+spills_gdf = gpd.GeoDataFrame(spills, crs='EPSG:4326')
+
+
+
+
+
+
+
+
+
+
+In [5]:
+
+
+# Ensure date columns are in datetime format
+spills_gdf['Date of Discovery'] = pd.to_datetime(spills_gdf['Date of Discovery'])
+spills_gdf['Initial Report Date'] = pd.to_datetime(spills_gdf['Initial Report Date'])
+
+
+
+
+
+
+
+
+
+
+In [6]:
+
+
+# drop 2024 data for date of discovery and initial report date
+spills_gdf = spills_gdf[spills_gdf['Date of Discovery'].dt.year < 2024]
+spills_gdf = spills_gdf[spills_gdf['Initial Report Date'].dt.year < 2024]
+
+
+
+
+
+
+
+
+
+
+In [7]:
+
+
+# Separate Historical vs. Recent Spills
+historical_spills = spills_gdf[spills_gdf['Spill Type'] == 'Historical']
+recent_spills = spills_gdf[spills_gdf['Spill Type'] == 'Recent']
+
+
+
+
+
+
+
+
+
+
+In [37]:
+
+
+import statsmodels.formula.api as smf
+
+# Create 'Period' column
+spills_gdf['Period'] = spills_gdf['Report Year'].apply(lambda x: 'Before 2020' if x < 2020 else '2020 and After')
+
+
+
+
+
+
+
+
+
+
+In [38]:
+
+
+spills_gdf = spills_gdf.rename(columns={
+ 'Report Delay (Days)': 'report_delay',
+ 'Spill Type': 'spill_type'
+})
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+OLS Regression¶
+
+
+
+
+
+
+
+
+In [16]:
+
+
+import statsmodels.formula.api as smf
+
+model = smf.ols("report_delay ~ C(spill_type) * C(Period)", data=spills_gdf).fit()
+print(model.summary())
+
+
+
+
+
+
+
+
+
+
+
+
+ OLS Regression Results
+==============================================================================
+Dep. Variable: report_delay R-squared: 0.006
+Model: OLS Adj. R-squared: 0.006
+Method: Least Squares F-statistic: 31.12
+Date: Fri, 18 Apr 2025 Prob (F-statistic): 4.78e-20
+Time: 21:03:48 Log-Likelihood: -99827.
+No. Observations: 15990 AIC: 1.997e+05
+Df Residuals: 15986 BIC: 1.997e+05
+Df Model: 3
+Covariance Type: nonrobust
+====================================================================================================================
+ coef std err t P>|t| [0.025 0.975]
+--------------------------------------------------------------------------------------------------------------------
+Intercept 13.5331 2.394 5.654 0.000 8.841 18.225
+C(spill_type)[T.Recent] -10.6062 3.032 -3.498 0.000 -16.550 -4.662
+C(Period)[T.Before 2020] 16.2130 3.417 4.745 0.000 9.516 22.910
+C(spill_type)[T.Recent]:C(Period)[T.Before 2020] -14.4276 4.200 -3.435 0.001 -22.660 -6.196
+==============================================================================
+Omnibus: 45995.174 Durbin-Watson: 1.996
+Prob(Omnibus): 0.000 Jarque-Bera (JB): 3525625525.882
+Skew: 38.959 Prob(JB): 0.00
+Kurtosis: 2302.059 Cond. No. 8.76
+==============================================================================
+
+Notes:
+[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
+
+
+
+
+
+
+
+
+
+
+
+
+Spill Reporting Analysis¶
What the OLS Says¶
This is a 2×2 model:
+
+- Spill Type:
Historical (reference) vs. Recent
+- Period:
2020 and After (reference) vs. Before 2020
+- With their interaction
+
+Coefficient interpretations:¶
+- Intercept (13.53): Avg delay for historical spills after 2020.
+- Recent (–10.61): Recent spills after 2020 had ~10.6 days faster reporting than historical ones.
+- Before 2020 (+16.21): Historical spills before 2020 were reported ~16.2 days slower than those after 2020.
+- Interaction (–14.43): Recent spills before 2020 were ~14.4 days faster than expected given the sum of the main effects. That’s key.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+Group
+Formula
+Result
+
+
+
+
+Historical, After 2020
+Intercept
+13.53 days
+
+
+Recent, After 2020
+Intercept + Recent
+2.92 days
+
+
+Historical, Before 2020
+Intercept + Before 2020
+29.74 days
+
+
+Recent, Before 2020
+Intercept + Before 2020 + Recent + Interaction
+4.70 days
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+Negative Binomial Regression¶
+
+
+
+
+
+
+
+
+In [23]:
+
+
+import statsmodels.formula.api as smf
+import statsmodels.api as sm
+
+# Make sure spill_type and Period are clean, and report_delay is numeric
+spills_gdf = spills_gdf.rename(columns={
+ 'Report Delay (Days)': 'report_delay',
+ 'Spill Type': 'spill_type'
+})
+
+# Fit the Negative Binomial model with interaction
+nb_model = smf.glm(
+ formula="report_delay ~ C(spill_type) * C(Period)",
+ data=spills_gdf,
+ family=sm.families.NegativeBinomial()
+).fit()
+
+# View the results
+print(nb_model.summary())
+
+
+
+
+
+
+
+
+
+
+
+
+ Generalized Linear Model Regression Results
+==============================================================================
+Dep. Variable: report_delay No. Observations: 15990
+Model: GLM Df Residuals: 15986
+Model Family: NegativeBinomial Df Model: 3
+Link Function: Log Scale: 1.0000
+Method: IRLS Log-Likelihood: -47751.
+Date: Fri, 18 Apr 2025 Deviance: 54508.
+Time: 21:14:16 Pearson chi2: 1.72e+06
+No. Iterations: 12 Pseudo R-squ. (CS): 0.4976
+Covariance Type: nonrobust
+====================================================================================================================
+ coef std err z P>|z| [0.025 0.975]
+--------------------------------------------------------------------------------------------------------------------
+Intercept 2.6051 0.020 130.748 0.000 2.566 2.644
+C(spill_type)[T.Recent] -1.5312 0.026 -57.998 0.000 -1.583 -1.479
+C(Period)[T.Before 2020] 0.7876 0.028 27.959 0.000 0.732 0.843
+C(spill_type)[T.Recent]:C(Period)[T.Before 2020] -0.3113 0.036 -8.672 0.000 -0.382 -0.241
+====================================================================================================================
+
+
+
+
+
+
+/home/dadams/miniconda3/envs/spatial_env2/lib/python3.10/site-packages/statsmodels/genmod/families/family.py:1367: ValueWarning: Negative binomial dispersion parameter alpha not set. Using default value alpha=1.0.
+ warnings.warn("Negative binomial dispersion parameter alpha not "
+
+
+
+
+
+
+
+
+
+
+
+
+Negative Binomial Regression Results¶
Key Coefficients (in log count space)¶
+
+
+Term
+Coef (log)
+Exp(Coef)
+Interpretation
+
+
+
+
+Intercept
+2.605
+13.53
+Expected delay for Historical spills after 2020: ~13.5 days
+
+
+Recent
+–1.531
+0.216
+Recent spills (after 2020) take ~78% less time to report
+
+
+Before 2020
+0.788
+2.20
+Historical spills before 2020 took ~2.2x longer
+
+
+Interaction
+–0.311
+0.733
+Recent spills before 2020 were ~27% faster than expected given the base effects
+
+
+
+
+Summary Interpretation¶
+- Recent reporting improved substantially after 2020, even after controlling for spill type.
+- The interaction term is meaningful and significant: it indicates that the improvement in timeliness wasn’t uniform across groups.
+- Model fit is strong for count data:
+- Log-likelihood is stable
+- Pseudo R² ≈ 0.498
+- No convergence issues or overdispersion errors yet (though
alpha=1.0 is default — you can estimate that if needed)
+
+
+
+
+
+
+
+
+
+
+
+In [24]:
+
+
+import numpy as np
+np.exp(nb_model.params)
+
+
+
+
+
+
+
+
+
+
+Out[24]:
+
+Intercept 13.533087
+C(spill_type)[T.Recent] 0.216276
+C(Period)[T.Before 2020] 2.198025
+C(spill_type)[T.Recent]:C(Period)[T.Before 2020] 0.732466
+dtype: float64
+
+
+
+
+
+
+
+
+
+
+
+Group-Level Predictions (Multiplicative Model)¶
Log link in a Negative Binomial model:¶
[
+\text{Expected Delay} = e^{\text{sum of coefficients for that group}}
+]
+Or, since you've already exponentiated the coefficients:
+[
+\text{Delay} = \text{Intercept} \times \text{Multipliers}
+]
+
+
+
+Group
+Calculation
+Predicted Delay (days)
+
+
+
+
+Historical, After 2020
+13.5313.53
+
+
+Recent, After 2020
+13.53 × 0.216≈ 2.93
+
+
+Historical, Before 2020
+13.53 × 2.20≈ 29.77
+
+
+Recent, Before 2020
+13.53 × 0.216 × 2.20 × 0.732≈ 4.26
+
+
+
+
+Interpretation¶
+- After 2020, recent spills are reported ~10.6 days faster than historical ones (2.93 vs. 13.53 days).
+- Historical reporting times worsened before 2020 (29.8 days).
+- The biggest drop is in recent spill reporting after 2020 — a likely policy win.
+
+
+
+
+
+
+
+
+
+In [25]:
+
+
+import matplotlib.pyplot as plt
+import pandas as pd
+
+# Predicted delays by group
+group_labels = [
+ "Historical, After 2020",
+ "Recent, After 2020",
+ "Historical, Before 2020",
+ "Recent, Before 2020"
+]
+
+# From your exponentiated results
+predicted_delays = [
+ 13.53, # Intercept only
+ 13.53 * 0.216, # Intercept * Recent
+ 13.53 * 2.198, # Intercept * Before 2020
+ 13.53 * 0.216 * 2.198 * 0.732 # Intercept * Recent * Before 2020 * Interaction
+]
+
+# Create DataFrame for plotting
+df_plot = pd.DataFrame({
+ "Group": group_labels,
+ "Predicted Delay (Days)": predicted_delays
+})
+
+# Plot
+plt.figure(figsize=(10, 6))
+bars = plt.bar(df_plot["Group"], df_plot["Predicted Delay (Days)"], alpha=0.8)
+plt.title("Predicted Report Delay by Spill Type and Period", fontsize=14)
+plt.ylabel("Predicted Delay (Days)", fontsize=12)
+plt.xticks(rotation=15, ha='right')
+plt.grid(axis='y', linestyle='--', alpha=0.7)
+plt.tight_layout()
+
+# Annotate bars
+for bar in bars:
+ height = bar.get_height()
+ plt.annotate(f"{height:.2f}",
+ xy=(bar.get_x() + bar.get_width() / 2, height),
+ xytext=(0, 3), # 3 points vertical offset
+ textcoords="offset points",
+ ha='center', va='bottom')
+
+plt.show()
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+What the Plot Highlights:¶
+- Historical spills before 2020 had the longest delays by far (~30 days).
+- Recent spills after 2020 saw the shortest (~3 days).
+- The interaction effect shows recent + before 2020 was better than historical, but still far worse than recent + after 2020.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+Interrupted Time Series¶
+
+
+
+
+
+
+
+
+In [30]:
+
+
+import pandas as pd
+import statsmodels.formula.api as smf
+import matplotlib.pyplot as plt
+
+# 1. Convert to monthly time series
+spills_gdf['report_month'] = spills_gdf['Initial Report Date'].dt.to_period('M')
+monthly = spills_gdf.groupby('report_month')['report_delay'].mean().reset_index()
+monthly['report_month'] = monthly['report_month'].dt.to_timestamp()
+
+# 2. Create ITS variables
+monthly['time'] = (monthly['report_month'] - monthly['report_month'].min()).dt.days // 30
+monthly['post_2020'] = (monthly['report_month'] >= pd.Timestamp('2020-01-01')).astype(int)
+monthly['time_post'] = monthly['time'] * monthly['post_2020']
+
+# 3. Run the ITS model
+its_model = smf.ols("report_delay ~ time + post_2020 + time_post", data=monthly).fit()
+print(its_model.summary())
+
+# 4. Predict and plot
+monthly['predicted'] = its_model.predict(monthly)
+
+plt.figure(figsize=(12, 6))
+plt.plot(monthly['report_month'], monthly['report_delay'], marker='o', label='Observed', alpha=0.6)
+plt.plot(monthly['report_month'], monthly['predicted'], linestyle='--', color='red', label='Predicted (ITS)')
+plt.axvline(x=pd.Timestamp('2020-01-01'), color='black', linestyle='--', label='Policy Change (2020)')
+plt.title('Interrupted Time Series: Monthly Report Delay')
+plt.xlabel('Month')
+plt.ylabel('Mean Report Delay (Days)')
+plt.legend()
+plt.grid(True)
+plt.tight_layout()
+plt.show()
+
+
+
+
+
+
+
+
+
+
+
+
+ OLS Regression Results
+==============================================================================
+Dep. Variable: report_delay R-squared: 0.040
+Model: OLS Adj. R-squared: 0.014
+Method: Least Squares F-statistic: 1.535
+Date: Fri, 18 Apr 2025 Prob (F-statistic): 0.209
+Time: 21:29:45 Log-Likelihood: -481.86
+No. Observations: 115 AIC: 971.7
+Df Residuals: 111 BIC: 982.7
+Df Model: 3
+Covariance Type: nonrobust
+==============================================================================
+ coef std err t P>|t| [0.025 0.975]
+------------------------------------------------------------------------------
+Intercept 16.6659 3.929 4.241 0.000 8.879 24.452
+time -0.1671 0.103 -1.627 0.107 -0.371 0.036
+post_2020 -5.0897 16.165 -0.315 0.753 -37.122 26.942
+time_post 0.1165 0.198 0.588 0.558 -0.276 0.509
+==============================================================================
+Omnibus: 125.885 Durbin-Watson: 1.893
+Prob(Omnibus): 0.000 Jarque-Bera (JB): 1986.555
+Skew: 3.880 Prob(JB): 0.00
+Kurtosis: 21.825 Cond. No. 931.
+==============================================================================
+
+Notes:
+[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+Interrupted Time Series Analysis¶
🔍 Model Findings¶
+
+
+Term
+Coefficient
+p-value
+Meaning
+
+
+
+
+Intercept
+16.67
+0.000
+Predicted delay at baseline (month 0)
+
+
+time–0.17
+0.107
+Slight (non-sig.) decline pre-2020
+
+
+post_2020–5.09
+0.753
+No significant level change
+
+
+time_post+0.12
+0.558
+No significant trend change after 2020
+
+
+
+R² = 0.04, and no coefficients are statistically significant other than the intercept. In short: ITS didn’t detect a strong policy effect.
+What the Plot Shows¶
The visual trend still tells a useful story:
+
+- High volatility pre-2020 with outliers and erratic reporting
+- A more stable pattern post-2020
+- No obvious immediate jump or trend break, but lower variance
+
+How to Report This¶
You can say:
+
+“We conducted an Interrupted Time Series analysis using monthly average delays and found no statistically significant level or trend change post-policy. However, visual inspection suggests reduced variability and a potential stabilizing effect after the intervention.”
+
+
+
+
+
+
+
+
+
+
+
+
+ITS by Spill Type¶
+
+
+
+
+
+
+
+
+In [32]:
+
+
+import pandas as pd
+import statsmodels.formula.api as smf
+import matplotlib.pyplot as plt
+
+# Group by month and spill type
+spills_gdf['report_month'] = spills_gdf['Initial Report Date'].dt.to_period('M')
+monthly_by_type = spills_gdf.groupby(['report_month', 'spill_type'])['report_delay'].mean().reset_index()
+monthly_by_type['report_month'] = monthly_by_type['report_month'].dt.to_timestamp()
+
+# ITS variables
+monthly_by_type['time'] = (monthly_by_type['report_month'] - monthly_by_type['report_month'].min()).dt.days // 30
+monthly_by_type['post_2020'] = (monthly_by_type['report_month'] >= pd.Timestamp('2020-01-01')).astype(int)
+monthly_by_type['time_post'] = monthly_by_type['time'] * monthly_by_type['post_2020']
+
+# Run ITS models and print summaries
+its_results_by_type = {}
+for stype in monthly_by_type['spill_type'].unique():
+ df = monthly_by_type[monthly_by_type['spill_type'] == stype].copy()
+ model = smf.ols("report_delay ~ time + post_2020 + time_post", data=df).fit()
+ df['predicted'] = model.predict(df)
+ its_results_by_type[stype] = (model, df)
+
+ print(f"\n=== ITS Model Summary for {stype} Spills ===\n")
+ print(model.summary())
+
+
+
+
+
+
+
+
+
+
+
+
+
+=== ITS Model Summary for Historical Spills ===
+
+ OLS Regression Results
+==============================================================================
+Dep. Variable: report_delay R-squared: 0.034
+Model: OLS Adj. R-squared: 0.008
+Method: Least Squares F-statistic: 1.300
+Date: Fri, 18 Apr 2025 Prob (F-statistic): 0.278
+Time: 21:32:41 Log-Likelihood: -624.01
+No. Observations: 115 AIC: 1256.
+Df Residuals: 111 BIC: 1267.
+Df Model: 3
+Covariance Type: nonrobust
+==============================================================================
+ coef std err t P>|t| [0.025 0.975]
+------------------------------------------------------------------------------
+Intercept 44.2342 13.526 3.270 0.001 17.432 71.036
+time -0.4759 0.354 -1.346 0.181 -1.177 0.225
+post_2020 -6.8325 55.641 -0.123 0.902 -117.089 103.424
+time_post 0.2196 0.682 0.322 0.748 -1.132 1.571
+==============================================================================
+Omnibus: 146.114 Durbin-Watson: 2.041
+Prob(Omnibus): 0.000 Jarque-Bera (JB): 3687.863
+Skew: 4.655 Prob(JB): 0.00
+Kurtosis: 29.133 Cond. No. 931.
+==============================================================================
+
+Notes:
+[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
+
+=== ITS Model Summary for Recent Spills ===
+
+ OLS Regression Results
+==============================================================================
+Dep. Variable: report_delay R-squared: 0.015
+Model: OLS Adj. R-squared: -0.012
+Method: Least Squares F-statistic: 0.5547
+Date: Fri, 18 Apr 2025 Prob (F-statistic): 0.646
+Time: 21:32:41 Log-Likelihood: -381.54
+No. Observations: 115 AIC: 771.1
+Df Residuals: 111 BIC: 782.1
+Df Model: 3
+Covariance Type: nonrobust
+==============================================================================
+ coef std err t P>|t| [0.025 0.975]
+------------------------------------------------------------------------------
+Intercept 4.6707 1.642 2.844 0.005 1.416 7.925
+time 0.0012 0.043 0.028 0.978 -0.084 0.086
+post_2020 -3.7044 6.757 -0.548 0.585 -17.093 9.684
+time_post 0.0222 0.083 0.268 0.789 -0.142 0.186
+==============================================================================
+Omnibus: 131.076 Durbin-Watson: 2.062
+Prob(Omnibus): 0.000 Jarque-Bera (JB): 2146.748
+Skew: 4.125 Prob(JB): 0.00
+Kurtosis: 22.493 Cond. No. 931.
+==============================================================================
+
+Notes:
+[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
+
+
+
+
+
+
+
+
+
+
+In [31]:
+
+
+import pandas as pd
+import statsmodels.formula.api as smf
+import matplotlib.pyplot as plt
+
+# 1. Group by month and spill type
+spills_gdf['report_month'] = spills_gdf['Initial Report Date'].dt.to_period('M')
+monthly_by_type = spills_gdf.groupby(['report_month', 'spill_type'])['report_delay'].mean().reset_index()
+monthly_by_type['report_month'] = monthly_by_type['report_month'].dt.to_timestamp()
+
+# 2. Create ITS variables
+monthly_by_type['time'] = (monthly_by_type['report_month'] - monthly_by_type['report_month'].min()).dt.days // 30
+monthly_by_type['post_2020'] = (monthly_by_type['report_month'] >= pd.Timestamp('2020-01-01')).astype(int)
+monthly_by_type['time_post'] = monthly_by_type['time'] * monthly_by_type['post_2020']
+
+# 3. Run ITS models separately by spill type
+its_results_by_type = {}
+for stype in monthly_by_type['spill_type'].unique():
+ df = monthly_by_type[monthly_by_type['spill_type'] == stype].copy()
+ model = smf.ols("report_delay ~ time + post_2020 + time_post", data=df).fit()
+ df['predicted'] = model.predict(df)
+ its_results_by_type[stype] = (model, df)
+
+# 4. Plot side-by-side
+fig, axs = plt.subplots(2, 1, figsize=(12, 10), sharex=True)
+
+for ax, (stype, (model, df)) in zip(axs, its_results_by_type.items()):
+ ax.plot(df['report_month'], df['report_delay'], marker='o', label='Observed', alpha=0.6)
+ ax.plot(df['report_month'], df['predicted'], linestyle='--', color='red', label='Predicted (ITS)')
+ ax.axvline(x=pd.Timestamp('2020-01-01'), color='black', linestyle='--', label='Policy Change (2020)')
+ ax.set_title(f"ITS Model for {stype} Spills")
+ ax.set_ylabel("Mean Report Delay (Days)")
+ ax.legend()
+ ax.grid(True)
+
+axs[1].set_xlabel("Month")
+plt.tight_layout()
+plt.show()
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+ITS Comparison: Historical vs. Recent Spills¶
+
+
+Term
+Historical Coef
+Recent Coef
+Interpretation
+
+
+
+
+Intercept
+44.23 (p = 0.001)
+4.67 (p = 0.005)
+Historical spills had much longer delays at baseline
+
+
+time–0.48 (p = 0.18)
+~0.00 (p = 0.98)
+Slight downward trend for historical, flat for recent
+
+
+post_2020–6.83 (p = 0.90)
+–3.70 (p = 0.58)
+No level shift for either group
+
+
+time_post+0.22 (p = 0.75)
++0.02 (p = 0.79)
+No slope change after policy for either group
+
+
+
+Interpretation¶
+- Historical spills started high and declined modestly pre-2020 but with lots of noise.
+- Recent spills were already fast pre-2020 (~5 days), with no real change after policy.
+- High skew and kurtosis in both models underline how volatile and sparse some months were.
+
+How to Report This¶
+We estimated interrupted time series models separately by spill type. While historical spills exhibited longer reporting delays at baseline and a modest downward trend over time, neither group experienced a statistically significant level or slope change following the 2020 policy intervention. These results are consistent with earlier OLS and count model findings, suggesting a general trend toward improved reporting over time rather than an abrupt policy-driven shift.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+Conclusion: Comparative Summary¶
+
+
+
+
+
+
+
+
+In [34]:
+
+
+def extract_regression_results(model, model_name):
+ coefs = model.params
+ ses = model.bse
+ pvals = model.pvalues
+
+ df = pd.DataFrame({
+ 'Coefficient': coefs,
+ 'Std. Error': ses,
+ 'P-Value': pvals
+ })
+ df['Model'] = model_name
+ return df.reset_index().rename(columns={'index': 'Term'})
+
+
+
+
+
+
+
+
+
+
+In [36]:
+
+
+# Collect regression summaries
+results_frames = []
+
+# Add your models — use your actual variable names
+results_frames.append(extract_regression_results(model, 'OLS (Interaction)'))
+results_frames.append(extract_regression_results(nb_model, 'Negative Binomial'))
+results_frames.append(extract_regression_results(its_model, 'ITS (Monthly, All Spills)'))
+
+
+# ITS by spill type (e.g., 'Historical', 'Recent')
+for spill_type, (model, df) in its_results_by_type.items():
+ label = f"ITS (Monthly, {spill_type})"
+ results_frames.append(extract_regression_results(model, label))
+
+# Combine all
+regression_summary_table = pd.concat(results_frames, ignore_index=True)
+
+# Pivot just the coefficients for display
+regression_coef_table = regression_summary_table.pivot_table(
+ index='Term',
+ columns='Model',
+ values='Coefficient',
+ aggfunc='first'
+).round(3)
+
+regression_coef_table
+
+
+
+
+
+
+
+
+
+
+Out[36]:
+
+
+
+
+
+
+Model
+ITS (Monthly, All Spills)
+ITS (Monthly, Historical)
+ITS (Monthly, Recent)
+Negative Binomial
+OLS (Interaction)
+
+
+Term
+
+
+
+
+C(Period)[T.Before 2020]
+NaN
+NaN
+NaN
+0.788
+NaN
+
+
+C(spill_type)[T.Recent]
+NaN
+NaN
+NaN
+-1.531
+NaN
+
+
+C(spill_type)[T.Recent]:C(Period)[T.Before 2020]
+NaN
+NaN
+NaN
+-0.311
+NaN
+
+
+Intercept
+16.666
+44.234
+4.671
+2.605
+4.671
+
+
+post_2020
+-5.090
+-6.832
+-3.704
+NaN
+-3.704
+
+
+time
+-0.167
+-0.476
+0.001
+NaN
+0.001
+
+
+time_post
+0.117
+0.220
+0.022
+NaN
+0.022
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+Regression Summary: Key Coefficients by Model¶
+
+
+Term
+OLS (Interaction)
+Negative Binomial
+ITS (Monthly, All Spills)
+ITS (Historical)
+ITS (Recent)
+
+
+
+
+Intercept
+4.671 ***
+2.605 ***
+16.666 ***
+44.234 ***
+4.671 ***
+
+
+C(spill_type)[T.Recent]-10.606 ***
+-1.531 ***
+—
+—
+—
+
+
+C(Period)[T.Before 2020]16.213 ***
+0.788 ***
+—
+—
+—
+
+
+C(Recent):C(Before 2020)-14.428 ***
+-0.311 ***
+—
+—
+—
+
+
+time—
+—
+-0.167
+-0.476
+0.001
+
+
+post_2020—
+—
+-5.090
+-6.832
+-3.704
+
+
+time_post—
+—
+0.117
+0.220
+0.022
+
+
+
+* p < .05, ** p < .01, *** p < .001
+Note: “—” indicates the variable was not included in that model.
+
+
+
+
+
+
+
+
+
+
+Results Summary¶
To assess the impact of the 2020 reporting policy on spill reporting delays, we estimated a series of regression models using both spill-level and aggregated monthly data.
+The interaction model (OLS) revealed that recent spills were associated with significantly shorter reporting delays compared to historical spills, with an estimated reduction of approximately 10.6 days (p < .001). Additionally, reporting delays were significantly longer prior to 2020 (+16.2 days, p < .001), and the interaction term was also negative and significant (–14.4 days, p < .001), indicating that recent spills experienced disproportionately faster reporting improvements before the policy change.
+The Negative Binomial model, appropriate for overdispersed count data, confirmed this pattern: recent spills were associated with a 78% decrease in expected delay (exp(–1.53) ≈ 0.22, p < .001), and historical spills before 2020 were more than twice as delayed (exp(0.788) ≈ 2.20, p < .001). The interaction effect remained statistically significant.
+Interrupted Time Series (ITS) models using monthly aggregated data showed no statistically significant level or trend change after the 2020 policy implementation. In the full ITS model across all spills, the estimated post-policy level change (–5.09 days) and slope change (+0.12) were not significant (p > .20). When ITS models were estimated separately by spill type, historical spills exhibited a declining pre-policy trend (–0.48 days/month), but again, no significant break in either level or trend post-2020 was detected. Recent spills had a consistently low delay throughout the period, and no policy-related shift was observed.
+Together, these findings suggest that improvements in reporting time were largely underway prior to the 2020 policy, particularly for recent spills, and that the policy may have contributed to continued stabilization rather than an abrupt structural shift. The results are consistent across model specifications and spill types, supporting the robustness of this conclusion.
+
+
+
+
+
+
+
diff --git a/analysis/Analysis Output - HTML Files/analysis7_OLS_NRB_ITS_2021.html b/analysis/Analysis Output - HTML Files/analysis7_OLS_NRB_ITS_2021.html
new file mode 100644
index 0000000..a0244eb
--- /dev/null
+++ b/analysis/Analysis Output - HTML Files/analysis7_OLS_NRB_ITS_2021.html
@@ -0,0 +1,8961 @@
+
+
+
+
+
+analysis7_OLS_NRB_ITS_2021
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+Colorado Spills¶
OLS, NBR, Interrupted Time Series Analysis¶
Author: David P. Adams, PhD¶
Date: 2025-04-18¶
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+Setup¶
+
+
+
+
+
+
+
+
+In [12]:
+
+
+import pandas as pd
+from sqlalchemy import create_engine
+import geopandas as gpd
+from dotenv import load_dotenv
+load_dotenv()
+
+import os
+
+# Database connection details from zshrc environment variables
+# use .env file for local development
+
+
+db_name = 'colorado_spills'
+user = os.getenv('DB_USER')
+password = os.getenv('DB_PASSWORD')
+host = os.getenv('DB_HOST')
+port = os.getenv('DB_PORT')
+
+
+# Create an engine to connect to the PostgreSQL database
+engine = create_engine(f'postgresql+psycopg2://{user}:{password}@{host}:{port}/{db_name}')
+
+# Read in the spills_with_demographics_geog as spills
+spills = pd.read_sql_table('spills_with_demographics_geog', engine)
+
+
+
+
+
+
+
+
+
+
+
+
+/home/dadams/miniconda3/envs/spatial_env2/lib/python3.10/site-packages/pandas/io/sql.py:1725: SAWarning: Did not recognize type 'geometry' of column 'geometry'
+ self.meta.reflect(bind=self.con, only=[table_name], views=True)
+
+
+
+
+
+
+
+
+
+
+In [13]:
+
+
+# use longitude and latitude to create a GeoDataFrame from spills data
+spills['geometry'] = gpd.points_from_xy(spills['Longitude'], spills['Latitude'])
+
+
+
+
+
+
+
+
+
+
+In [14]:
+
+
+spills_gdf = gpd.GeoDataFrame(spills, crs='EPSG:4326')
+
+
+
+
+
+
+
+
+
+
+In [15]:
+
+
+# Ensure date columns are in datetime format
+spills_gdf['Date of Discovery'] = pd.to_datetime(spills_gdf['Date of Discovery'])
+spills_gdf['Initial Report Date'] = pd.to_datetime(spills_gdf['Initial Report Date'])
+
+
+
+
+
+
+
+
+
+
+In [16]:
+
+
+# drop 2024 data for date of discovery and initial report date
+spills_gdf = spills_gdf[spills_gdf['Date of Discovery'].dt.year < 2024]
+spills_gdf = spills_gdf[spills_gdf['Initial Report Date'].dt.year < 2024]
+
+
+
+
+
+
+
+
+
+
+In [17]:
+
+
+# Separate Historical vs. Recent Spills
+historical_spills = spills_gdf[spills_gdf['Spill Type'] == 'Historical']
+recent_spills = spills_gdf[spills_gdf['Spill Type'] == 'Recent']
+
+
+
+
+
+
+
+
+
+
+In [18]:
+
+
+import statsmodels.formula.api as smf
+
+# Create 'Period' column
+spills_gdf['Period'] = spills_gdf['Report Year'].apply(lambda x: 'Before 2020' if x < 2020 else '2020 and After')
+
+
+
+
+
+
+
+
+
+
+In [19]:
+
+
+spills_gdf = spills_gdf.rename(columns={
+ 'Report Delay (Days)': 'report_delay',
+ 'Spill Type': 'spill_type'
+})
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+OLS Regression¶
+
+
+
+
+
+
+
+
+In [20]:
+
+
+import statsmodels.formula.api as smf
+
+model = smf.ols("report_delay ~ C(spill_type) * C(Period)", data=spills_gdf).fit()
+print(model.summary())
+
+
+
+
+
+
+
+
+
+
+
+
+ OLS Regression Results
+==============================================================================
+Dep. Variable: report_delay R-squared: 0.006
+Model: OLS Adj. R-squared: 0.006
+Method: Least Squares F-statistic: 31.12
+Date: Fri, 18 Apr 2025 Prob (F-statistic): 4.78e-20
+Time: 22:00:49 Log-Likelihood: -99827.
+No. Observations: 15990 AIC: 1.997e+05
+Df Residuals: 15986 BIC: 1.997e+05
+Df Model: 3
+Covariance Type: nonrobust
+====================================================================================================================
+ coef std err t P>|t| [0.025 0.975]
+--------------------------------------------------------------------------------------------------------------------
+Intercept 13.5331 2.394 5.654 0.000 8.841 18.225
+C(spill_type)[T.Recent] -10.6062 3.032 -3.498 0.000 -16.550 -4.662
+C(Period)[T.Before 2020] 16.2130 3.417 4.745 0.000 9.516 22.910
+C(spill_type)[T.Recent]:C(Period)[T.Before 2020] -14.4276 4.200 -3.435 0.001 -22.660 -6.196
+==============================================================================
+Omnibus: 45995.174 Durbin-Watson: 1.996
+Prob(Omnibus): 0.000 Jarque-Bera (JB): 3525625525.882
+Skew: 38.959 Prob(JB): 0.00
+Kurtosis: 2302.059 Cond. No. 8.76
+==============================================================================
+
+Notes:
+[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
+
+
+
+
+
+
+
+
+
+
+
+
+Spill Reporting Analysis¶
What the OLS Says¶
This is a 2×2 model:
+
+- Spill Type:
Historical (reference) vs. Recent
+- Period:
2021 and After (reference) vs. Before 2021
+- With their interaction
+
+Coefficient interpretations:¶
+- Intercept (13.53): Avg delay for historical spills after 2021.
+- Recent (–10.61): Recent spills after 2021 had ~10.6 days faster reporting than historical ones.
+- Before 2021 (+16.21): Historical spills before 2021 were reported ~16.2 days slower than those after 2021.
+- Interaction (–14.43): Recent spills before 2021 were ~14.4 days faster than expected given the sum of the main effects. That’s key.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+Group
+Formula
+Result
+
+
+
+
+Historical, After 2021
+Intercept
+13.53 days
+
+
+Recent, After 2021
+Intercept + Recent
+2.92 days
+
+
+Historical, Before 2021
+Intercept + Before 2021
+29.74 days
+
+
+Recent, Before 2021
+Intercept + Before 2021 + Recent + Interaction
+4.70 days
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+Negative Binomial Regression¶
+
+
+
+
+
+
+
+
+In [21]:
+
+
+import statsmodels.formula.api as smf
+import statsmodels.api as sm
+
+# Make sure spill_type and Period are clean, and report_delay is numeric
+spills_gdf = spills_gdf.rename(columns={
+ 'Report Delay (Days)': 'report_delay',
+ 'Spill Type': 'spill_type'
+})
+
+# Fit the Negative Binomial model with interaction
+nb_model = smf.glm(
+ formula="report_delay ~ C(spill_type) * C(Period)",
+ data=spills_gdf,
+ family=sm.families.NegativeBinomial()
+).fit()
+
+# View the results
+print(nb_model.summary())
+
+
+
+
+
+
+
+
+
+
+
+
+ Generalized Linear Model Regression Results
+==============================================================================
+Dep. Variable: report_delay No. Observations: 15990
+Model: GLM Df Residuals: 15986
+Model Family: NegativeBinomial Df Model: 3
+Link Function: Log Scale: 1.0000
+Method: IRLS Log-Likelihood: -47751.
+Date: Fri, 18 Apr 2025 Deviance: 54508.
+Time: 22:00:49 Pearson chi2: 1.72e+06
+No. Iterations: 12 Pseudo R-squ. (CS): 0.4976
+Covariance Type: nonrobust
+====================================================================================================================
+ coef std err z P>|z| [0.025 0.975]
+--------------------------------------------------------------------------------------------------------------------
+Intercept 2.6051 0.020 130.748 0.000 2.566 2.644
+C(spill_type)[T.Recent] -1.5312 0.026 -57.998 0.000 -1.583 -1.479
+C(Period)[T.Before 2020] 0.7876 0.028 27.959 0.000 0.732 0.843
+C(spill_type)[T.Recent]:C(Period)[T.Before 2020] -0.3113 0.036 -8.672 0.000 -0.382 -0.241
+====================================================================================================================
+
+
+
+
+
+
+/home/dadams/miniconda3/envs/spatial_env2/lib/python3.10/site-packages/statsmodels/genmod/families/family.py:1367: ValueWarning: Negative binomial dispersion parameter alpha not set. Using default value alpha=1.0.
+ warnings.warn("Negative binomial dispersion parameter alpha not "
+
+
+
+
+
+
+
+
+
+
+
+
+Negative Binomial Regression Results¶
Key Coefficients (in log count space)¶
+
+
+Term
+Coef (log)
+Exp(Coef)
+Interpretation
+
+
+
+
+Intercept
+2.605
+13.53
+Expected delay for Historical spills after 2021: ~13.5 days
+
+
+Recent
+–1.531
+0.216
+Recent spills (after 2021) take ~78% less time to report
+
+
+Before 2021
+0.788
+2.20
+Historical spills before 2021 took ~2.2x longer
+
+
+Interaction
+–0.311
+0.733
+Recent spills before 2021 were ~27% faster than expected given the base effects
+
+
+
+
+Summary Interpretation¶
+- Recent reporting improved substantially after 2021, even after controlling for spill type.
+- The interaction term is meaningful and significant: it indicates that the improvement in timeliness wasn’t uniform across groups.
+- Model fit is strong for count data:
+- Log-likelihood is stable
+- Pseudo R² ≈ 0.498
+- No convergence issues or overdispersion errors yet (though
alpha=1.0 is default — you can estimate that if needed)
+
+
+
+
+
+
+
+
+
+
+
+In [22]:
+
+
+import numpy as np
+np.exp(nb_model.params)
+
+
+
+
+
+
+
+
+
+
+Out[22]:
+
+Intercept 13.533087
+C(spill_type)[T.Recent] 0.216276
+C(Period)[T.Before 2020] 2.198025
+C(spill_type)[T.Recent]:C(Period)[T.Before 2020] 0.732466
+dtype: float64
+
+
+
+
+
+
+
+
+
+
+
+Group-Level Predictions (Multiplicative Model)¶
Log link in a Negative Binomial model:¶
[
+\text{Expected Delay} = e^{\text{sum of coefficients for that group}}
+]
+Or, since you've already exponentiated the coefficients:
+[
+\text{Delay} = \text{Intercept} \times \text{Multipliers}
+]
+
+
+
+Group
+Calculation
+Predicted Delay (days)
+
+
+
+
+Historical, After 2021
+13.5313.53
+
+
+Recent, After 2021
+13.53 × 0.216≈ 2.93
+
+
+Historical, Before 2021
+13.53 × 2.20≈ 29.77
+
+
+Recent, Before 2021
+13.53 × 0.216 × 2.20 × 0.732≈ 4.26
+
+
+
+
+Interpretation¶
+- After 2021, recent spills are reported ~10.6 days faster than historical ones (2.93 vs. 13.53 days).
+- Historical reporting times worsened before 2021 (29.8 days).
+- The biggest drop is in recent spill reporting after 2021 — a likely policy win.
+
+
+
+
+
+
+
+
+
+In [23]:
+
+
+import matplotlib.pyplot as plt
+import pandas as pd
+
+# Predicted delays by group
+group_labels = [
+ "Historical, After 2021",
+ "Recent, After 2021",
+ "Historical, Before 2021",
+ "Recent, Before 2021"
+]
+
+# From your exponentiated results
+predicted_delays = [
+ 13.53, # Intercept only
+ 13.53 * 0.216, # Intercept * Recent
+ 13.53 * 2.198, # Intercept * Before 2021
+ 13.53 * 0.216 * 2.198 * 0.732 # Intercept * Recent * Before 2021 * Interaction
+]
+
+# Create DataFrame for plotting
+df_plot = pd.DataFrame({
+ "Group": group_labels,
+ "Predicted Delay (Days)": predicted_delays
+})
+
+# Plot
+plt.figure(figsize=(10, 6))
+bars = plt.bar(df_plot["Group"], df_plot["Predicted Delay (Days)"], alpha=0.8)
+plt.title("Predicted Report Delay by Spill Type and Period", fontsize=14)
+plt.ylabel("Predicted Delay (Days)", fontsize=12)
+plt.xticks(rotation=15, ha='right')
+plt.grid(axis='y', linestyle='--', alpha=0.7)
+plt.tight_layout()
+
+# Annotate bars
+for bar in bars:
+ height = bar.get_height()
+ plt.annotate(f"{height:.2f}",
+ xy=(bar.get_x() + bar.get_width() / 2, height),
+ xytext=(0, 3), # 3 points vertical offset
+ textcoords="offset points",
+ ha='center', va='bottom')
+
+plt.show()
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+What the Plot Highlights:¶
+- Historical spills before 2021 had the longest delays by far (~30 days).
+- Recent spills after 2021 saw the shortest (~3 days).
+- The interaction effect shows recent + before 2021 was better than historical, but still far worse than recent + after 2021.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+Interrupted Time Series¶
+
+
+
+
+
+
+
+
+In [24]:
+
+
+import pandas as pd
+import statsmodels.formula.api as smf
+import matplotlib.pyplot as plt
+
+# 1. Convert to monthly time series
+spills_gdf['report_month'] = spills_gdf['Initial Report Date'].dt.to_period('M')
+monthly = spills_gdf.groupby('report_month')['report_delay'].mean().reset_index()
+monthly['report_month'] = monthly['report_month'].dt.to_timestamp()
+
+# 2. Create ITS variables
+monthly['time'] = (monthly['report_month'] - monthly['report_month'].min()).dt.days // 30
+monthly['post_2021'] = (monthly['report_month'] >= pd.Timestamp('2021-01-01')).astype(int)
+monthly['time_post'] = monthly['time'] * monthly['post_2021']
+
+# 3. Run the ITS model
+its_model = smf.ols("report_delay ~ time + post_2021 + time_post", data=monthly).fit()
+print(its_model.summary())
+
+# 4. Predict and plot
+monthly['predicted'] = its_model.predict(monthly)
+
+plt.figure(figsize=(12, 6))
+plt.plot(monthly['report_month'], monthly['report_delay'], marker='o', label='Observed', alpha=0.6)
+plt.plot(monthly['report_month'], monthly['predicted'], linestyle='--', color='red', label='Predicted (ITS)')
+plt.axvline(x=pd.Timestamp('2021-01-01'), color='black', linestyle='--', label='Policy Change (2021)')
+plt.title('Interrupted Time Series: Monthly Report Delay')
+plt.xlabel('Month')
+plt.ylabel('Mean Report Delay (Days)')
+plt.legend()
+plt.grid(True)
+plt.tight_layout()
+plt.show()
+
+
+
+
+
+
+
+
+
+
+
+
+ OLS Regression Results
+==============================================================================
+Dep. Variable: report_delay R-squared: 0.044
+Model: OLS Adj. R-squared: 0.018
+Method: Least Squares F-statistic: 1.708
+Date: Fri, 18 Apr 2025 Prob (F-statistic): 0.169
+Time: 22:00:49 Log-Likelihood: -481.60
+No. Observations: 115 AIC: 971.2
+Df Residuals: 111 BIC: 982.2
+Df Model: 3
+Covariance Type: nonrobust
+==============================================================================
+ coef std err t P>|t| [0.025 0.975]
+------------------------------------------------------------------------------
+Intercept 15.5960 3.601 4.331 0.000 8.461 22.731
+time -0.1178 0.079 -1.486 0.140 -0.275 0.039
+post_2021 -27.6725 25.778 -1.073 0.285 -78.753 23.409
+time_post 0.3024 0.272 1.111 0.269 -0.237 0.842
+==============================================================================
+Omnibus: 127.364 Durbin-Watson: 1.901
+Prob(Omnibus): 0.000 Jarque-Bera (JB): 2109.313
+Skew: 3.925 Prob(JB): 0.00
+Kurtosis: 22.458 Cond. No. 1.41e+03
+==============================================================================
+
+Notes:
+[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
+[2] The condition number is large, 1.41e+03. This might indicate that there are
+strong multicollinearity or other numerical problems.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+🧠 Reinterpreting the ITS (Breakpoint: Jan 2021)¶
📊 Model Coefficients:¶
+
+
+Term
+Coefficient
+p-value
+Interpretation
+
+
+
+
+Intercept
+15.60
+0.000
+Average delay in the first month (baseline)
+
+
+time–0.118
+0.140
+Pre-2021 monthly decline in delay (not significant)
+
+
+post_2021–27.67
+0.285
+Immediate drop in Jan 2021 (not significant, but large)
+
+
+time_post+0.30
+0.269
+Slope increase after Jan 2021 (not significant)
+
+
+
+
+Visual Takeaways from the Plot¶
+- There’s a clear downward trend pre-2021, with sharp outliers but general decline.
+- The predicted line (ITS) reflects a notable drop at 2021 followed by a flattening or slight uptick.
+- This matches your narrative: the policy went into effect in 2020, but the agency's implementation (culture/process) adapted in 2021.
+
+
+Interpretation for Results Section¶
+We re-estimated the interrupted time series model using January 2021 as the intervention point to reflect the agency's internal adaptation to the 2020 regulatory rule change. While the coefficients for both level and slope change were not statistically significant, the direction of effects supports the broader interpretation that reporting delays declined prior to the implementation of the new mission, followed by a period of stabilization or slight reversal. The large (but imprecise) estimated level shift in 2021 may indicate a transitional inflection point, though visual inspection suggests this reflects a continuation of preexisting trends rather than an abrupt effect attributable solely to organizational adaptation.
+
+
+Final Thought¶
Well-supported claim: the formal policy didn’t create an immediate break, but organizational change may have reinforced an ongoing improvement trajectory. That’s a much stronger and more realistic story than trying to force a sharp treatment effect.
+
+
+
+
+
+
+
+
+
+
+Interrupted Time Series Analysis¶
🔍 Model Findings¶
+
+
+Term
+Coefficient
+p-value
+Meaning
+
+
+
+
+Intercept
+16.67
+0.000
+Predicted delay at baseline (month 0)
+
+
+time–0.17
+0.107
+Slight (non-sig.) decline pre-2021
+
+
+post_2021–5.09
+0.753
+No significant level change
+
+
+time_post+0.12
+0.558
+No significant trend change after 2021
+
+
+
+R² = 0.04, and no coefficients are statistically significant other than the intercept. In short: ITS didn’t detect a strong policy effect.
+What the Plot Shows¶
The visual trend still tells a useful story:
+
+- High volatility pre-2021 with outliers and erratic reporting
+- A more stable pattern post-2021
+- No obvious immediate jump or trend break, but lower variance
+
+How to Report This¶
You can say:
+
+“We conducted an Interrupted Time Series analysis using monthly average delays and found no statistically significant level or trend change post-policy. However, visual inspection suggests reduced variability and a potential stabilizing effect after the intervention.”
+
+
+
+
+
+
+
+
+
+
+
+
+ITS by Spill Type¶
+
+
+
+
+
+
+
+
+In [25]:
+
+
+import pandas as pd
+import statsmodels.formula.api as smf
+import matplotlib.pyplot as plt
+
+# Group by month and spill type
+spills_gdf['report_month'] = spills_gdf['Initial Report Date'].dt.to_period('M')
+monthly_by_type = spills_gdf.groupby(['report_month', 'spill_type'])['report_delay'].mean().reset_index()
+monthly_by_type['report_month'] = monthly_by_type['report_month'].dt.to_timestamp()
+
+# ITS variables
+monthly_by_type['time'] = (monthly_by_type['report_month'] - monthly_by_type['report_month'].min()).dt.days // 30
+monthly_by_type['post_2021'] = (monthly_by_type['report_month'] >= pd.Timestamp('2021-01-01')).astype(int)
+monthly_by_type['time_post'] = monthly_by_type['time'] * monthly_by_type['post_2021']
+
+# Run ITS models and print summaries
+its_results_by_type = {}
+for stype in monthly_by_type['spill_type'].unique():
+ df = monthly_by_type[monthly_by_type['spill_type'] == stype].copy()
+ model = smf.ols("report_delay ~ time + post_2021 + time_post", data=df).fit()
+ df['predicted'] = model.predict(df)
+ its_results_by_type[stype] = (model, df)
+
+ print(f"\n=== ITS Model Summary for {stype} Spills ===\n")
+ print(model.summary())
+
+
+
+
+
+
+
+
+
+
+
+
+
+=== ITS Model Summary for Historical Spills ===
+
+ OLS Regression Results
+==============================================================================
+Dep. Variable: report_delay R-squared: 0.037
+Model: OLS Adj. R-squared: 0.011
+Method: Least Squares F-statistic: 1.420
+Date: Fri, 18 Apr 2025 Prob (F-statistic): 0.241
+Time: 22:00:50 Log-Likelihood: -623.83
+No. Observations: 115 AIC: 1256.
+Df Residuals: 111 BIC: 1267.
+Df Model: 3
+Covariance Type: nonrobust
+==============================================================================
+ coef std err t P>|t| [0.025 0.975]
+------------------------------------------------------------------------------
+Intercept 41.4175 12.403 3.339 0.001 16.840 65.995
+time -0.3440 0.273 -1.260 0.210 -0.885 0.197
+post_2021 -72.4348 88.791 -0.816 0.416 -248.380 103.511
+time_post 0.7685 0.937 0.820 0.414 -1.089 2.626
+==============================================================================
+Omnibus: 146.900 Durbin-Watson: 2.047
+Prob(Omnibus): 0.000 Jarque-Bera (JB): 3812.237
+Skew: 4.681 Prob(JB): 0.00
+Kurtosis: 29.607 Cond. No. 1.41e+03
+==============================================================================
+
+Notes:
+[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
+[2] The condition number is large, 1.41e+03. This might indicate that there are
+strong multicollinearity or other numerical problems.
+
+=== ITS Model Summary for Recent Spills ===
+
+ OLS Regression Results
+==============================================================================
+Dep. Variable: report_delay R-squared: 0.011
+Model: OLS Adj. R-squared: -0.016
+Method: Least Squares F-statistic: 0.4120
+Date: Fri, 18 Apr 2025 Prob (F-statistic): 0.745
+Time: 22:00:50 Log-Likelihood: -381.76
+No. Observations: 115 AIC: 771.5
+Df Residuals: 111 BIC: 782.5
+Df Model: 3
+Covariance Type: nonrobust
+==============================================================================
+ coef std err t P>|t| [0.025 0.975]
+------------------------------------------------------------------------------
+Intercept 5.1035 1.511 3.377 0.001 2.109 8.098
+time -0.0171 0.033 -0.513 0.609 -0.083 0.049
+post_2021 -5.7731 10.819 -0.534 0.595 -27.212 15.666
+time_post 0.0565 0.114 0.495 0.622 -0.170 0.283
+==============================================================================
+Omnibus: 133.689 Durbin-Watson: 2.054
+Prob(Omnibus): 0.000 Jarque-Bera (JB): 2325.882
+Skew: 4.223 Prob(JB): 0.00
+Kurtosis: 23.349 Cond. No. 1.41e+03
+==============================================================================
+
+Notes:
+[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
+[2] The condition number is large, 1.41e+03. This might indicate that there are
+strong multicollinearity or other numerical problems.
+
+
+
+
+
+
+
+
+
+
+In [26]:
+
+
+import pandas as pd
+import statsmodels.formula.api as smf
+import matplotlib.pyplot as plt
+
+# 1. Group by month and spill type
+spills_gdf['report_month'] = spills_gdf['Initial Report Date'].dt.to_period('M')
+monthly_by_type = spills_gdf.groupby(['report_month', 'spill_type'])['report_delay'].mean().reset_index()
+monthly_by_type['report_month'] = monthly_by_type['report_month'].dt.to_timestamp()
+
+# 2. Create ITS variables
+monthly_by_type['time'] = (monthly_by_type['report_month'] - monthly_by_type['report_month'].min()).dt.days // 30
+monthly_by_type['post_2021'] = (monthly_by_type['report_month'] >= pd.Timestamp('2021-01-01')).astype(int)
+monthly_by_type['time_post'] = monthly_by_type['time'] * monthly_by_type['post_2021']
+
+# 3. Run ITS models separately by spill type
+its_results_by_type = {}
+for stype in monthly_by_type['spill_type'].unique():
+ df = monthly_by_type[monthly_by_type['spill_type'] == stype].copy()
+ model = smf.ols("report_delay ~ time + post_2021 + time_post", data=df).fit()
+ df['predicted'] = model.predict(df)
+ its_results_by_type[stype] = (model, df)
+
+# 4. Plot side-by-side
+fig, axs = plt.subplots(2, 1, figsize=(12, 10), sharex=True)
+
+for ax, (stype, (model, df)) in zip(axs, its_results_by_type.items()):
+ ax.plot(df['report_month'], df['report_delay'], marker='o', label='Observed', alpha=0.6)
+ ax.plot(df['report_month'], df['predicted'], linestyle='--', color='red', label='Predicted (ITS)')
+ ax.axvline(x=pd.Timestamp('2021-01-01'), color='black', linestyle='--', label='Policy Change (2021)')
+ ax.set_title(f"ITS Model for {stype} Spills")
+ ax.set_ylabel("Mean Report Delay (Days)")
+ ax.legend()
+ ax.grid(True)
+
+axs[1].set_xlabel("Month")
+plt.tight_layout()
+plt.show()
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+Reinterpreting the Spill-Type-Specific ITS (2021 Policy Shift)¶
+
+
+Spill Type
+Key Takeaways
+
+
+
+
+Historical
+Delays trended downward pre-2021, but the level change after 2021 was large (–72 days) and not statistically significant due to wide variance and outliers. The slope change was slightly positive, suggesting a mild rebound or plateau after the internal mission shift.
+
+
+Recent
+Consistently lower delays throughout the time series. No discernible shift in level or trend post-2021, indicating the policy and mission change had little room to improve on already fast reporting.
+
+
+
+Despite large coefficients (especially for post_2021 in historical spills), none of the changes are statistically significant, likely due to high skew and extreme outliers (note the 400+ day delays).
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+Conclusion: Comparative Summary¶
+
+
+
+
+
+
+
+
+In [29]:
+
+
+def extract_regression_results(model, model_name):
+ coefs = model.params
+ ses = model.bse
+ pvals = model.pvalues
+
+ df = pd.DataFrame({
+ 'Coefficient': coefs,
+ 'Std. Error': ses,
+ 'P-Value': pvals
+ })
+ df['Model'] = model_name
+ return df.reset_index().rename(columns={'index': 'Term'})
+
+
+
+
+
+
+
+
+
+
+In [30]:
+
+
+# Collect regression summaries
+results_frames = []
+
+# Add your models — use your actual variable names
+results_frames.append(extract_regression_results(model, 'OLS (Interaction)'))
+results_frames.append(extract_regression_results(nb_model, 'Negative Binomial'))
+results_frames.append(extract_regression_results(its_model, 'ITS (Monthly, All Spills)'))
+
+
+# ITS by spill type (e.g., 'Historical', 'Recent')
+for spill_type, (model, df) in its_results_by_type.items():
+ label = f"ITS (Monthly, {spill_type})"
+ results_frames.append(extract_regression_results(model, label))
+
+# Combine all
+regression_summary_table = pd.concat(results_frames, ignore_index=True)
+
+# Pivot just the coefficients for display
+regression_coef_table = regression_summary_table.pivot_table(
+ index='Term',
+ columns='Model',
+ values='Coefficient',
+ aggfunc='first'
+).round(3)
+
+regression_coef_table
+
+
+
+
+
+
+
+
+
+
+Out[30]:
+
+
+
+
+
+
+Model
+ITS (Monthly, All Spills)
+ITS (Monthly, Historical)
+ITS (Monthly, Recent)
+Negative Binomial
+OLS (Interaction)
+
+
+Term
+
+
+
+
+C(Period)[T.Before 2020]
+NaN
+NaN
+NaN
+0.788
+NaN
+
+
+C(spill_type)[T.Recent]
+NaN
+NaN
+NaN
+-1.531
+NaN
+
+
+C(spill_type)[T.Recent]:C(Period)[T.Before 2020]
+NaN
+NaN
+NaN
+-0.311
+NaN
+
+
+Intercept
+15.596
+41.418
+5.104
+2.605
+5.104
+
+
+post_2021
+-27.672
+-72.435
+-5.773
+NaN
+-5.773
+
+
+time
+-0.118
+-0.344
+-0.017
+NaN
+-0.017
+
+
+time_post
+0.302
+0.768
+0.057
+NaN
+0.057
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+Summary of Regression Models (Post-2021 Breakpoint)¶
The change from a 2020 to a 2021 breakpoint in the ITS models allowed us to align more precisely with the agency’s operational adaptation to the new policy. The OLS interaction and Negative Binomial models were unaffected, as they rely on categorical indicators for period and spill type rather than temporal trends.
+The ITS models showed minor adjustments in intercepts and trend coefficients. For historical spills, the predicted level drop post-2021 was large (–72.4 days) but not statistically significant due to high variance. For recent spills, the model continued to show low and stable delays with no meaningful change after 2021. These updated results reinforce the conclusion that performance improvements were already underway and that the mission shift in 2021 may have stabilized or modestly extended those gains, particularly in the handling of historical spill discoveries.
+
+
+
+
+
+
+
+
+
+
+Regression Summary: Key Coefficients by Model (2021 ITS Breakpoint)¶
+
+
+Term
+OLS (Interaction)
+Negative Binomial
+ITS (Monthly, All Spills)
+ITS (Historical)
+ITS (Recent)
+
+
+
+
+Intercept
+5.104 ***
+2.605 ***
+15.596 ***
+41.418 ***
+5.104 ***
+
+
+C(spill_type)[T.Recent]-10.606 ***
+-1.531 ***
+—
+—
+—
+
+
+C(Period)[T.Before 2020]16.213 ***
+0.788 ***
+—
+—
+—
+
+
+C(Recent):C(Before 2020)-14.428 ***
+-0.311 ***
+—
+—
+—
+
+
+time-0.017
+—
+-0.118
+-0.344
+-0.017
+
+
+post_2021-5.773
+—
+-27.672
+-72.435
+-5.773
+
+
+time_post0.057
+—
+0.302
+0.768
+0.057
+
+
+
+* p < .05, ** p < .01, *** p < .001
+Note: “—” indicates the variable was not included in that model. All ITS models use January 2021 as the intervention point.
+
+
+
+
+
+
+
+
+
+
+Limitations¶
Several limitations should be noted when interpreting these findings. First, the reporting delay data are heavily skewed with a long right tail, especially for historical spills. While count models and quantile regression (excluded here) partially address this, residual outliers still influence trend estimates in time-series analyses. Second, the categorization into "recent" vs. "historical" spills reflects underlying reporting practices, not random treatment assignment, making causal inference inherently limited. Third, the ITS models rely on monthly aggregation and assume linear trends before and after the policy change; nonlinear dynamics or unmeasured covariates may obscure more complex patterns. Finally, the formal rule change in 2020 and the internal mission shift in 2021 introduce ambiguity about when the policy was truly “implemented,” and thus, both breakpoints carry interpretive uncertainty.
+
+
+
+
+
+
+
diff --git a/analysis/analysis5.ipynb b/analysis/analysis5.ipynb
index c901f93..b42c440 100644
--- a/analysis/analysis5.ipynb
+++ b/analysis/analysis5.ipynb
@@ -1,5 +1,37 @@
{
"cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Colorado Spills Project\n",
+ "## Outline\n",
+ "\n",
+ "1. **Data Import and Preparation**\n",
+ " - Import necessary libraries and load data from the database.\n",
+ " - Convert data into GeoDataFrame and ensure proper formatting.\n",
+ "\n",
+ "2. **Data Cleaning**\n",
+ " - Handle missing or incorrect data.\n",
+ " - Filter out data for the year 2024.\n",
+ "\n",
+ "3. **Exploratory Data Analysis**\n",
+ " - Separate historical and recent spills.\n",
+ " - Analyze reporting delays and trends over time.\n",
+ "\n",
+ "4. **Visualization**\n",
+ " - Create spatial distribution maps.\n",
+ " - Plot trends and differences in reporting delays.\n",
+ "\n",
+ "5. **Statistical Analysis**\n",
+ " - Perform t-tests to compare response times.\n",
+ " - Use regression models to analyze differences and trends.\n",
+ "\n",
+ "6. **Results and Interpretation**\n",
+ " - Summarize findings in markdown cells.\n",
+ " - Highlight key takeaways and policy implications.\n"
+ ]
+ },
{
"cell_type": "code",
"execution_count": 3,
@@ -1413,9 +1445,17 @@
"\n",
"The fixed-effects model confirmed our core finding: **recent spills diverged significantly from historical trends in key policy-relevant years**. In 2020, historical spills experienced a 33-day increase in reporting delay, while recent spills showed a 33-day smaller delay—resulting in a significant 66-day relative difference (p = 0.006). Significant differences were also observed in 2015 and 2019, further reinforcing the conclusion that recent spills were less affected by structural or reporting challenges.\n",
"\n",
- "### **Summary**\n",
+ "### **Reasons DiD Model Is Inappropriate**\n",
"\n",
- "Across all models, results consistently show that **the 2021 policy change was associated with improved reporting timeliness**, particularly for historical spills. While recent spills already had shorter delays and showed smaller improvements, the fixed-effects model demonstrated that these trends were distinct and consistent with industry adaptation and improved detection technologies. Together, the findings suggest that the policy change not only reduced delays but also helped standardize reporting practices across different types of spills."
+ "A difference-in-differences (DiD) design assumes parallel pre-policy trends between treated and control groups. In our case, *historical* and *recent* spills are fundamentally different: recent spills are reported shortly after discovery, while historical spills reflect delayed discoveries. Because their reporting dynamics are structurally distinct and not plausibly parallel over time, DiD is not an appropriate design.\n",
+ "\n",
+ "### **Key Takeaways**\n",
+ "- **Historical spills**: Showed significant improvement in reporting delays post-2020, with a 16.2-day reduction.\n",
+ "- **Recent spills**: Experienced a smaller, marginally significant reduction of 1.8 days.\n",
+ "- **DiD model**: Indicated that historical spills improved significantly more than recent spills, but the parallel trends assumption is violated.\n",
+ "- **Fixed-effects model**: Confirmed that recent spills diverged from historical trends, particularly in key years (2015, 2019, 2020), reinforcing the idea that recent spills are less affected by structural challenges.\n",
+ "- **Policy implications**: The 2021 policy change appears to have standardized reporting practices and improved timeliness, particularly for historical spills, suggesting that regulatory changes can effectively address reporting delays.\n",
+ "\n"
]
}
],
diff --git a/analysis/analysis7.ipynb b/analysis/analysis7.ipynb
new file mode 100644
index 0000000..e67be9b
--- /dev/null
+++ b/analysis/analysis7.ipynb
@@ -0,0 +1,1234 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "id": "1b85622b",
+ "metadata": {},
+ "source": [
+ "# Colorado Spills \n",
+ "## OLS, NBR, Interrupted Time Series Analysis\n",
+ "### Author: [David P. Adams, PhD](https://dadams.io)\n",
+ "#### Date: 2025-04-18"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "d456a510",
+ "metadata": {},
+ "source": [
+ "---\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "9814ecd7",
+ "metadata": {},
+ "source": [
+ "# Setup"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "id": "eba79ece",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "/home/dadams/miniconda3/envs/spatial_env2/lib/python3.10/site-packages/pandas/io/sql.py:1725: SAWarning: Did not recognize type 'geometry' of column 'geometry'\n",
+ " self.meta.reflect(bind=self.con, only=[table_name], views=True)\n"
+ ]
+ }
+ ],
+ "source": [
+ "import pandas as pd\n",
+ "from sqlalchemy import create_engine\n",
+ "import geopandas as gpd\n",
+ "from dotenv import load_dotenv\n",
+ "load_dotenv()\n",
+ "\n",
+ "import os\n",
+ "\n",
+ "# Database connection details from zshrc environment variables\n",
+ "# use .env file for local development\n",
+ "\n",
+ "\n",
+ "db_name = 'colorado_spills'\n",
+ "user = os.getenv('DB_USER')\n",
+ "password = os.getenv('DB_PASSWORD')\n",
+ "host = os.getenv('DB_HOST')\n",
+ "port = os.getenv('DB_PORT')\n",
+ "\n",
+ "\n",
+ "# Create an engine to connect to the PostgreSQL database\n",
+ "engine = create_engine(f'postgresql+psycopg2://{user}:{password}@{host}:{port}/{db_name}')\n",
+ "\n",
+ "# Read in the spills_with_demographics_geog as spills\n",
+ "spills = pd.read_sql_table('spills_with_demographics_geog', engine)\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "id": "da21b89e",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# use longitude and latitude to create a GeoDataFrame from spills data\n",
+ "spills['geometry'] = gpd.points_from_xy(spills['Longitude'], spills['Latitude'])"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "id": "789eaccb",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "spills_gdf = gpd.GeoDataFrame(spills, crs='EPSG:4326') \n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "id": "d041a419",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "\n",
+ "# Ensure date columns are in datetime format\n",
+ "spills_gdf['Date of Discovery'] = pd.to_datetime(spills_gdf['Date of Discovery'])\n",
+ "spills_gdf['Initial Report Date'] = pd.to_datetime(spills_gdf['Initial Report Date'])\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "id": "3106f3c5",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# drop 2024 data for date of discovery and initial report date\n",
+ "spills_gdf = spills_gdf[spills_gdf['Date of Discovery'].dt.year < 2024]\n",
+ "spills_gdf = spills_gdf[spills_gdf['Initial Report Date'].dt.year < 2024]\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "id": "139de8e5",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# Separate Historical vs. Recent Spills\n",
+ "historical_spills = spills_gdf[spills_gdf['Spill Type'] == 'Historical']\n",
+ "recent_spills = spills_gdf[spills_gdf['Spill Type'] == 'Recent']\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 37,
+ "id": "ab3680c1",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "import statsmodels.formula.api as smf\n",
+ "\n",
+ "# Create 'Period' column\n",
+ "spills_gdf['Period'] = spills_gdf['Report Year'].apply(lambda x: 'Before 2020' if x < 2020 else '2020 and After')\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 38,
+ "id": "2ccf5c3a",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "spills_gdf = spills_gdf.rename(columns={\n",
+ " 'Report Delay (Days)': 'report_delay',\n",
+ " 'Spill Type': 'spill_type'\n",
+ "})\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "1da721cd",
+ "metadata": {},
+ "source": [
+ "---"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "1b9a3766",
+ "metadata": {},
+ "source": [
+ "# OLS Regression"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 16,
+ "id": "1b722ad7",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " OLS Regression Results \n",
+ "==============================================================================\n",
+ "Dep. Variable: report_delay R-squared: 0.006\n",
+ "Model: OLS Adj. R-squared: 0.006\n",
+ "Method: Least Squares F-statistic: 31.12\n",
+ "Date: Fri, 18 Apr 2025 Prob (F-statistic): 4.78e-20\n",
+ "Time: 21:03:48 Log-Likelihood: -99827.\n",
+ "No. Observations: 15990 AIC: 1.997e+05\n",
+ "Df Residuals: 15986 BIC: 1.997e+05\n",
+ "Df Model: 3 \n",
+ "Covariance Type: nonrobust \n",
+ "====================================================================================================================\n",
+ " coef std err t P>|t| [0.025 0.975]\n",
+ "--------------------------------------------------------------------------------------------------------------------\n",
+ "Intercept 13.5331 2.394 5.654 0.000 8.841 18.225\n",
+ "C(spill_type)[T.Recent] -10.6062 3.032 -3.498 0.000 -16.550 -4.662\n",
+ "C(Period)[T.Before 2020] 16.2130 3.417 4.745 0.000 9.516 22.910\n",
+ "C(spill_type)[T.Recent]:C(Period)[T.Before 2020] -14.4276 4.200 -3.435 0.001 -22.660 -6.196\n",
+ "==============================================================================\n",
+ "Omnibus: 45995.174 Durbin-Watson: 1.996\n",
+ "Prob(Omnibus): 0.000 Jarque-Bera (JB): 3525625525.882\n",
+ "Skew: 38.959 Prob(JB): 0.00\n",
+ "Kurtosis: 2302.059 Cond. No. 8.76\n",
+ "==============================================================================\n",
+ "\n",
+ "Notes:\n",
+ "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
+ ]
+ }
+ ],
+ "source": [
+ "import statsmodels.formula.api as smf\n",
+ "\n",
+ "model = smf.ols(\"report_delay ~ C(spill_type) * C(Period)\", data=spills_gdf).fit()\n",
+ "print(model.summary())\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "e52e1b08",
+ "metadata": {},
+ "source": [
+ "## Spill Reporting Analysis\n",
+ "### **What the OLS Says**\n",
+ "\n",
+ "This is a 2×2 model:\n",
+ "- **Spill Type**: `Historical` (reference) vs. `Recent`\n",
+ "- **Period**: `2020 and After` (reference) vs. `Before 2020`\n",
+ "- With their **interaction**\n",
+ "\n",
+ "#### Coefficient interpretations:\n",
+ "- **Intercept (13.53)**: Avg delay for *historical* spills *after 2020*.\n",
+ "- **Recent (–10.61)**: Recent spills after 2020 had **~10.6 days faster** reporting than historical ones.\n",
+ "- **Before 2020 (+16.21)**: Historical spills before 2020 were reported **~16.2 days slower** than those after 2020.\n",
+ "- **Interaction (–14.43)**: Recent spills before 2020 were **~14.4 days faster** than expected given the sum of the main effects. That’s key."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "3585fb92",
+ "metadata": {},
+ "source": [
+ "| Group | Formula | Result |\n",
+ "|-----------------------------|----------------------------------------------------------------------|------------|\n",
+ "| Historical, After 2020 | Intercept | 13.53 days |\n",
+ "| Recent, After 2020 | Intercept + Recent | 2.92 days |\n",
+ "| Historical, Before 2020 | Intercept + Before 2020 | 29.74 days |\n",
+ "| Recent, Before 2020 | Intercept + Before 2020 + Recent + Interaction | 4.70 days |\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "0cef322c",
+ "metadata": {},
+ "source": [
+ "---"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "ee48e488",
+ "metadata": {},
+ "source": [
+ "# Negative Binomial Regression"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 23,
+ "id": "813a6d9c",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " Generalized Linear Model Regression Results \n",
+ "==============================================================================\n",
+ "Dep. Variable: report_delay No. Observations: 15990\n",
+ "Model: GLM Df Residuals: 15986\n",
+ "Model Family: NegativeBinomial Df Model: 3\n",
+ "Link Function: Log Scale: 1.0000\n",
+ "Method: IRLS Log-Likelihood: -47751.\n",
+ "Date: Fri, 18 Apr 2025 Deviance: 54508.\n",
+ "Time: 21:14:16 Pearson chi2: 1.72e+06\n",
+ "No. Iterations: 12 Pseudo R-squ. (CS): 0.4976\n",
+ "Covariance Type: nonrobust \n",
+ "====================================================================================================================\n",
+ " coef std err z P>|z| [0.025 0.975]\n",
+ "--------------------------------------------------------------------------------------------------------------------\n",
+ "Intercept 2.6051 0.020 130.748 0.000 2.566 2.644\n",
+ "C(spill_type)[T.Recent] -1.5312 0.026 -57.998 0.000 -1.583 -1.479\n",
+ "C(Period)[T.Before 2020] 0.7876 0.028 27.959 0.000 0.732 0.843\n",
+ "C(spill_type)[T.Recent]:C(Period)[T.Before 2020] -0.3113 0.036 -8.672 0.000 -0.382 -0.241\n",
+ "====================================================================================================================\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "/home/dadams/miniconda3/envs/spatial_env2/lib/python3.10/site-packages/statsmodels/genmod/families/family.py:1367: ValueWarning: Negative binomial dispersion parameter alpha not set. Using default value alpha=1.0.\n",
+ " warnings.warn(\"Negative binomial dispersion parameter alpha not \"\n"
+ ]
+ }
+ ],
+ "source": [
+ "import statsmodels.formula.api as smf\n",
+ "import statsmodels.api as sm\n",
+ "\n",
+ "# Make sure spill_type and Period are clean, and report_delay is numeric\n",
+ "spills_gdf = spills_gdf.rename(columns={\n",
+ " 'Report Delay (Days)': 'report_delay',\n",
+ " 'Spill Type': 'spill_type'\n",
+ "})\n",
+ "\n",
+ "# Fit the Negative Binomial model with interaction\n",
+ "nb_model = smf.glm(\n",
+ " formula=\"report_delay ~ C(spill_type) * C(Period)\",\n",
+ " data=spills_gdf,\n",
+ " family=sm.families.NegativeBinomial()\n",
+ ").fit()\n",
+ "\n",
+ "# View the results\n",
+ "print(nb_model.summary())\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "89571910",
+ "metadata": {},
+ "source": [
+ "## Negative Binomial Regression Results\n",
+ "### **Key Coefficients (in log count space)**\n",
+ "\n",
+ "| Term | Coef (log) | Exp(Coef) | Interpretation |\n",
+ "|------|------------|-----------|----------------|\n",
+ "| **Intercept** | 2.605 | 13.53 | Expected delay for **Historical spills after 2020**: ~13.5 days |\n",
+ "| **Recent** | –1.531 | 0.216 | Recent spills (after 2020) take ~78% *less time* to report |\n",
+ "| **Before 2020** | 0.788 | 2.20 | Historical spills before 2020 took ~2.2x *longer* |\n",
+ "| **Interaction** | –0.311 | 0.733 | Recent spills before 2020 were ~27% *faster* than expected given the base effects |\n",
+ "\n",
+ "---\n",
+ "\n",
+ "### Summary Interpretation\n",
+ "\n",
+ "- **Recent reporting improved substantially** after 2020, even after controlling for spill type.\n",
+ "- The **interaction term** is meaningful and significant: it indicates that the improvement in timeliness wasn’t uniform across groups.\n",
+ "- Model fit is strong for count data: \n",
+ " - Log-likelihood is stable \n",
+ " - Pseudo R² ≈ 0.498 \n",
+ " - No convergence issues or overdispersion errors yet (though `alpha=1.0` is default — you can estimate that if needed)\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 24,
+ "id": "4343c273",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "Intercept 13.533087\n",
+ "C(spill_type)[T.Recent] 0.216276\n",
+ "C(Period)[T.Before 2020] 2.198025\n",
+ "C(spill_type)[T.Recent]:C(Period)[T.Before 2020] 0.732466\n",
+ "dtype: float64"
+ ]
+ },
+ "execution_count": 24,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "import numpy as np\n",
+ "np.exp(nb_model.params)\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "3459a91b",
+ "metadata": {},
+ "source": [
+ "### **Group-Level Predictions (Multiplicative Model)**\n",
+ "#### Log link in a Negative Binomial model:\n",
+ "\n",
+ "\\[\n",
+ "\\text{Expected Delay} = e^{\\text{sum of coefficients for that group}}\n",
+ "\\]\n",
+ "\n",
+ "Or, since you've already exponentiated the coefficients:\n",
+ "\n",
+ "\\[\n",
+ "\\text{Delay} = \\text{Intercept} \\times \\text{Multipliers}\n",
+ "\\]\n",
+ "\n",
+ "| Group | Calculation | Predicted Delay (days) |\n",
+ "|------------------------------|----------------------------------------------------------|-------------------------|\n",
+ "| **Historical, After 2020** | `13.53` | 13.53 |\n",
+ "| **Recent, After 2020** | `13.53 × 0.216` | ≈ 2.93 |\n",
+ "| **Historical, Before 2020** | `13.53 × 2.20` | ≈ 29.77 |\n",
+ "| **Recent, Before 2020** | `13.53 × 0.216 × 2.20 × 0.732` | ≈ 4.26 |\n",
+ "\n",
+ "---\n",
+ "\n",
+ "### Interpretation\n",
+ "\n",
+ "- After 2020, recent spills are reported **~10.6 days faster** than historical ones (2.93 vs. 13.53 days).\n",
+ "- Historical reporting times **worsened before 2020** (29.8 days).\n",
+ "- The **biggest drop** is in **recent spill reporting after 2020** — a likely policy win."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 25,
+ "id": "f09f08d0",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "import matplotlib.pyplot as plt\n",
+ "import pandas as pd\n",
+ "\n",
+ "# Predicted delays by group\n",
+ "group_labels = [\n",
+ " \"Historical, After 2020\",\n",
+ " \"Recent, After 2020\",\n",
+ " \"Historical, Before 2020\",\n",
+ " \"Recent, Before 2020\"\n",
+ "]\n",
+ "\n",
+ "# From your exponentiated results\n",
+ "predicted_delays = [\n",
+ " 13.53, # Intercept only\n",
+ " 13.53 * 0.216, # Intercept * Recent\n",
+ " 13.53 * 2.198, # Intercept * Before 2020\n",
+ " 13.53 * 0.216 * 2.198 * 0.732 # Intercept * Recent * Before 2020 * Interaction\n",
+ "]\n",
+ "\n",
+ "# Create DataFrame for plotting\n",
+ "df_plot = pd.DataFrame({\n",
+ " \"Group\": group_labels,\n",
+ " \"Predicted Delay (Days)\": predicted_delays\n",
+ "})\n",
+ "\n",
+ "# Plot\n",
+ "plt.figure(figsize=(10, 6))\n",
+ "bars = plt.bar(df_plot[\"Group\"], df_plot[\"Predicted Delay (Days)\"], alpha=0.8)\n",
+ "plt.title(\"Predicted Report Delay by Spill Type and Period\", fontsize=14)\n",
+ "plt.ylabel(\"Predicted Delay (Days)\", fontsize=12)\n",
+ "plt.xticks(rotation=15, ha='right')\n",
+ "plt.grid(axis='y', linestyle='--', alpha=0.7)\n",
+ "plt.tight_layout()\n",
+ "\n",
+ "# Annotate bars\n",
+ "for bar in bars:\n",
+ " height = bar.get_height()\n",
+ " plt.annotate(f\"{height:.2f}\",\n",
+ " xy=(bar.get_x() + bar.get_width() / 2, height),\n",
+ " xytext=(0, 3), # 3 points vertical offset\n",
+ " textcoords=\"offset points\",\n",
+ " ha='center', va='bottom')\n",
+ "\n",
+ "plt.show()\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "0ef55d60",
+ "metadata": {},
+ "source": [
+ "### What the Plot Highlights:\n",
+ "- **Historical spills before 2020** had the longest delays by far (~30 days).\n",
+ "- **Recent spills after 2020** saw the shortest (~3 days).\n",
+ "- The interaction effect shows **recent + before 2020** was better than historical, but still far worse than recent + after 2020."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "daa4b710",
+ "metadata": {},
+ "source": []
+ },
+ {
+ "cell_type": "markdown",
+ "id": "a9251055",
+ "metadata": {},
+ "source": [
+ "---\n",
+ "\n",
+ "# Compare OLS and Negative Binomial\n",
+ "\n",
+ "### **Comparison Table**\n",
+ "\n",
+ "| Group | OLS Prediction (Days) | NBR Prediction (Days) |\n",
+ "|---------------------------|------------------------|------------------------|\n",
+ "| Historical, After 2020 | 13.53 | 13.53 |\n",
+ "| Recent, After 2020 | 2.92 | 2.93 |\n",
+ "| Historical, Before 2020 | 29.74 | 29.77 |\n",
+ "| Recent, Before 2020 | 4.70 | 4.26 |\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "3238be11",
+ "metadata": {},
+ "source": [
+ "---"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "f6c679b7",
+ "metadata": {},
+ "source": [
+ "# Interrupted Time Series"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 30,
+ "id": "fc8dc178",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " OLS Regression Results \n",
+ "==============================================================================\n",
+ "Dep. Variable: report_delay R-squared: 0.040\n",
+ "Model: OLS Adj. R-squared: 0.014\n",
+ "Method: Least Squares F-statistic: 1.535\n",
+ "Date: Fri, 18 Apr 2025 Prob (F-statistic): 0.209\n",
+ "Time: 21:29:45 Log-Likelihood: -481.86\n",
+ "No. Observations: 115 AIC: 971.7\n",
+ "Df Residuals: 111 BIC: 982.7\n",
+ "Df Model: 3 \n",
+ "Covariance Type: nonrobust \n",
+ "==============================================================================\n",
+ " coef std err t P>|t| [0.025 0.975]\n",
+ "------------------------------------------------------------------------------\n",
+ "Intercept 16.6659 3.929 4.241 0.000 8.879 24.452\n",
+ "time -0.1671 0.103 -1.627 0.107 -0.371 0.036\n",
+ "post_2020 -5.0897 16.165 -0.315 0.753 -37.122 26.942\n",
+ "time_post 0.1165 0.198 0.588 0.558 -0.276 0.509\n",
+ "==============================================================================\n",
+ "Omnibus: 125.885 Durbin-Watson: 1.893\n",
+ "Prob(Omnibus): 0.000 Jarque-Bera (JB): 1986.555\n",
+ "Skew: 3.880 Prob(JB): 0.00\n",
+ "Kurtosis: 21.825 Cond. No. 931.\n",
+ "==============================================================================\n",
+ "\n",
+ "Notes:\n",
+ "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
+ ]
+ },
+ {
+ "data": {
+ "image/png": "iVBORw0KGgoAAAANSUhEUgAABKUAAAJOCAYAAABm7rQwAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8hTgPZAAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOzdd3gUZdcG8Hu2bwoJLSQhIQlN6UUEAemEJqCCoB8gIEWsgFSxUFSqiLQXEERApKg0BQEpCoiAVKUpNaFDQgvpW2a+P5IZdpPdZDekbbh/18VFdmZ29myeLBnOnOc8giRJEoiIiIiIiIiIiPKRqqADICIiIiIiIiKixw+TUkRERERERERElO+YlCIiIiIiIiIionzHpBQREREREREREeU7JqWIiIiIiIiIiCjfMSlFRERERERERET5jkkpIiIiIiIiIiLKd0xKERERERERERFRvmNSioiIiIiIiIiI8h2TUkRE5FGWLl0KQRBw+PBht5+blJSE8ePHY9euXbkfWAHYvHkzxo8fnyfnbt68OZo3b+50//jx4yEIQrZ/mjdvjujoaAiCgKVLl+ZJrDkhSRJWr16NJk2aICAgAAaDASEhIWjbti2+/vrrXH+98PBw9O3bN9fP6y55LARBcPqz069fP+WYvHT9+nWMHz8ef//9d6Z9ffv2hY+Pj0vnyeq9uMv2+yMIAlQqFYoXL45WrVph27ZtufIaeW3fvn0YP3487t+/79LxGT/LXl5eymdhzpw5iI+Pz3Es8r/X0dHROT4HEREVbUxKERHRYyMpKQkTJkwoUkmpCRMmFMhrDxgwAPv371f+rFu3DgDw7rvv2m2fN28egoKCsH//fjz33HMFEqsjY8aMwf/93/+hSpUq+Prrr7FlyxZ89tlnKFOmDH766adcf73169fj448/zvXz5pSvry+WLl0KURTttickJODHH39EsWLF8jyG69evY8KECQ6TUgVN/jn+448/MH36dJw7dw4dOnTAnj17Cjq0bO3btw8TJkxwOSkl27p1K/bv34+tW7di+vTpKFeuHEaNGoVq1arhn3/+yZtgiYjosacp6ACIiIg8ndlshiAI0Ggy/1pNSkqCl5dXAUSVt0JCQhASEqI8lishypUrh2eeeSbT8Y62FZTk5GTMnDkTvXv3xsKFC+329e3bN1Oi5lFfy2g0ok6dOrl2ztzw8ssv4+uvv8bOnTsRGRmpbP/+++9htVrxwgsv4LvvvivACAuW7c9x48aNUalSJTRr1gyLFy9G06ZNCzg6x5KTk2EwGHL8/KeeegqlSpVSHr/yyit455130KxZM3Tu3Blnz56FXq/PjVCJiIgUrJQiIiKPJ0/1OX/+PDp06AAfHx+EhoZi+PDhSE1NBZCWNCldujQAYMKECcpUFdspVefOnUOPHj0QEBAAvV6PKlWq4H//+5/da+3atQuCIGD58uUYPnw4ypYtC71ej/PnzytxnDhxAm3atIGvry9atWoFwPn0rYzT5OTzf/fddxg2bBgCAwNhNBrRrFkzHDt2zO49y7HZTr2Rk0OSJGHevHmoXbs2jEYjihcvjpdeegkXL160e31JkjBt2jSEhYXBYDCgbt262LJlS47GwRlH0/fkKUPHjx9Ht27d4OfnhxIlSmDYsGGwWCw4c+YM2rVrB19fX4SHh2PatGmZzvvgwQOMGDECERER0Ol0KFu2LIYOHYrExMQs40lMTERqaiqCgoIc7lep7C+PTCYTPvvsMzz55JPQ6/UoXbo0XnvtNcTGxtodFx4ejo4dO2LdunWoU6cODAaDUsnmaPxdjf/HH39EgwYN4OfnBy8vL5QvXx79+vXL8j1m54knnkCjRo3wzTff2G3/5ptv0KVLF/j5+WV6jiiKmDZtmvJ9CAgIQO/evXH16lW745o3b47q1avj0KFDaNKkiRLzlClTlITfrl278PTTTwMAXnvtNadTCrP6TDsSHR0NjUaDyZMnZ9q3Z88eCIKAH3/80aXvka169eoBAG7dumW3/ebNmxg0aBBCQkKg0+kQERGBCRMmwGKx2MUkCAKmTZuGiRMnoly5cjAYDKhXrx527tyZ6bX27t2LVq1awdfXF15eXmjUqBF++eUXu2PkaXHbtm1Dv379ULp0aXh5eWHMmDEYOXIkACAiIkL5vua0OrRWrVr48MMPcfnyZXz//fd2+3bs2IFWrVqhWLFi8PLyQuPGjR2+n4y2b9+O559/HiEhITAYDKhYsSIGDRqE27dvK8f88ccfEAQBq1atyvT8b7/9FoIg4NChQzl6T0REVLgwKUVEREWC2WxG586d0apVK/z000/o168fvvzyS0ydOhUAEBQUhK1btwIA+vfvr0wvk6dUnT59Gk8//TROnjyJL774Aps2bcJzzz2HwYMHO5wiN2bMGFy+fBkLFizAxo0bERAQACAtgdG5c2e0bNkSP/30U46n133wwQe4ePEivv76a3z99de4fv06mjdvriSVPv74Y7z00ksAYDddTk60DBo0CEOHDkXr1q2xYcMGzJs3D6dOnUKjRo3s/mM9YcIEjB49GpGRkdiwYQPefPNNDBw4EGfOnMlR3O7q3r07atWqhbVr12LgwIH48ssv8d577+GFF17Ac889h/Xr16Nly5YYPXq0MkUQSKtAa9asGZYtW4bBgwdjy5YtGD16NJYuXYrOnTtDkiSnr1mqVClUrFgR8+bNw4wZM/Dff/85PV4URTz//POYMmUKevTogV9++QVTpkzB9u3b0bx5cyQnJ9sdf/ToUYwcORKDBw/G1q1b0bVrV4fndTX+/fv34+WXX0b58uWxevVq/PLLLxg7dqxd0gNISwS52wOqf//+2LBhA+7duwcAOHPmDPbt24f+/fs7PP7NN99UflZ+/vlnfPrpp9i6dSsaNWpkl1AA0pI1PXv2RK9evfDzzz+jffv2GDNmjFJ9VbduXSxZsgQA8NFHHyk/vwMGDFDOkd1n2pHw8HB07twZCxYsgNVqtds3d+5cBAcH48UXX3Tr+wQAUVFRAIDKlSvbvcf69evj119/xdixY7Flyxb0798fkydPxsCBAzOdY+7cudi6dStmzpyJ7777DiqVCu3bt8f+/fuVY3bv3o2WLVsiLi4OixcvxqpVq+Dr64tOnTplSgoBaf2/tFotli9fjjVr1uDNN9/Eu+++CwBYt26d8n2tW7eu2+9Z1rlzZwCwm7r43XffoU2bNihWrBiWLVuGH374ASVKlEDbtm2zTUxduHABDRs2xPz587Ft2zaMHTsWf/31F5599lmYzWYAQJMmTVCnTp1MNwWAtO/j008/rSQ1iYjIw0lEREQeZMmSJRIA6dChQ8q2Pn36SACkH374we7YDh06SE888YTyODY2VgIgjRs3LtN527ZtK4WEhEhxcXF229955x3JYDBId+/elSRJkn7//XcJgNS0adNM55Dj+OabbzLtCwsLk/r06ZNpe7NmzaRmzZopj+Xz161bVxJFUdkeHR0tabVaacCAAcq2t99+W3L0q3z//v0SAOmLL76w237lyhXJaDRKo0aNkiRJku7duycZDAbpxRdftDvuzz//lADYxZWdqKgoCYD0+eefO923ZMkSZdu4ceMcxli7dm0JgLRu3Tplm9lslkqXLi116dJF2TZ58mRJpVLZ/RxIkiStWbNGAiBt3rw5y3gPHjwolStXTgIgAZB8fX2ljh07St9++63d933VqlUSAGnt2rV2zz906JAEQJo3b56yLSwsTFKr1dKZM2cyvV7G8Xc1/unTp0sApPv372f5flq2bCmp1eosj5Ek+3GKj4+XfHx8pLlz50qSJEkjR46UIiIiJFEUM/1s/fvvvxIA6a233rI7319//SUBkD744ANlW7NmzSQA0l9//WV3bNWqVaW2bdsqj+Xvoe3PhczVz7QkSZk+0/JnaP369cq2a9euSRqNRpowYYJL35+pU6dKZrNZSklJkf7++2+pYcOGUlBQkBQVFaUcO2jQIMnHx0e6dOmS3TnkMTt16pTdOYODg6Xk5GTluAcPHkglSpSQWrdurWx75plnpICAACk+Pl7ZZrFYpOrVq0shISHKz6b872Dv3r0zvYfPP/9cAmAXa1bkz2JsbKzD/cnJyRIAqX379pIkSVJiYqJUokQJqVOnTnbHWa1WqVatWlL9+vWVbXKczmIRRVEym83SpUuXJADSTz/9lOm5x44dU7YdPHhQAiAtW7bMpfdGRESFHyuliIioSBAEAZ06dbLbVrNmTVy6dCnb56akpGDnzp148cUX4eXlBYvFovzp0KEDUlJScODAAbvnOKuAyW6fq3r06GFX+RIWFoZGjRrh999/z/a5mzZtgiAI6NWrl917CQwMRK1atZSpPPv370dKSgp69uxp9/xGjRohLCzskd+DKzp27Gj3uEqVKhAEAe3bt1e2aTQaVKxY0W4sN23ahOrVq6N27dp277Ft27YuTVd6+umncf78eWzduhUffPABGjZsiJ07d6J37952lUqbNm2Cv78/OnXqZPc6tWvXRmBgYKbXqVmzpl01jTOuxi9Xg3Tv3h0//PADrl275vB8O3fuzFQ9lR0fHx9069YN33zzDSwWC7799ltlKl1G8s9dximI9evXR5UqVTJVxwQGBqJ+/fp221z9PMpy+plu3rw5atWqZVdls2DBAgiCgNdff92l1x49ejS0Wi0MBgNq166NkydPYuPGjQgPD1eO2bRpE1q0aIHg4GC7MZR/dnfv3m13zi5dutj1fJIroPbs2QOr1YrExET89ddfeOmll+xWHlSr1Xj11Vdx9erVTBWMufFvTXakDFWE+/btw927d9GnTx+79y2KItq1a4dDhw5lOYU2JiYGb7zxBkJDQ6HRaKDVapV/b/7991/luP/7v/9DQECA3TjOmTMHpUuXxssvv5zL75KIiAoKk1JERFQkeHl5ZWryq9frkZKSku1z79y5A4vFgjlz5kCr1dr96dChAwBkmp7krB+Rl5dXrqxcFhgY6HDbnTt3sn3urVu3IEkSypQpk+n9HDhwQHkv8rmcvVZ+KFGihN1jnU7ncCx1Op3dWN66dQvHjx/P9P58fX0hSVKm8XJEq9Wibdu2mDhxIn799VdcuXIFzZs3x6ZNm5S+Wrdu3cL9+/eh0+kyvdbNmzdd/rnIyNX4mzZtig0bNsBisaB3794ICQlB9erVHfbayYn+/fvj6NGjmDhxImJjYx32PQMe/qw4en/BwcGZfi5LliyZ6Ti9Xp9pumNWHuUzPXjwYOzcuRNnzpyB2WzGokWL8NJLL7n8cz1kyBAcOnQIe/fuxfTp02E2m/H888/bvc9bt25h48aNmcawWrVqADL/m+Hsc2YymZCQkIB79+5BkiSn32MAmb7Prv68PQo5CSjHIE//femllzK996lTp0KSJNy9e9fhuURRRJs2bbBu3TqMGjUKO3fuxMGDB5Wkv+3Ph16vx6BBg7By5Urcv38fsbGx+OGHHzBgwAA2XCciKkK4+h4RET32ihcvrlQjvP322w6PiYiIsHvsrH+Ps+0Gg8Fhg+bbt2/brXglu3nzpsNtjv6zn1GpUqUgCAL++OMPh/95k7fJ53L2WrZVIYVNqVKlYDQaMzXqtt3vrpIlS2Lo0KHYtWsXTp48iQ4dOqBUqVIoWbKk0o8sI19fX7vHrvZ1cif+559/Hs8//zxSU1Nx4MABTJ48GT169EB4eDgaNmzo4rtzrHHjxnjiiSfwySefIDIyEqGhoQ6Pk39Wbty4YbfqIgBcv349R9/vvNSjRw+MHj0a//vf//DMM8/g5s2bTj/bjoSEhCjNzRs3bozAwED06tUL48aNw9y5cwGkjVHNmjUxceJEh+eQkzgyZ58znU4HHx8faDQaqFQq3LhxI9Nx169fV17Tlrt9xHLi559/BgBlQQY5hjlz5jhdVbNMmTIOt588eRL//PMPli5dij59+ijbz58/7/D4N998E1OmTME333yDlJQUWCwWvPHGGzl9K0REVAgxKUVERI8NORmTsVrDy8sLLVq0wLFjx1CzZk3odLpcf+3w8HAcP37cbtvZs2dx5swZh/+hX7VqFYYNG6b8p/PSpUvYt28fevfurRxj+36MRqOyvWPHjpgyZQquXbuG7t27O43pmWeegcFgwIoVK+ymAe3btw+XLl0q1Empjh07YtKkSShZsmSmhGF2zGYzHjx44DDBJ08fkhMKHTt2xOrVq2G1WtGgQYNHDzxdTuLX6/Vo1qwZ/P398euvv+LYsWOPnJQC0hqNr1mzJsukTcuWLQGkNbi2bTB96NAh/Pvvv/jwww/dfl1nn8fcYDAY8Prrr2Pu3LnYt28fateujcaNG+f4fD179sTXX3+NRYsWYeTIkQgLC0PHjh2xefNmVKhQAcWLF8/2HOvWrcPnn3+uVH/Fx8dj48aNaNKkCdRqNby9vdGgQQOsW7cO06dPVz7Toijiu+++Q0hIiEtTQ3Pz+/rPP/9g0qRJCA8PV/4tady4Mfz9/XH69Gm88847bp1P/vcsY7L8q6++cnh8UFAQunXrhnnz5sFkMqFTp04oV65cDt4JEREVVkxKERHRY8PX1xdhYWH46aef0KpVK5QoUQKlSpVCeHg4Zs2ahWeffRZNmjTBm2++ifDwcMTHx+P8+fPYuHEjfvvtt0d67VdffRW9evXCW2+9ha5du+LSpUuYNm0aSpcu7fD4mJgYvPjiixg4cCDi4uIwbtw4GAwGjBkzRjmmRo0aAICpU6eiffv2UKvVqFmzJho3bozXX38dr732Gg4fPoymTZvC29sbN27cwN69e1GjRg28+eabKF68OEaMGIHPPvsMAwYMQLdu3XDlyhWMHz8+36bv5dTQoUOxdu1aNG3aFO+99x5q1qwJURRx+fJlbNu2DcOHD3eaRIqLi0N4eDi6deuG1q1bIzQ0FAkJCdi1axdmzZqFKlWqoEuXLgCAV155BStWrECHDh0wZMgQ1K9fH1qtFlevXsXvv/+O559/Pkerubka/9ixY3H16lW0atUKISEhuH//PmbNmgWtVotmzZop52vVqhV2797tdl8pAOjVqxd69eqV5TFPPPEEXn/9dcyZM0dZNS46Ohoff/wxQkND8d5777n9uhUqVIDRaMSKFStQpUoV+Pj4IDg4OFOFUU699dZbmDZtGo4cOYKvv/76kc83depUNGjQAJ9++im+/vprfPLJJ9i+fTsaNWqEwYMH44knnkBKSgqio6OxefNmLFiwwK6qTK1WIzIyEsOGDYMoipg6dSoePHhgt0Ln5MmTERkZiRYtWmDEiBHQ6XSYN28eTp48iVWrVrlUGSX/uzBr1iz06dMHWq0WTzzxRKaqvoyOHDkCPz8/mM1mXL9+HTt37sTy5csREBCAjRs3Ksl6Hx8fzJkzB3369MHdu3fx0ksvISAgALGxsfjnn38QGxuL+fPnO3yNJ598EhUqVMD7778PSZJQokQJbNy4Edu3b3ca15AhQ5TPsrxiIxERFR1MShER0WNl8eLFGDlyJDp37ozU1FT06dMHS5cuRdWqVXH06FF8+umn+OijjxATEwN/f39UqlRJ6Sv1KHr06IHr169jwYIFWLJkCapXr4758+fb/YfU1qRJk3Do0CG89tprePDgAerXr4/Vq1ejQoUKduf8888/MW/ePHzyySeQJAlRUVEIDw/HV199hWeeeQZfffUV5s2bB1EUERwcjMaNG9s1oP7kk0/g7e2NefPmYfny5XjyySexYMECTJ8+/ZHfc17y9vbGH3/8gSlTpmDhwoWIioqC0WhEuXLl0Lp16yyrvIoVK4YJEyZg586d+OCDD3Dr1i0IgoCIiAgMHToUo0ePhpeXF4C0RMLPP/+MWbNmYfny5Zg8eTI0Gg1CQkLQrFkzJQGQV/E3aNAAhw8fxujRoxEbGwt/f3/Uq1cPv/32m9K7CACsViusVmuOYnHV/PnzUaFCBSxevBj/+9//4Ofnh3bt2mHy5MkuTSvNyMvLC9988w0mTJiANm3awGw2Y9y4cRg/fnyuxFu2bFk8++yzOH78OHr06PHI56tfvz66deuGZcuWYcyYMahQoQIOHz6MTz/9FJ9//jmuXr0KX19fREREoF27dpmqp9555x2kpKRg8ODBiImJQbVq1fDLL7/YVXA1a9YMv/32G8aNG4e+fftCFEXUqlULP//8c6ZFAZxp3rw5xowZg2XLlmHRokUQRRG///67Mv3OmXbt2gFIq2IqUaIEatSogalTp+K1117LlNDq1asXypUrh2nTpmHQoEGIj49HQEAAateu7bQvGZDWx23jxo0YMmQIBg0aBI1Gg9atW2PHjh1OK6Dq16+P8PBwGI1GtGrVyqXvAREReQ5ByrikBhERERWYXbt2oUWLFvjxxx/x0ksvFXQ4RB4rJiYGYWFhePfddzFt2rQCiyM6OhoRERH4/PPPMWLEiAKLw1MdP35cWU3xrbfeKuhwiIgol7FSioiIiIiKjKtXr+LixYv4/PPPoVKpMGTIkIIOiXLgwoULuHTpEj744AMEBQVlWYFFRESeS1XQARARERER5Zavv/4azZs3x6lTp7BixQqULVu2oEOiHPj0008RGRmJhIQE/Pjjj8qUWiIiKlo4fY+IiIiIiIiIiPIdK6WIiIiIiIiIiCjfMSlFRERERERERET5jkkpIiIiIiIiIiLKd1x9D4Aoirh+/Tp8fX0hCEJBh0NERERERERE5LEkSUJ8fDyCg4OhUjmvh2JSCsD169cRGhpa0GEQERERERERERUZV65cQUhIiNP9TEoB8PX1BZD2zSpWrJiy3Ww2Y9u2bWjTpg20Wm1BhUcu4Fh5Do6VZ+A4eQ6OlWfgOHkOjpXn4Fh5Bo6T5+BYeQZPGacHDx4gNDRUybc4w6QUoEzZK1asWKaklJeXF4oVK1aoB5s4Vp6EY+UZOE6eg2PlGThOniE1NRWjR4/GpUuX0KlTJ/j4+BR0SJQFfq48A8fJc3CsPIOnjVN2LZKYlCIiIiIiAmCxWLBgwQLlayIiIspbXH2PiIiIiIiIiIjyHZNSRERERERERESU7zh9j4iIiIiIiMgFoijCZDIVdBh5wmw2Q6PRICUlBVartaDDIScKyzhptVqo1epHPg+TUkRERERERETZMJlMiIqKgiiKBR1KnpAkCYGBgbhy5Uq2zamp4BSmcfL390dgYOAjxcGkFBEREREREVEWJEnCjRs3oFarERoaCpWq6HXCEUURCQkJ8PHxKZLvr6goDOMkSRKSkpIQExMDAAgKCsrxuZiUIiIiIiIiIsqCxWJBUlISgoOD4eXlVdDh5Al5aqLBYGBSqhArLONkNBoBADExMQgICMjxVD4mpYiIiIiIkHaBffbsWfz+++/KxTYREQCld49OpyvgSIgKDzlBazabc5yUYvqTiIiIiAiASqVCeHg4ypQpwyoBInKooHv4EBUmufF54G9bIiIiIiIiIiLKd0xKEREREREhbWWt999/H0uXLi2yS74TETlTvnx5zJ8/v6DDyDXh4eGYOXNmQYdB2WBSioiIiIgIaT0xZsyYgQ0bNsBsNhd0OERUBImihIuxCfjnyn1cjE2AKEr58rpXrlxB//79ERwcDJ1Oh7CwMAwZMgR37tzJl9cncoaNzomIiIiIiIjy2MlrcVh79CrOxyQg1SxCr1WhYoAPutYNQfWyfnn2uhcvXkTDhg1RuXJlrFq1ChERETh16hRGjhyJLVu24MCBAyhRokSevb4zVqsVgiCwh99jjqNPRERERERElIdOXovD7J3ncOJqHPyNOoSX8oa/UYcTV9O2n7wWl2ev/fbbb0On02Hbtm1o1qwZypUrh/bt22PHjh24du0aPvzwQ+XYhIQE9OzZEz4+PggODsacOXPszjV+/HiUK1cOer0ewcHBGDx4sLLPZDJh1KhRKFu2LLy9vdGgQQPs2rVL2b906VL4+/tj06ZNqFq1KvR6PRYtWgSDwYD79+/bvc7gwYPRrFkz5fG+ffvQtGlTGI1GhIaGYvDgwUhMTFT2x8TEoFOnTjAajYiIiMCKFSty6btHeY1JKSIiIiIiIiI3SJKEVIvVpT/JJgt+OHwFdxJTUb6UF7x0KgiQ4KVToXwpL9xJTMWPR64g2WRx6XyS5PqUv7t37+LXX3/FW2+9BaPRaLcvMDAQPXv2xPfff6+cc86cOahZsyaOHj2KMWPG4L333sP27dsBAGvWrMGXX36Jr776CufOncOGDRtQo0YN5XyvvfYa/vzzT6xevRrHjx9Ht27d0K5dO5w7d045JikpCZMnT8bXX3+NU6dOoVevXvD398fatWuVY6xWK3744Qf07NkTAHDixAm0bdsWXbp0wfHjx/H9999j7969eOedd5Tn9O3bF9HR0fjtt9+wZs0azJs3DzExMW6MKBUUTt8jjyeKEqJuJ+JSAhB1OxEVy/hBpeJSrURERERElDdMVhHjfz7t0rEPks04cukedBoV7idZMu03W0X8evIWbsebUMyozfZ84ztXhV6jdum1z507B0mSUKVKFYf7q1Spgnv37iE2NhYAUL9+fYwePRoqlQqVK1fGn3/+iS+//BKRkZG4fPkyAgMD0bp1a2i1WpQrVw7169cHAFy4cAGrVq3C1atXERwcDAAYMWIEtm7diiVLlmDSpElp79Vsxrx581CrVi0lhpdffhkrV65E//79AQA7d+7EvXv30K1bNwDA559/jh49emDo0KEAgEqVKmH27Nlo1qwZ5s+fj8uXLyvTEBs0aAAAWLx4sdP3TIULk1Lk0eR52eduxuNGjAoHN/2HSoG+eT4vm4iIiIiIyBVmqwirKEHj5Ma5WiXAKkowW8V8jgxKhZQgpMUmJ5lkDRs2VFaw69atG2bOnIny5cujXbt26NChAzp16gSNRoOjR49CkiRUrlzZ7vmpqakoWbKk8lin06FmzZp2x/Ts2RMNGzbE9evXERwcjBUrVqBDhw4oXrw4AODIkSM4f/683ZQ8SZIgiiKioqJw9uxZaDQa1KtXT9n/5JNPwt/f/9G+OZQvmJQijyXPy76baEKZYnpYDICflxYnrsbh2r1kDG5ViYkpIiIiIiLKdTq1CuM7V3Xp2Kjbifh002n4G7Xw0Wf+L3hCqgX3k80Y1qYyIkp5u/TarqpYsSIEQcDp06fxwgsvZNr/33//oXjx4ihVqpTTc8gJq9DQUJw5cwbbt2/Hjh078NZbb+Hzzz/H7t27IYoi1Go1jhw5ArXavorLx8dH+dpoNCrnk9WvXx8VKlTA6tWr8eabb2L9+vVYsmSJsl8URQwaNMiuf5WsXLlyOHPmjF2c5FmYlCKPJIoS1h69iruJJlQM8IEkSbgnAD56DXwDtDgfk4B1R6+halAxTuUjIiIilxiNRhw7dgx//PFHpt4rRES2BEFweQpd5QBfVC7jixNX4+AboLVLnkiShFsPUlEzxB+VA3xz/f8uJUuWRGRkJObNm4f33nvP7t+2mzdvYsWKFejdu7cS06FDh+yef+DAATz55JPKY6PRiM6dO6Nz5854++238eSTT+LEiROoU6cOrFYrYmJi0KRJE7fj7NGjB1asWIGQkBCoVCo899xzyr66devi1KlTqFixosPnVqlSBRaLBYcPH1Yqvc6cOZOpeToVTmx0Th4p+k4izsckIMgvLdMe8yAF901p+wRBQJCfEedi4hF9JzHrExERERGlU6lUqFatGsqVK8clyoko16hUArrWDUEJbx3OxyQgIcUCqyghIcWC8zEJKOGtQ5e6ZfPsZvrcuXORmpqKtm3bYs+ePbhy5Qq2bt2KyMhIlC1bFhMnTlSO/euvv/D555/j7Nmz+N///ocff/wRQ4YMAZC2et7ixYtx8uRJXLx4EcuXL4fRaERYWBgqV66Mnj17onfv3li3bh2ioqJw6NAhTJ06FZs3b842xp49e+Lo0aOYOHEiXnrpJRgMBmXf6NGjsX//frz99tv4+++/ce7cOfz888949913AQBPPPEE2rVrh4EDB+Kvv/7CkSNHMGDAAN5c8BD8bUseKT7FglSzCKNODVGScPleMu6kChDFtDnRRp0aqWYR8SmZGwkSERERERHlp+pl/TC4VSXUCPHD/WQTom8n4n6yCTVD/PO87UilSpVw+PBhVKhQAS+//DIqVKiA119/HS1atMD+/ftRokQJ5dh33nkHR44cQZ06dfDpp5/iiy++QNu2bQEA/v7+WLRoERo3boyaNWti586d2Lhxo9IzasmSJejduzeGDx+OJ554Ap07d8Zff/2F0NBQl2J8+umncfz4cWXVPVnNmjWxe/dunDt3Dk2aNEGdOnXw8ccfIygoSDlmyZIlCA0NRbNmzdClSxe8/vrrCAgIyI1vH+UxTt8jj+Rr0ECvVSHZZIVRp4IkAZAAEWlJqWSTFXqtCr4G/ogTERGRa0wmEz799FOcO3dOWV2KiCi3VC/rh6pBxRB9JxHxKRb4GjQIL+mdL+1GwsLC7Po0OXLx4kU8ePAAxYoVc1gt+sILLzjsSyXTarWYMGECJkyY4HB/37590bdvX6fPP3jwoNN9Tz/9NLZt2+Z0f2BgIDZt2mS37dVXX3V6PBUe/B87eaTwkt6oGOCDE1fjEF7SS9kuSWnzsm/EJaNmiD/CS2bfKJCIiIgISFuq/LPPPgMALFiwoICjIaKiSKUSUL60T/YHEj0mmJQijyTPy752LxkXYhNhtoqQJCAhxYK7Scl5Pi+biIiIiIiIiB5NgfaU2rNnDzp16oTg4GAIgoANGzYo+8xmM0aPHo0aNWrA29sbwcHB6N27N65fv253jtTUVLz77rsoVaoUvL290blzZ1y9ejWf3wkVBHle9pNBvjBZRaSIQFyyOV/mZRMRERERERHRoynQpFRiYiJq1aqFuXPnZtqXlJSEo0eP4uOPP8bRo0exbt06nD17Fp07d7Y7bujQoVi/fj1Wr16NvXv3IiEhAR07doTVas2vt0EFqHpZPwxtXRlPlfNHmI+EYZEV8dFzVZiQIiIiIiIiIirkCnT6Xvv27dG+fXuH+/z8/LB9+3a7bXPmzEH9+vVx+fJllCtXDnFxcVi8eDGWL1+O1q1bAwC+++47hIaGYseOHcoqAVT0+Rq08NUCZf2NnLJHRERERERE5AE8qqdUXFwcBEGAv78/AODIkSMwm81o06aNckxwcDCqV6+Offv2OU1KpaamIjU1VXn84MEDAGlTBs1ms7Jd/tp2GxU+qSYzRFEEAKSYzByvQo6fK8/AcfIcHCvPwHHyDBmvAzlehRs/V56hqIyT2WyGJEkQRVH5v0dRI0mS8ndRfY9FQWEaJ1EUIUkSzGYz1Gq13T5XP/Mek5RKSUnB+++/jx49eqBYsWIAgJs3b0Kn06F48eJ2x5YpUwY3b950eq7Jkyc7XKZy27Zt8PLyyrQ9Y8UWFS53UoBr19Jmou7Z8wdKGgo4IHIJP1eegePkOThWnoHjVLilpKQoX//2228wGHhR4Qn4ufIMnj5OGo0GgYGBSEhIgMlkKuhw8lR8fHxBh0AuKAzjZDKZkJycjD179sBisdjtS0pKcukcHpGUMpvNeOWVVyCKIubNm5ft8ZIkQRCcT+EaM2YMhg0bpjx+8OABQkND0aZNGyXhJb/u9u3bERkZCa1W+2hvgvLMpTtJ+HfPRVy7ehUNGzdGxTLsJ1WY8XPlGThOnoNj5Rk4Tp7BarViz549+Ouvv9ChQwcmpQo5fq48Q1EZp5SUFFy5cgU+Pj5F9t8GSZIQHx8PX1/fLP8/TQWrMI1TSkoKjEYjmjZtmulzIc9Iy06hT0qZzWZ0794dUVFR+O233+ySRoGBgTCZTLh3755dtVRMTAwaNWrk9Jx6vR56vT7Tdq1W6/AfSmfbqXAQ1GqoVGmVUiqVhmPlIfi58gwcJ8/BsfIMHKfCTavV4plnnsHdu3dhMBg4Vh6CnyvP4OnjZLVaIQgCVCqV8n+PokaeCia/TyqcCtM4qVQqCILg8PPt6ue9UP+kyQmpc+fOYceOHShZsqTd/qeeegpardauFPTGjRs4efJklkkpKlqsouTwayIiIiIiIsof48ePR+3atZXHffv2xQsvvJDvcURHR0MQBPz9999ZHnfmzBkEBgbm6TS41NRUlCtXDkeOHMmz1/B0BZqUSkhIwN9//638sERFReHvv//G5cuXYbFY8NJLL+Hw4cNYsWIFrFYrbt68iZs3bypzeP38/NC/f38MHz4cO3fuxLFjx9CrVy/UqFFDWY2Pij67pJTEpBQRERHljMlkwhdffIH169cX+Z4xRPR46Nu3LwRBUKpZypcvjxEjRiAxMTHPX3vWrFlYunSpS8e6mkjKTR9++CHefvtt+Pr6AgB27doFQRBw//59u++bsz9A2iytQYMGoVy5ctDr9QgMDETbtm2xf/9+AGmztEaMGIHRo0fn2/vyNAU6fe/w4cNo0aKF8lju89SnTx+MHz8eP//8MwDYZVsB4Pfff0fz5s0BAF9++SU0Gg26d++O5ORktGrVCkuXLs3U+Z2KLlZKERERUW4wm80YM2YMAGDmzJkFGwwRUS5p164dlixZArPZjD/++AMDBgxAYmIi5s+fn+nY3Fwl0c+v8Pb6vXr1Kn7++Wen/9bPmjULU6ZMUR4HBQVhyZIlaNeund1xXbt2hdlsxrJly1C+fHncunULO3fuxN27d5VjevbsiZEjR+Lff/9FlSpV8uT9eLICrZRq3rw5JEnK9Gfp0qUIDw93uE+SJCUhBQAGgwFz5szBnTt3kJSUhI0bNyI0NLTg3hTlO9GmOkpkpRQREREREZFCruAJDQ1Fjx490LNnT2zYsAHAwyl333zzDSpWrIgyZcpAkiTExcXh9ddfR0BAAIoVK4aWLVvin3/+sTvvlClTUKZMGfj6+qJ///52K5gCmafviaKIqVOnomLFitDr9ShXrhwmTpwIAIiIiAAA1KlTB4Ig2P2ff8mSJahSpQoMBgOefPLJTIufHTx4EHXq1IHBYEC9evVw7NixbL8nP/zwA2rVqoWQkBCH+/38/BAYGKj8AQB/f3+7bffv38fevXsxdepUtGjRAmFhYahfvz7GjBmD5557TjlXyZIl0ahRI6xatSrbuB5Hhb7ROVF2LKyUIiIiIiKigpDVNDi1GrBdkSyrY1UqwGjM/lhvb/fic8BoNNpVRJ0/fx4//PADfvzxRyQnJwMAnnvuOZQoUQKbN2+Gn58fvvrqK7Rq1Qpnz55FiRIl8MMPP2DcuHH43//+hyZNmmD58uWYPXs2ypcv7/R1x4wZg0WLFuHLL7/Es88+ixs3buC///4DkJZYql+/Pnbs2IFq1apBp9MBABYtWoRx48Zh7ty5qFOnDo4dO4aBAwfC29sbffr0QWJiIjp27IiWLVviu+++Q1RUFIYMGZLt92DPnj2oV6/eo3wb4ePjAx8fH2zYsAHPPPOMw8XUZPXr18cff/zxSK9XVDEpRR5PFG0rpQowECIiIiIierz4+Djf16ED8MsvDx8HBABJSY6PbdYM2LXr4ePwcOD27czHPeLMkIMHD2LlypVo1aqVss1kMmH58uUoWbIkHjx4gN9//x0nTpxATEyMkmiZPn06NmzYgDVr1uD111/HzJkz0a9fPwwYMAAA8Nlnn2HHjh2ZqqVk8fHxmDVrFubOnYs+ffoAACpUqIBnn30WAFC6dGkAaVVFcmUSAHz66af44osv0KVLFwBpFVWnT5/GV199hT59+ij9p7/55ht4eXmhWrVquHr1Kt58880svw/R0dF46qmncvItVGg0GixduhQDBw7EggULULduXTRr1gyvvPIKatasaXds2bJlER0d/UivV1QV6tX3iFzBnlJERERERESObdq0CT4+PjAYDGjYsCGaNm2KOXPmKPvDwsKUpBAAHD16FAkJCShZsqRSDeTj44OoqChcuHABAPDvv/+iYcOGdq+T8bGtf//9F6mpqXbJsOzExsbiypUr6N+/v10cn332mV0ctWrVgpeXl0txyJKTk2GwrWLLoa5du+L69ev4+eef0bZtW+zatQt169bN1ODdaDQiyVlC8jHHSinyeFb2lCIiIiIiooKQkOB8X8bFt2JinB+rylAvkotVNS1atMD8+fOh1WoRHBwMrVZrt987w5RAURQRFBSEXbaVW+n8/f1zFIPRdmqii0RRBJA2ha9BgwZ2++SFzaQc/v+vVKlSuHfvXo6em5HBYEBkZCQiIyMxduxYDBgwAOPGjUPfvn2VY+7evWuX+KOHmJQij8dKKSIiIiIiKhDu9HjKq2OzPZU3Klas6PLxderUwc2bN6HRaBAeHu7wmCpVquDAgQPo3bu3su3AgQNOz1mpUiUYjUbs3LlTmfJnS+4hZbValW1lypRB2bJlcfHiRfTs2dPheatWrYrly5cjOTlZSXxlFYftezx9+nS2x+VE1apVlUbyspMnT6JOnTp58nqejtP3yOMxKUVERES5wWAwYPv27fj0009zZVoHEZEnat26NRo2bIgXXngBv/76K6Kjo7Fv3z589NFHOHz4MABgyJAh+Oabb/DNN9/g7NmzGDduHE6dOuX0nAaDAaNHj8aoUaPw7bff4sKFCzhw4AAWL14MAAgICIDRaMTWrVtx69YtxMXFAUhbHXDy5MmYNWsWzp49ixMnTmDJkiWYMWMGAKBHjx5QqVTo378/Tp8+jc2bN2P69OnZvse2bdti//79dkkwd925c0dpsH78+HFERUXhxx9/xLRp0/D888/bHfvHH3+gTZs2OX6tooyVUuTx7JJSnL5HREREOaRWq9GsWTMkJiYqU0OIiB43giBg8+bN+PDDD9GvXz/ExsYiMDAQTZs2RZkyZQAAL7/8Mi5cuIDRo0cjJSUFXbt2xZtvvolff/3V6Xk//vhjaDQajB07FtevX0dQUBDeeOMNAGlNw2fPno1PPvkEY8eORZMmTbBr1y4MGDAAXl5e+PzzzzFq1Ch4e3ujRo0aGDp0KIC0FfA2btyIN954A3Xq1EHVqlUxdepUdO3aNcv32KFDB2i1WuzYsQNt27bN0ffJx8cHDRo0wJdffokLFy7AbDYjNDQUAwcOxAcffKAct3//fsTFxeGll17K0esUdYKU00mYRciDBw/g5+eHuLg4FCtWTNluNpuxefNm5QeWCqdtp25i57+3cOXyZfSOrIt2NcoWdEiUBX6uPAPHyXNwrDwDx8lzcKw8B8fKMxSVcUpJSUFUVBQiIiKKbCWlKIp48OABihUrBlXGHldF0Lx58/DTTz9lmUjLDd26dUOdOnXsElWPojCNU1afC2d5loyK/k8aFXm2zc3Te+ERERERuc1sNmP+/PnYvHkzzGZzQYdDRER56PXXX0fTpk0RHx+fZ6+RmpqKWrVq4b333suz1/B0nL5HHs9qk4ji9D0iIiLKKZPJhCFDhgAApk6darfEOBERFS0ajQYffvhhnr6GXq/HRx99lKev4elYKUUez2pXKcWkFBEREREREZEnYFKKPJ7VZs4eK6WIiIiIiIiIPAOTUuTxbKfvsVKKiIiIiIiIyDMwKUUezzYRZWFSioiIiIiIiMgjMClFHs+upxSn7xERERERERF5BCalyOPZVkfZTuUjIiIiIiIiosJLU9ABED0q2+l7bHROREREOaXX67FhwwYcPnwYer2+oMMhIiIq8lgpRR7PvlKKSSkiIiLKGY1Ggw4dOqBevXrQaHjvloho165dEAQB9+/fBwAsXboU/v7+BRoTAIwfPx61a9cu6DAeiclkQsWKFfHnn38WdCgOjRgxAoMHD87z12FSijyeyKQUERERERFRJn379oUgCBAEAVqtFuXLl8eIESOQmJiYo/O9/PLLOHv2bC5HmdnatWvRvHlz+Pn5wcfHBzVr1sQnn3yCu3fv5vlr55eFCxciLCwMjRs3BgBER0ejf//+iIiIgNFoRIUKFTBu3DiYTCa7512+fBmvvPIKfH19UapUKQwePNjumF27duH5559HUFAQvL29Ubt2baxYsSLT6+/evRtPPfUUDAYDypcvjwULFtjtHzVqFJYsWYKoqKg8ePcPMSlFHo+NzomIiCg3mM1mfPvtt9i5cyfMZnNBh0NElCvatWuHGzdu4OLFi/jss88wb948jBgxIkfnMhqNCAgIyOUI7X344Yd4+eWX8fTTT2PLli04efIkvvjiC/zzzz9Yvnx5nr52fpozZw4GDBigPP7vv/8giiK++uornDp1Cl9++SUWLFiADz74QDnGarWiU6dOSEpKwp49e7B69WqsXbsWw4cPV47Zt28fatasibVr1+L48ePo168fevfujY0bNyrHREVFoUOHDmjSpAmOHTuGDz74AIMHD8batWuVYwICAtCmTZtMyarcxqQUeTwrK6WIiIgoF5hMJgwYMABz5szJdGeaiMhT6fV6BAYGIjQ0FD169EDPnj2xYcMGAEBqaioGDx6MgIAAeHl5oV27djh06JDTczmavvfzzz+jXr16MBgMKFWqFLp06QIA+OSTT1CjRo1M53jqqacwduxYh+c/ePAgJk2ahC+++AKff/45GjVqhPDwcERGRmLt2rXo06eP3fHLly9HeHg4/Pz88MorryA+Pl7Zt3XrVjz77LPw9/dHyZIl0bFjR1y4cEHZHx0dDUEQsG7dOrRo0QJeXl6oVasW9u/fb/caixYtQmhoKLy8vPDiiy9ixowZmb4HGzdutKs6mjBhAiwWi9Pv49GjR3H+/Hk899xzyrZ27dphyZIlaNOmDcqXL4/OnTtjxIgRWLdunXLMtm3bcPr0aXz11VeoU6cOWrdujS+++AKLFi3CgwcPAAAffPABPv30UzRq1AgVKlTA4MGD0a5dO6xfv145z4IFC1CuXDnMnDkTVapUwYABA9CvXz9Mnz7dLs7OnTtj1apVTt9HbmBSijyebSJKZFKKiIiIiIjySWJiotM/KSkpLh+bnJzs0rG5wWg0KtWgo0aNwtq1a7Fs2TIcPnwY5cuXR/v27V2eJvfLL7+gS5cueO6553Ds2DHs3LkT9erVAwD069cPp0+ftktyHT9+HMeOHUPfvn0dnm/FihXw8fHBW2+95XC/bTLowoUL2LBhAzZt2oRNmzZh9+7dmDJlirI/MTERw4YNw6FDh7Bz506oVCq8+OKLEEX7Jds//PBDjBgxAn///TcqV66M//u//1MSSn/++SfeeOMNDBkyBH///TciIyMxceJEu+f/+uuv6NWrFwYPHqwkjJYuXZrpOFt79uxB5cqVUaxYMafHAEBcXBxKlCihPN6/fz+qV6+OoKAgZVvbtm2RmpqKI0eOuHWeNm3a2B3Ttm1bHD582K5SuH79+rhy5QouXbqUZZyPgh0cyePZVUoxJ0VERERERPnEx8fH6b4OHTrgl19+UR4HBAQgKSnJ4bHNmjXDrl27lMfh4eG4fft2puOkR2xXcvDgQaxcuRKtWrVCYmIi5s+fj6VLl6J9+/YQRRGzZs1C7dq1sXjxYowcOTLb802cOBGvvPIKJkyYoGyrVasWACAkJARt27bFkiVL8PTTTwMAlixZgmbNmqF8+fIOz3fu3DmUL18eWq0229cWRRFLly6Fr68vAODVV1/Fzp07lWRQ165d7Y5fvHgxAgICcPr0aVSvXl3ZPmLECKViacKECahWrRrOnz+PJ598EnPmzEH79u2V6Y6VK1fGvn37sGnTJrvvwfvvv69UcZUvXx6ffvopRo0ahXHjxjmMPTo6GsHBwVm+vwsXLmDOnDn44osvlG03b97MNH2yePHi0Ol0uHnzpsPzrFmzBocOHcJXX31ld54yZcrYHVemTBlYLBbcvn1bSXqVLVtWiTcsLCzLeHOKlVLk8Wz7SLFSioiIiIiI6KFNmzbBx8cHBoMBDRs2RNOmTTFnzhxcuHABZrNZabQNAFqtFk8//TT+/fdfl879999/o1WrVk73Dxw4EKtWrUJKSgrMZjNWrFiBfv36OT1ekiQIguDSa4eHhysJKQAICgpCTEyM8vjChQvo0aMHypcvj2LFiiEiIgJAWqNwWzVr1rQ7BwDlPGfOnEH9+vXtjs/4+MiRI/jkk0/g4+Oj/Bk4cCBu3LjhNAmZnJwMg8Hg9L1dv34d7dq1Q7du3ez6TgFw+P1x9n3btWsX+vbti0WLFqFatWpZnkdOeNpuNxqNAOD0feQGVkqRx7OvlGJSioiIiIiI8kdCQoLTfWq12u6xbcIkI5XKvl4kOjr6keKy1aJFC8yfPx9arRbBwcFKFdKNGzcAOE5OuJoYkpMWznTq1Al6vR7r16+HXq9HampqpgomW5UrV8bevXthNpuzrZbKuF8QBLupeZ06dUJoaCgWLVqE4OBgiKKI6tWrZ+oZaHse+X3L53H0vchYrSaKIiZMmKD00rLlLPFUqlQpnDhxwuG+69evo0WLFmjYsCEWLlxoty8wMBB//fWX3bZ79+7BbDZnqnzavXs3OnXqhBkzZqB3796ZzpOxsiomJgYajQYlS5ZUtsnTOEuXLu0w1tzASinyeLaJKDY6JyIiIiKi/OLt7e30T8aERFbHZkzuODsupzFWrFgRYWFhdgmYihUrQqfTYe/evco2s9mMI0eOoEqVKi6du2bNmti5c6fT/RqNBn369MGSJUuwZMkSvPLKK/Dy8nJ6fI8ePZCQkIB58+Y53H///n2X4rpz5w7+/fdffPTRR2jVqhWqVKmCe/fuufRcW08++SQOHjxot+3w4cN2j+vWrYszZ86gYsWKmf5kTDbK6tSpg//++y9TguvatWto3rw56tatiyVLlmR6fsOGDXHy5Em7hNK2bdug1+vx1FNPKdt27dqF5557DlOmTMHrr7+e6fUbNmyI7du3223btm0b6tWrZ/czcvLkSWi12kxVVrmJlVLk8axWJqWIiIiIiIjc4e3tjTfffBMjR45EiRIlEBISgkmTJiEpKQn9+/d36Rzjxo1Dq1atUKFCBbzyyiuwWCzYsmULRo0apRwzYMAAJcn1559/Znm+Bg0aYNSoURg+fDiuXbuGF198EcHBwTh//jwWLFiAZ599FkOGDMk2ruLFi6NkyZJYuHAhgoKCcPnyZbz//vsuvSdb7777Lpo2bYoZM2agU6dO+O2337Blyxa76qmxY8eiY8eOCA0NRbdu3aBSqXD8+HGcOHECn332mcPztmjRAomJiTh16pTS3+r69eto3rw5ypUrh+nTpyM2NlY5PjAwEADQpk0bVK1aFW+88Qa++OIL3L9/HyNGjMDAgQOVpulyQmrIkCHo2rWrksDS6XRKs/M33ngDc+fOxbBhwzBw4EDs378fixcvzrTS3h9//IEmTZpkWxH3KFgpRR7PtlJK5PQ9IiIiyiG9Xo+VK1di5MiR0Ov1BR0OEVGemzJlCrp27YpXX30V9erVw8WLF7FlyxYUL17cpec3b94cP/74I37++WfUrl0bLVu2zDS9rFKlSmjUqBGeeOIJNGjQINtzTp06FStXrsRff/2Ftm3bolq1ahg2bBhq1qypNBPPjkqlwurVq3HkyBFUr14d7733Hj7//HOXnmurcePGWLBgAWbMmIFatWph69ateO+99+yq4Nq2bYtNmzZh+/btePrpp/HMM89gxowZWTYGL1myJLp06YIVK1Yo27Zt24bz58/jt99+Q0hICIKCgpQ/MrVajY0bN0Kv16NJkybo3r07XnjhBUyfPl05ZunSpUhKSsLkyZPtzmE7vTAiIgKbN2/Grl27ULt2bXz66aeYPXt2pqmVq1atwsCBA93+vrlDkB61fX8R8ODBA/j5+SEuLs5uSUaz2YzNmzejQ4cOLnX/p4Ix9qeTSDVbceXyZVSuEI7xz9co6JAoC/xceQaOk+fgWHkGjpPn4Fh5Do6VZygq45SSkoKoqChERERk2aDak4miiAcPHqBYsWJOp53lhCRJePLJJzFo0CAMGzYs185bUAYOHIj//vsPf/zxxyOd58SJE2jdujXOnz9v17A9O3k1Thn98ssvGDlyJI4fPw6NxvEku6w+F87yLBlx+h55PPtG5wUYCBERERERESliYmKwfPlyXLt2Da+99lpBh5Mj06dPR2RkJLy9vbFlyxYsW7bMac8rd9SoUQPTpk1DdHQ0atQofIUViYmJWLJkidOEVG5hUoo8miRJsG0jJbKnFBEREeWQxWLBmjVrcOzYMbRp08ajqzqIiAqDMmXKoFSpUli4cKHLUwILm4MHD2LatGmIj49H+fLlMXv2bAwYMCBXzu3qdMSC0L1793x5HSalyKNlbGxu5WxUIiIiyqHU1FT06NEDAPDBBx/kaWNXIqLHQVHoFvTDDz8UdAhFGhudk0fLmISSJFZLEREREREREXkCJqXIo4li5m2sliIiIiIiIiIq/JiUIo9mcZCVyjilj4iIiIiIKDcUheloRLlFdFQl4ib2lCKPJn8GBMFmG39REBERERFRLtJqtRAEAbGxsShdujQE2/+AFBGiKMJkMiElJQUqFetXCqvCME6SJMFkMiE2NhYqlQo6nS7H52JSijyaPFVPq1YB6b8XWClFRERERES5Sa1WIyQkBFevXkV0dHRBh5MnJElCcnIyjEZjkUy6FRWFaZy8vLxQrly5R0qOMSlFHk1OQKmFh3NRmZQiIiIiIqLc5uPjg0qVKsFsNhd0KHnCbDZjz549aNq0KbRabUGHQ04UlnFSq9XQaDSPnBhjUoo8mjxVT60SoGKlFBERET0CnU6Hr7/+Gv/8888jTUUgoqJLrVZDrVYXdBh5Qq1Ww2KxwGAwMClViBW1ceJEUfJolvQElEoQlL5SXH2PiIiIckKr1aJ3795o1apVkbjQJyIiKuyYlCKPJoo2lVLKtoKLh4iIiIiIiIhcw6QUeTSLmHn6noVZKSIiIsoBi8WCzZs34/Dhw7BYLAUdDhERUZHHnlLk0R42On+YlGJOioiIiHIiNTUVL7zwAgBg+PDhMBqNBRsQERFREcdKKfJocqNzlc30PfaUIiIiIiIiIir8mJQij/awUgoQBMluGxEREREREREVXkxKkUdTklJqm+l7rJQiIiIiIiIiKvSYlCKP5qinFCuliIiIiIiIiAo/JqXIo8n9o9S2PaWYlCIiIiIiIiIq9JiUIo8mig+TUgKn7xERERERERF5DE1BB0D0KCzpSSmVzfQ9CyuliIiIKAd0Oh1mzZqFU6dOQafTFXQ4RERERR4rpcij2VZKqTJsIyIiInKHVqvFm2++iQ4dOkCr1RZ0OEREREUek1Lk0ex6SrHROREREREREZHHYFKKPJrF0ep77ClFREREOWC1WrF7926cOHECVqu1oMMhIiIq8thTijyaPFVPpRIgAJAAiGKBhkREREQeKiUlBZGRkQCAd955BwaDoYAjIiIiKtpYKUUezapUSqX9AVgpRUREREREROQJmJQijyba9JQS0pNSbHROREREREREVPgxKUUezSKy0TkRERERERGRJ2JSijya1bbRefo2C5NSRERERERERIUek1Lk0ZTpe2qb6XvsKUVERERERERU6DEpRR7Nmr7Snlrg9D0iIiIiIiIiT6Ip6ACIHoVVTMtKqVQPp++xUoqIiIhyQqvVYvLkyfjvv/+g1WoLOhwiIqIij0kp8mi2lVLy9D2LlUkpIiIicp9Op8Pw4cOxefNm6HS6gg6HiIioyOP0PfJocqWUWiVALU/fY6UUERERERERUaHHpBR5NGX1PZVNo3P2lCIiIqIcsFqtOHz4MM6dOwer1VrQ4RARERV5nL5HHk2eqacSHmZYWSlFREREOZGSkoJGjRoBAAYMGACDwVDAERERERVtBVoptWfPHnTq1AnBwcEQBAEbNmyw2y9JEsaPH4/g4GAYjUY0b94cp06dsjsmNTUV7777LkqVKgVvb2907twZV69ezcd3QQVJtKmUUgmS3TYiIiIiIiIiKrwKNCmVmJiIWrVqYe7cuQ73T5s2DTNmzMDcuXNx6NAhBAYGIjIyEvHx8coxQ4cOxfr167F69Wrs3bsXCQkJ6NixI0uuHxNyVVRaUip9G5NSRERERERERIVegU7fa9++Pdq3b+9wnyRJmDlzJj788EN06dIFALBs2TKUKVMGK1euxKBBgxAXF4fFixdj+fLlaN26NQDgu+++Q2hoKHbs2IG2bdvm23uhgqH0lBIECPI25qSIiIiIiIiICr1C21MqKioKN2/eRJs2bZRter0ezZo1w759+zBo0CAcOXIEZrPZ7pjg4GBUr14d+/btc5qUSk1NRWpqqvL4wYMHAACz2Qyz2axsl7+23UaFi8lsgSiKkEQrVAIgiiLMFgvHrBDj58ozcJw8B8fKM3CcPEPG60COV+HGz5Vn4Dh5Do6VZ/CUcXI1vkKblLp58yYAoEyZMnbby5Qpg0uXLinH6HQ6FC9ePNMx8vMdmTx5MiZMmJBp+7Zt2+Dl5ZVp+/bt292On/LHmSsqPDABByzRUAnAtatXkXJbwua40wUdGmWDnyvPwHHyHBwrz8BxKtxSUlKUr3/77Tc2OvcQ/Fx5Bo6T5+BYeYbCPk5JSUkuHVdok1IyQRDsHkuSlGlbRtkdM2bMGAwbNkx5/ODBA4SGhqJNmzYoVqyYst1sNmP79u2IjIyEVqvN4TugvHRmx3ncSTShyTMh+H3PXpQNCUFoCS90aFa+oEMjJ/i58gwcJ8/BsfIMHCfPkJiYqHzdsmVL+Pv7F1wwlC1+rjwDx8lzcKw8g6eMkzwjLTuFNikVGBgIIK0aKigoSNkeExOjVE8FBgbCZDLh3r17dtVSMTExynK+juj1euj1+kzbtVqtw0F1tp0KniQIUKlU0Ou0UAmASqUCBBXHywPwc+UZOE6eg2PlGThOhZuXlxc++ugjnDt3Dl5eXhwrD8HPlWfgOHkOjpVnKOzj5GpsBbr6XlYiIiIQGBhoV5JmMpmwe/duJeH01FNPQavV2h1z48YNnDx5MsukFBUdovhw9T2Bq+8RERHRI9DpdBg7diz+7//+DzqdrqDDISIiKvIKtFIqISEB58+fVx5HRUXh77//RokSJVCuXDkMHToUkyZNQqVKlVCpUiVMmjQJXl5e6NGjBwDAz88P/fv3x/Dhw1GyZEmUKFECI0aMQI0aNZTV+Khos9gkpeQMqygxKUVERERERERU2BVoUurw4cNo0aKF8lju89SnTx8sXboUo0aNQnJyMt566y3cu3cPDRo0wLZt2+Dr66s858svv4RGo0H37t2RnJyMVq1aYenSpVCr1fn+fij/yVVRagFQsVKKiIiIHoEoijh16hQuX74MURQLOhwiIqIir0CTUs2bN4eURVWLIAgYP348xo8f7/QYg8GAOXPmYM6cOXkQIRV2clWUSiU8TEqxUoqIiIhyIDk5GXXq1AEAvPrqqw57kBIREVHuKbQ9pYhc8bBS6mFSSmSlFBEREREREVGhx6QUeSxJkiDnn9QqAek5KaXPFBEREREREREVXkxKkcey7R2lVrFSioiIiIiIiMiTMClFHsviJCnFnlJEREREREREhR+TUuSxRJvkk1oQlB9mq4gsG+gTERERERERUcFjUoo8ljx9TxAe/pFxBh8RERERERFR4aYp6ACIckoU0/5WCwIEQYDaJillFSWoVYLjJxIRERE5oNVqMWzYMFy8eBFarbagwyEiIirymJQijyX3jpKTT/aVUiyVIiIiIvfodDpMmTIFmzdvhk6nK+hwiIiIijxO3yOPZUkvlVKlZ6Nsf5itnL9HREREREREVKgxKUUeS5m+l/5TbNtXysKkFBEREblJFEVER0fj1q1bEOULDSIiIsoznL5HHkuulFKrHuZWNSoBVgkQmZQiIiIiNyUnJ6Ny5coAgO7du0Ov1xdwREREREUbK6XIY2WslAIeTuWzsqcUERERERERUaHGpBR5LKXRuU2Hc7npOSuliIiIiIiIiAo3JqXIY8nNzFWqh0kp+UtWShEREREREREVbkxKkceSk1IaVeZKKa6+R0RERERERFS4MSlFHstRpZQ8lY8L5hAREREREREVbkxKkccSpcyVUnKCitP3iIiIiIiIiAo3TUEHQJRTSqWUkLlSyspSKSIiInKTRqPBG2+8gUuXLkGj4WUyERFRXuNvW/JYlvSklNphT6kCCYmIiIg8mF6vx+zZs7F582bo9fqCDoeIiKjI4/Q98ljy9D3HSSlO3yMiIiIiIiIqzJiUIo9ldVAppUr/iRbZU4qIiIjcJEkSYmNjERcXB4nXEkRERHmO0/fIYylJKYc9pXghSURERO5JSkpC2bJlAQCdO3eGTqcr4IiIiIiKNlZKkcdSGp07mr7Hu5tEREREREREhRqTUuSx5MSTXaVUelJKZKUUERERERERUaHGpBR5LNFRTylO3yMiIiIiIiLyCExKkceyOEhKcfU9IiIiIiIiIs/ApBR5rCwrpdhTioiIiIiIiKhQY1KKPJbSU8quUip9HyuliIiIiIiIiAo1TUEHQJRTcuLJrtF5+tciK6WIiIjITRqNBq+++iquXr0KjYaXyURERHmNv23JY1kdTd9L/9piZVKKiIiI3KPX67F48WJs3rwZer2+oMMhIiIq8jh9jzyWnJRS2SSlNCpWShERERERERF5AialyGPJSSmNo0bnYoGERERERB5MkiQkJiYiJSUFEm9wERER5TlO3yOPJTc6V9n2lFJx9T0iIiLKmaSkJBQvXhwAcO/ePeh0ugKOiIiIqGhjpRR5LDGLnlIiV98jIiIiIiIiKtSYlCKP9bDR+cNtasF+HxEREREREREVTkxKkceSF9iznb6n4vQ9IiIiIiIiIo/ApBR5LKuY1s1co3r4Y6wWOH2PiIiIiIiIyBMwKUUeS15hT2U7fS+9UsrCpBQRERERERFRocakFHksMX2Knl2llNzonNP3iIiIiIiIiAo1TUEHQJRTjhudC3b7iIiIiFylVqvRpUsX3Lx5E2q1uqDDISIiKvKYlCKPJU/Rc9jonEkpIiIicpPBYMDq1auxefNmGAyGgg6HiIioyOP0PfJYolIp9TAppTQ65/Q9IiIiIiIiokKNSSnyWBZHSSmlUqpAQiIiIiIiIiIiF3H6HnksuRrKPillv4+IiIjIVYmJifDx8QEA3Lt3D/7+/gUbEBERURHHSinyWEqjc9ueUmx0TkREREREROQRmJQijyUnnlQOp+8xKUVERERERERUmDEpRR5LTjxpVFx9j4iIiIiIiMjTMClFHssqOaiUkqfvsacUERERERERUaHGpBR5JFGUIOedNA6m74mslCIiIiIiIiIq1JiUIo9kWwmlsmt0nr6fSSkiIiIiIiKiQk3jzsGSJGH37t34448/EB0djaSkJJQuXRp16tRB69atERoamldxEtmxTTqpVQIgpn0tV01ZmJQiIiIiN6nVarRv3x4xMTFQq9UFHQ4REVGR51KlVHJyMiZNmoTQ0FC0b98ev/zyC+7fvw+1Wo3z589j3LhxiIiIQIcOHXDgwIG8jpkIok2llFrI3OhcZE8pIiIicpPBYMBPP/2Ejz/+GAaDoaDDISIiKvJcqpSqXLkyGjRogAULFqBt27bQarWZjrl06RJWrlyJl19+GR999BEGDhyY68ESyeRKKUFIS0RZrWnblUbnrJQiIiIiIiIiKtRcSkpt2bIF1atXz/KYsLAwjBkzBsOHD8elS5dyJTgiZ+Skk22VFGBbKZU23VTIsJ+IiIiIiIiICgeXklLZJaRs6XQ6VKpUKccBEblCSUqp7JNOtkkqqyhBo2ZSioiIiFyTmJiIgIAAWK1W3Lx5E/7+/gUdEhERUZHm9up7W7duxd69e5XH//vf/1C7dm306NED9+7dy9XgiJxxlpRS2Ty2sq8UERERuSkpKQmpqakFHQYREdFjwe2k1MiRI/HgwQMAwIkTJzB8+HB06NABFy9exLBhw3I9QCJH5IRT5kqph1+LYn5GRERERERERETucGn6nq2oqChUrVoVALB27Vp07NgRkyZNwtGjR9GhQ4dcD5DIEblSSpWhZ5SalVJEREREREREHsHtSimdToekpCQAwI4dO9CmTRsAQIkSJZQKKqK8JldBqTP8BAuCADkvxRX4iIiIiIiIiAovtyulnn32WQwbNgyNGzfGwYMH8f333wMAzp49i5CQkFwPkMgRS3pWSq3KnFdVqwSIVgkik1JEREREREREhZbblVJz586FRqPBmjVrMH/+fJQtWxYAsGXLFrRr1y7XAyRyRJR7SgmZV9eTp/Rx+h4RERERERFR4eV2pVS5cuWwadOmTNu//PLLXAmIyBXW9Ol7GnXmpJTcV4qVUkREROQOlUqFpk2b4s6dO1A5qMYmIiKi3OX2b9vmzZvj22+/RXJycl7EQ+QSZ43OgYdJKQuTUkREROQGo9GIHTt2YOLEiTAajQUdDhERUZHndlLqqaeewqhRoxAYGIiBAwfiwIEDeREXUZbkpFTGRudp2wS7Y4iIiIiIiIio8HE7KfXFF1/g2rVr+PbbbxEbG4umTZuiatWqmD59Om7dupUXMRJlIveLclgplb5NZE8pIiIiIiIiokIrR5Pl1Wo1nn/+eWzYsAHXrl1Djx498PHHHyM0NBQvvPACfvvtt1wJzmKx4KOPPkJERASMRiPKly+PTz75BGL6ymsAIEkSxo8fj+DgYBiNRjRv3hynTp3KldenwuthpZSDRueslCIiIqIcSExMRHBwMHr37o3ExMSCDoeIiKjIe6QOjgcPHsTYsWMxffp0BAQEYMyYMQgICECnTp0wYsSIRw5u6tSpWLBgAebOnYt///0X06ZNw+eff445c+Yox0ybNg0zZszA3LlzcejQIQQGBiIyMhLx8fGP/PpUeMkJJ42DpBQrpYiIiCinbt++jQcPHhR0GERERI8Ft1ffi4mJwfLly7FkyRKcO3cOnTp1wurVq9G2bVsI6cmA7t2744UXXsD06dMfKbj9+/fj+eefx3PPPQcACA8Px6pVq3D48GEAaVVSM2fOxIcffoguXboAAJYtW4YyZcpg5cqVGDRo0CO9PhVeSqNzB0kpeUU+q5hpFxEREREREREVEm4npUJCQlChQgX069cPffv2RenSpTMdU79+fTz99NOPHNyzzz6LBQsW4OzZs6hcuTL++ecf7N27FzNnzgQAREVF4ebNm2jTpo3yHL1ej2bNmmHfvn1Ok1KpqalITU1VHst3w8xmM8xms7Jd/tp2GxUOJrM5bRqnKNqNm9lshiSKEEURqSYzx64Q4ufKM3CcPAfHyjNwnDxDxutAjlfhxs+VZ+A4eQ6OlWfwlHFyNT5Bktyb4/THH3+gSZMmOQrKXZIk4YMPPsDUqVOhVqthtVoxceJEjBkzBgCwb98+NG7cGNeuXUNwcLDyvNdffx2XLl3Cr7/+6vC848ePx4QJEzJtX7lyJby8vPLmzVCu+u++gH/uCAjzlfBMgP2P8G/XBcQmC2hURkSoTwEFSERERB4nJSUFr7zyCgBg9erVMBgMBRwRERGRZ0pKSkKPHj0QFxeHYsWKOT3O7Uqp/EpIAcD333+P7777DitXrkS1atXw999/Y+jQoQgODkafPn2U44QMK7BJkpRpm60xY8Zg2LBhyuMHDx4gNDQUbdq0sftmmc1mbN++HZGRkdBqtbn4zuhReZ2Nxd1/Y1E3zB8dagfbjdWtQ9dxITYRzzxVFrVC/Ao6VMqAnyvPwHHyHBwrz8Bx8gy2zc1btmwJf3//gguGssXPlWfgOHkOjpVn8JRxcrU/o9tJKQBYs2YNfvjhB1y+fBkmk8lu39GjR3NySodGjhyJ999/X7ljVaNGDVy6dAmTJ09Gnz59EBgYCAC4efMmgoKClOfFxMSgTJkyTs+r1+uh1+szbddqtQ4H1dl2KjiCoIZKpYJOq7EbG61WC60mbZ+gUnPcCjF+rjwDx8lzcKw8A8epcMt0TcGx8ggcK8/AcfIcHCvPUNjHydXY3F59b/bs2XjttdcQEBCAY8eOoX79+ihZsiQuXryI9u3bux1oVpKSkqBS2YeoVqvTegkBiIiIQGBgILZv367sN5lM2L17Nxo1apSrsVDhYk2fdap2tPqeiqvvERERkftUKhWeeuopVKxYMdM1KBEREeU+tyul5s2bh4ULF+L//u//sGzZMowaNQrly5fH2LFjcffu3VwNrlOnTpg4cSLKlSuHatWq4dixY5gxYwb69esHIG3a3tChQzFp0iRUqlQJlSpVwqRJk+Dl5YUePXrkaixUuIjpq++pHUzTVAny6ntMShEREZHrjEYj9u/fj82bN8NoNBZ0OEREREWe20mpy5cvK1VIRqMR8fHxAIBXX30VzzzzDObOnZtrwc2ZMwcff/wx3nrrLcTExCA4OBiDBg3C2LFjlWNGjRqF5ORkvPXWW7h37x4aNGiAbdu2wdfXN9fioMLHkp5wUmVRKcWkFBEREREREVHh5XZSKjAwEHfu3EFYWBjCwsJw4MAB1KpVC1FRUXBzIb9s+fr6YubMmZg5c6bTYwRBwPjx4zF+/PhcfW0q3OSpeY4qpZiUIiIiIiIiIir83E5KtWzZEhs3bkTdunXRv39/vPfee1izZg0OHz6MLl265EWMRJlYrOlJKbWDpJQ8fY89pYiIiMgNSUlJqFq1KpKSknDu3Dn4+bm/iq8oSoi+k4j4FAt8DRqEl/R2WNlNREREOUhKLVy4UGk0/sYbb6BEiRLYu3cvOnXqhDfeeCPXAyRyxOpCpZTISikiIiJygyRJuHTpkvK1u05ei8Pao1dxPiYBqWYReq0KFQN80LVuCKqXdT/BRUREVNS5nZRSqVR2q5F0794d3bt3z9WgiLKjNDp3cOdRxel7RERElM9OXovD7J3ncDfRhCA/I4x+aiSbrDhxNQ7X7iVjcKtKTEwRERFl4HZS6ty5c/jpp58QHR0NQRBQvnx5vPDCC4iIiMiL+IgckiulVI4qpdK3iZy+R0RERPlAFCWsPXoVdxNNqBjgA0EAUi0ifAxqVNT74HxMAtYdvYaqQcU4lY+IiMiGW0mpyZMnY+zYsRBFEQEBAZAkCbGxsRg9ejQmTZqEESNG5FWcRHayqpRSpxfyWcX8jIiIiIgeV9F3EnE+JgFBfkYIgoCYBymIvpOEiFJeKO1rQJCfEedi4hF9JxHlS/sUdLhERESFhir7Q9L8/vvv+Oijj/Dhhx/i9u3buHHjBm7evInY2Fi8//77eP/997Fnz568jJVIYclq+h4bnRMREVE+ik+xINUswqhTAwCSzVYAQIo57Q6ZUadGqllEfIqlwGIkIiIqjFyulFqwYAEGDBiA8ePH220vUaIEPvnkE9y8eRPz589H06ZNcztGokysWVZKyT2lWCpFREREec/XoIFeq0KyyQofg0ZpISD/nWyyQq9VwdfgducMIiKiIs3lSqmDBw/i1Vdfdbr/1VdfxYEDB3IlKKLsyBd5miyTUvkaEhEREXk4QRBQpUoVhIaGQnDQt9KZ8JLeqBjggxtxyZAkCfJ9MVGSIEkSbsQlo1KAL8JLeudR5ERERJ7J5ds1t27dQnh4uNP9ERERuHnzZm7ERJQtOeHksNF5elJK5Op7RERE5AYvLy/8888/2Lx5M7y8vFx+nkoloGvdEFy7l4zzMQlItVghShKSTVacj0lACW8dutQtyybnREREGbhcKZWSkgKdTud0v1arhclkypWgiLIjT81zOH2PPaWIiIgon1Uv64fBrSqhRogfElOtSEixIDHVipoh/hjcqhKql/Ur6BCJiIgKHbcmtn/99dfw8XG8Ykh8fHyuBETkCrlSSu0grapSpu8xKUVERET5p3pZP1QNKgad+j9E3U5E5TI+GBb5BCukiIiInHA5KVWuXDksWrQo22OI8sPDSqnMWSm5z5TISikiIiJyQ1JSEurVq4eEhAQ0b94cfn7uVzepVAKKGbUo6aOHv5eOCSkiIqIsuJyUio6OzsMwiNwjT81TO+gpxUopIiIiyglJkvDvv/8qX+eUxZr2XAuvRYiIiLLkck8posJEaXTu4CdY6SnFC0EiIiIqAJb0im4LlwImIiLKkktJqdWrV7t8witXruDPP//McUBErpCn5jlsdM5KKSIiIipA5vRKKflvIiIicsylpNT8+fPx5JNPYurUqUpJs624uDhs3rwZPXr0wFNPPYW7d+/meqBEtuSyeEdJKRVX3yMiIqICZE6vkJIrpoiIiMgxl3pK7d69G5s2bcKcOXPwwQcfwNvbG2XKlIHBYMC9e/dw8+ZNlC5dGq+99hpOnjyJgICAvI6bHnNiFj2l5ESVyEopIiIiKgAWJSnFaxEiIqKsuNzovGPHjujYsSPu3LmDvXv3Ijo6GsnJyShVqhTq1KmDOnXqQOWowQ9RHpCn5jmevicfk58REREREaU1SDenX6dYOH2PiIgoSy4npWQlS5bE888/nxexELnMmkVPKWX6HkvmiYiIyA2CICAsLAxJSUkQHFRju8IqSpA7CLDRORERUdbcTkoRFTTR5mLPUVJKk16xx55SRERE5A4vLy+cO3cOmzdvhpeXV47OYTtlzyxKkCQpxwkuIiKioo7z7cjj2CabVA4u8lScvkdEREQFxGxzASJJANtKEREROcekFHkcq83VneOeUumNzlkpRURERPksYx8pM++SEREROcXpe+Rx7JJSjlbfU3pKMSlFRERErktOTkaTJk0QFxeHFi1aQKvVun2OjEkorsBHRETknNuVUrt27cqDMIhcJ0/fEwRA5ajRuYpJKSIiInKfKIo4cuQIzp8/DzGHC6aYM1x/WLkCHxERkVNuJ6XatWuHChUq4LPPPsOVK1fyIiaiLMkXdxoHCSngYaUUp+8RERFRfsu44p6ZqwETERE55XZS6vr16xgyZAjWrVuHiIgItG3bFj/88ANMJlNexEeUiVwp5ajJOcBKKSIiIio45gyVURl7TBEREdFDbielSpQogcGDB+Po0aM4fPgwnnjiCbz99tsICgrC4MGD8c8//+RFnEQKMT3Z5KjJue12UQIkVksRERFRPsrYU4qNzomIiJx7pNX3ateujffffx9vv/02EhMT8c033+Cpp55CkyZNcOrUqdyKkciOXCnlLCllO62P1VJERESUnzJWRrHRORERkXM5SkqZzWasWbMGHTp0QFhYGH799VfMnTsXt27dQlRUFEJDQ9GtW7fcjpUIwMOLPWdJKdtpfVZWShEREVE+ythDysqeUkRERE5p3H3Cu+++i1WrVgEAevXqhWnTpqF69erKfm9vb0yZMgXh4eG5FiSRLbmBudpJTynbZBWvA4mIiMgdpUqVeqReqRkrpTL2mCIiIqKH3E5KnT59GnPmzEHXrl2h0+kcHhMcHIzff//9kYMjckSekqdyWin18GuLKAJQ50NURERE5Om8vb1x/fp1bN68Gd7e3jk6B3tKERERuc7tpNTOnTuzP6lGg2bNmuUoIKLsyJVSGidJKUEQoFYBVpGVUkRERJS/MvaQYk8pIiIi59xOSslOnz6Ny5cvZypv7ty58yMHRZQVSzar7wFpU/uskNhTioiIiPKV2WJ/RyzjdD4iIiJ6yO2k1MWLF/Hiiy/ixIkTEAQBUvp/+oX0/j5WqzV3IyTKQJm+56SnFJA+tc8qcfU9IiIicllycjLatWuHO3fuoEWLFtBqtW6fwyJmTEqxbJuIiMgZt1ffGzJkCCIiInDr1i14eXnh1KlT2LNnD+rVq4ddu3blQYhE9uRrPXUWP71yE3SRlVJERETkIlEUsWfPHpw6dQpiDnsAZGxsbuYNMiIiIqfcrpTav38/fvvtN5QuXRoqlQoqlQrPPvssJk+ejMGDB+PYsWN5ESeRQr4DqVY5z0qp1emVe7wQJCIionzESikiIiLXuV0pZbVa4ePjAyBtydzr168DAMLCwnDmzJncjY7IAbn6Se189p5SKcWkFBEREeUns4WNzomIiFzldqVU9erVcfz4cZQvXx4NGjTAtGnToNPpsHDhQpQvXz4vYiSyY1Wm72XR6FzF6XtERESU/8zplVJ6jQqpFpGNzomIiLLgdlLqo48+QmJiIgDgs88+Q8eOHdGkSROULFkS33//fa4HSJSR0ug8i6SU3ASddyeJiIgoP8lJKINWnZaUymFvKiIioseB20mptm3bKl+XL18ep0+fxt27d1G8eHFlBT6ivCQnpTSuVEoxKUVERET5yJxe0u2lUyMu2Zyp8TkRERE95HZSypESJUrkxmmIXGJNn5KnyiIJKielrJy+R0RERG7w8vKC1WrN8fPNSqVUWutWNjonIiJyzqWkVJcuXVw+4bp163IcDJEr5OonTRadzlVsdE5ERERu8vb2xv3797F582Z4e3vn6BxyEsqoVac95rUIERGRUy4lpfz8/PI6DiKXKT2lsqyUsj+WiIiIKD+Y0689jLq0y2xWShERETnnUlJqyZIleR0HkcvkO45Zr76XlpViUoqIiIjyEyuliIiIXJejnlIWiwW7du3ChQsX0KNHD/j6+uL69esoVqwYfHx8cjtGIjtiep8odVaVUoL9sURERETZSUlJQZcuXRATE4OWLVtCq9W6fQ650blRp0p/zGsRIiIiZ9xOSl26dAnt2rXD5cuXkZqaisjISPj6+mLatGlISUnBggUL8iJOIoV8x1Hlwup7rJgnIiIiV1mtVmzZskX5OiceNjpPr5TixQgREZFTKnefMGTIENSrVw/37t2D0WhUtr/44ovYuXNnrgZH5IjS6DyLpJRKxUbnRERElP8sIqfvERERucrtSqm9e/fizz//hE6ns9seFhaGa9eu5VpgRM5YXamUSp/ax+l7RERElF9EUVKqtI06OSnFSikiIiJn3K6UEkXRYTnz1atX4evrmytBEWXF6kJPKVZKERERUX4z2ySglEop9pQiIiJyyu2kVGRkJGbOnKk8FgQBCQkJGDduHDp06JCbsRE5JLqy+l56wsrKSikiIiLKJ7YJKDkpxUbnREREzrk9fe/LL79EixYtULVqVaSkpKBHjx44d+4cSpUqhVWrVuVFjER2LK4kpeRKKV4IEhERUT6Rk1JqFaBVp9375fQ9IiIi59xOSgUHB+Pvv//GqlWrcPToUYiiiP79+6Nnz552jc+J8orcJ8qlpBQrpYiIiCifyNP3NCoVNOq0axGzVYIkSRCyaDtARET0uHI7KQUARqMR/fr1Q79+/XI7HqJsWd2olBLZU4qIiIhc5O3tDZPJhM2bN8Pb29vt55vTu5zrNCqlUgpIu3aRk1RERET0kNtJqd9++w3r1q1DdHQ0BEFA+fLl0bVrVzRt2jQv4iPKRElKZdXonD2liIiIKJ/J0/c0KgEam5tnFlGCRl1QURERERVebjU6f+ONN9C6dWusWrUKd+7cQWxsLL777ju0aNEC7777bl7FSGTHnUoprr5HRERE+UWulNKoVXbXKfJ2IiIisudypdT69euxZMkSfPPNN+jTp48yL14URSxduhRvvvkmIiMj0blz5zwLlgh4WP2kyqJSSr47KbJSioiIiFyUkpKCnj174ubNm2jZsiW0Wq1bz5cXY9GqBAiCAK1agNkq2a3KR0RERA+5nJRasmQJhg0bhr59+9ptV6lU6NevH86cOYPFixczKUV5Tl5RL6veDCqlUipfQiIiIqIiwGq1Yt26dcrX7jJZ0i48tJq0yQgalQpmq1VJVhEREZE9l6fvHT16FC+++KLT/V27dsWRI0dyJSiirLhSKaUW2OiciIiI8pecfJIrtuUbaBaRd8mIiIgccTkpdfv2bZQtW9bp/rJly+LOnTu5EhRRVkQXekqp0n+yeWeSiIiI8oslvURbXnlPTk5x+h4REZFjLielTCYTdDqd0/0ajQYmkylXgiLKilwpldXqe2quvkdERET5zJyhxYAmPTnFRudERESOudxTCgA+/vhjeHl5OdyXlJSUKwERZUeuflJlkVKVq6g4fY+IiIjyizlDpZRWrpTi9QgREZFDLielmjZtijNnzmR7DFFeE5V+Dc6zUg8bnfMikIiIiPKHfN2hzVApxel7REREjrmclNq1a1cehkHkOrkCPstKKYFJKSIiIspfpvSLFPnGmdJTio3OiYiIHHJr+h5RYSBK2VdKydP32FOKiIiIXOXl5YV79+7h119/ddqyIityRdTDSqm0v82slCIiInLI5UbnRIWFXP2UZaNzTt8jIiIiNwmCAG9vbxgMBghZXGc4Y85YKaVM32OlFBERkSOFPil17do19OrVCyVLloSXlxdq166NI0eOKPslScL48eMRHBwMo9GI5s2b49SpUwUYMeUlSZLY6JyIiIgKJfkaRatho3MiIiJXFOqk1L1799C4cWNotVps2bIFp0+fxhdffAF/f3/lmGnTpmHGjBmYO3cuDh06hMDAQERGRiI+Pr7gAqc8YzsbT048OaISOH2PiIiI3JOamor+/ftj1qxZSE1Ndfv5yup7qgyNzpmUIiIicsjtnlKXL19GaGhoppJmSZJw5coVlCtXLteCmzp1KkJDQ7FkyRJlW3h4uN1rzpw5Ex9++CG6dOkCAFi2bBnKlCmDlStXYtCgQbkWCxUOthd1WSWlWClFRERE7rJYLFi+fLnytdvPl6fvpSej5N5SnL5HRETkmNtJqYiICNy4cQMBAQF22+/evYuIiAhYrdZcC+7nn39G27Zt0a1bN+zevRtly5bFW2+9hYEDBwIAoqKicPPmTbRp00Z5jl6vR7NmzbBv3z6nSanU1FS7u18PHjwAAJjNZpjNZmW7/LXtNipYqWYrxPQVbESLBWYp7euMYyVaLRBFEWaLleNXyPBz5Rk4Tp6DY+UZOE6eIeN1oLvjlWJOu/6AlHb9IUgiRFFEisn9c1H2+LnyDBwnz8Gx8gyeMk6uxidIknvzm1QqFW7duoXSpUvbbb906RKqVq2KxMREd06XJYPBAAAYNmwYunXrhoMHD2Lo0KH46quv0Lt3b+zbtw+NGzfGtWvXEBwcrDzv9ddfx6VLl/Drr786PO/48eMxYcKETNtXrlyZo5VWKP+kWIGfotPuPnYvL8JZD9K7qcD2qyoYNUDnMN6dJCIiouylpKTglVdeAQCsXr1auRZ11W/XBcQmC2hURkSoD3D8roB/7wmo7CehTilWbxMR0eMjKSkJPXr0QFxcHIoVK+b0OJcrpYYNGwYgbVWSjz/+2C55Y7Va8ddff6F27do5j9gBURRRr149TJo0CQBQp04dnDp1CvPnz0fv3r2V4xxNJcxqxZQxY8Yo7wdIq5QKDQ1FmzZt7L5ZZrMZ27dvR2RkJLRabW69LXoED5LNOLrtHFQC8NxzVZXtGcfqRlwK/tt1ET56NTq0e6IAI6aM+LnyDBwnz8Gx8gwcJ89ge3O1ZcuWdn1MXXFp90UY7qegeYNQPBHoC8OZWCT8F4vq4cXRoVZQLkdL/Fx5Bo6T5+BYeQZPGSd5Rlp2XE5KHTt2DEBawufEiRPQ6XTKPp1Oh1q1amHEiBFuhpm1oKAgVK1a1W5blSpVsHbtWgBAYGAgAODmzZsICnr4iz4mJgZlypRxel69Xg+9Xp9pu1ardTiozrZT/lOZJKhUKmjVQpZjZdBboVKpIEHFsSuk+LnyDBwnz8Gx8gwcp8LNdmxyMlYiBKhUKhj0urTrEZ02/XrE8XUL5Q5+rjwDx8lzcKw8Q2EfJ1djczkp9fvvvwMA+vbtizlz5sDX1zdnkbmhcePGOHPmjN22s2fPIiwsDEBaf6vAwEBs374dderUAQCYTCbs3r0bU6dOzfP4KP/Jq+mpsqiEAwB1+n6Rq+8RERFRPrFY06475Abn8ip8XH2PiIjIMZU7B1ssFnz33Xe4dOlSXsVj57333sOBAwcwadIknD9/HitXrsTChQvx9ttvA0ibtjd06FBMmjQJ69evx8mTJ9G3b194eXmhR48e+RIj5S9r+kVdVivv2e638iKQiIiI8ok5fTEWbfrqe/IqfFx9j4iIyDG3Vt/TaDQICwvL1RX2svL0009j/fr1GDNmDD755BNERERg5syZ6Nmzp3LMqFGjkJycjLfeegv37t1DgwYNsG3btnyp5KL8JyeZNNkkpVRyUoqVUkREROQiLy8vXLt2DTt27MjR4jdmi/11ikbNSikiIqKsuJWUAoCPPvoIY8aMwXfffYcSJUrkRUx2OnbsiI4dOzrdLwgCxo8fj/Hjx+d5LFTw5KSUKpuklHwxKEmAKErZHk9EREQkCAJKly4NPz+/LBfNccaSsVJKnr5nZVKKiIjIEbeTUrNnz8b58+cRHByMsLAweHt72+0/evRorgVHlJHcIyrbSimbC0mrJEEFJqWIiIgo70iSBHN68kmukNKo0pJT8rQ+IiIisud2UuqFF17IgzCIXCOXv2fb6NwmaWUVJWjVeRoWERERFQGpqakYOnQoLl26hFatWrm1qpHtFD25UkpueM5KKSIiIsfcTkqNGzcuL+IgconoaqNzm6QVV+AjIiIiV1gsFixYsED52q3n2iSeHvaUYqNzIiKirLidlJIdOXIE//77LwRBQNWqVVGnTp3cjIvIIblxeXZJKUFI+yNJXIGPiIiI8p48RU8QHl6nKD2leC1CRETkkNtJqZiYGLzyyivYtWsX/P39IUkS4uLi0KJFC6xevRqlS5fOiziJADy8C5l9UkqAWhBgkSSwjQMRERHlNfkaRadWKU3SufoeERFR1lTuPuHdd9/FgwcPcOrUKdy9exf37t3DyZMn8eDBAwwePDgvYiRSyFPx1C6siCMnrqwuTN8TRQkXYxPwz5X7uBiboEwTJCIiInKFOX2Knu2NM6XROafvEREROeR2pdTWrVuxY8cOVKlSRdlWtWpV/O9//0ObNm1yNTiijOSpeKpsKqWAh83Qs5u+d/JaHNYevYrzMQlINYvQa1WoGOCDrnVDUL2s36MHTUREREWenHiSm5ynfc1G50RERFlxOykliqLDlUi0Wi1EzpOiPPawUir7Y+VrwqySUievxWH2znO4m2hCkJ8RRj81kk1WnLgah2v3kjG4VSUmpoiIiChbcuJJa3ORojQ6FyVIkqRM6yMiIqI0bk/fa9myJYYMGYLr168r265du4b33nsPrVq1ytXgiDJSekqps//RVWUzfU8UJaw9ehV3E02oGOADH4MGapUAH4MGFQN8cDfRhHVHr3EqHxEREWXLkn5zVp6yl/a1YLOf1xNEREQZuZ2Umjt3LuLj4xEeHo4KFSqgYsWKiIiIQHx8PObMmZMXMRIprO70lEo/xllSKfpOIs7HJCDIzwhBEHDtXhJOXL0Pi1WEIAgI8jPiXEw8ou8k5t4bICIiokLLaDTi7Nmz+Oqrr2A0Gt16rjn9xpnGtlLKNinFKXxERESZuD19LzQ0FEePHsX27dvx33//QZIkVK1aFa1bt86L+IjsyDNENS70lFIanTtJSsWnWJBqFmH0UwMA7iaakGwWkZhqgZ+XDkadGrceiIhPseRO8ERERFSoqVQqhIeHo0yZMlCp3Lt3K/eU0tlUc6tVAgQBkCS5kkqdm+ESERF5PLeTUrLIyEhERkbmZixE2ZIrpdxpdO6sXN7XoIFeq0KyyQofgwbyDUz5NZJNVui1KvgacvwxISIioseEo0opQRCgVQkwWSVWShERETng9vQ9ANi5cyc6duyoTN/r2LEjduzYkduxEWViFeXllrM/Vq6mEp30lAov6Y2KAT64EZcMSZKU46zpzUhvxCWjUoAvwkt6507wREREVKiZTCa8//77WLp0KUwmk1vPtaRXSmkyXKTIj81cEIiIiCiTHPWUateuHXx9fTFkyBAMHjwYxYoVQ4cOHTB37ty8iJFIkX69p1RBZUWVzfQ9lUpA17ohKOGtw/mYBKSarRAlCQkpVpyPSUAJbx261C3rUlUWEREReT6z2YwZM2Zgw4YNMJvNbj1XrszWZrhukCunWClFRESUmdvzkiZPnowvv/wS77zzjrJt8ODBaNy4MSZOnGi3nSi3PayUevSeUgBQvawfBreqhLVHrmLLyZuwihKMOjOeiSiJLnXLonpZv9wJnIiIiIo0U/qdM23GSikVk1JERETOuJ2UevDgAdq1a5dpe5s2bTB69OhcCYrIGas7jc6FrKfvyaqX9UP5Ut6IiU+F2SqiUYWS6N0wnBVSRERE5DKLg55SAKBRcfoeERGRM25P3+vcuTPWr1+faftPP/2ETp065UpQRM4ojc5zYfqeLbMooZhRi5I+evh76ZiQIiIiyoIoSrgYm4B/rtzHxdgEiC78ri3qLE4qpbScvkdEROSU25VSVapUwcSJE7Fr1y40bNgQAHDgwAH8+eefGD58OGbPnq0cO3jw4NyLlAhQLnpdmr6Xfkh2lVLAw2WcASDVYs1ZcERERI+Bk9fisPbo1fR+jCL0WhUqBviga92Qx3rauzn9GiVjNbfS6NzKSikiIqKM3E5KLV68GMWLF8fp06dx+vRpZbu/vz8WL16sPBYEgUkpynVy1ZMrlUwPe0plf16TxTYpxYtGIiIiR05ei8PsnedwN9GEID8jjH5qJJusOHE1DtfuJWNwq0qPbWLKbMm6p5QrldtERESPG7eTUlFRUXkRB5FLrE7uQjoiJ64sLvRwsKuUMjMpRURElJEoSlh79CruJppQMcAHQvpUeh+DBhX1Pjgfk4B1R6+halCxx3IavHyN4rTROXtKERERZeJ2TymZyWTCmTNnYLFYcjMeoizJPaXULvSUki8CXbkGtK2UMrG8noiIKJPoO4k4H5OAID8jBEHA7YRUnLn5ABZRhCAICPIz4lxMPKLvJBZ0qDlmNBpx7NgxzJ49G0aj0a3nytcPmRqdK9P3WClFRESUkdtJqaSkJPTv3x9eXl6oVq0aLl++DCCtf9SUKVNyPUAiW1Y3ekrJzdCtLvSUsp2yl2pmTykiIqKM4lMsSDWLMOrUAIAb95MRl2xBfHLaDUqjTo1Us4j4FM+9YalSqVCtWjWUK1cOKpV7l8kPG53bX6Ow0TkREZFzbielxowZg3/++Qe7du2CwWBQtrdu3Rrff/99rgZHlJHctNylRudKpZS7jc5ZKUVERJSRr0EDvVaFZJMVkCTl96V88yfZZIVeq4Kvwe3uEEWCRWkxkHH6XnqlFKfvERERZeJ2UmrDhg2YO3cunn32WaWXAABUrVoVFy5cyNXgiDKS7zK61+g8+6QUG50TERFlLbykNyoG+OBGXDJSLFbIv15FUYIkSbgRl4xKAb4IL+ldsIE+ApPJhE8++QSrVq2CyWRy67ny9LxMPaXSK6WsrJQiIiLKxO2kVGxsLAICAjJtT0xMtEtSEeUF0Y2eUkpSyoXpe6YMlVKSC88hIiJ6nKhUArrWDUEJbx3O3UqA2SpClCQkpFpwPiYBJbx16FK3rEc3OTebzfjss8/w/fffw2w2u/dcJ9P35EopNjonIiLKzO2k1NNPP41ffvlFeSwnohYtWoSGDRvmXmREDljc6CklJ65cmb5nylAdxWopIiKizKqX9cPgVpUQUtwIk0VEQooF8SkW1Azxx+BWlVC9rF9Bh1hgLEqjc8eVUmx0TkRElJnbk/4nT56Mdu3a4fTp07BYLJg1axZOnTqF/fv3Y/fu3XkRI5HCrUbnblRKma2Zk1IGrToHERIRERVt1cv6oW31QKhVKpitIhqUL4HXGkV4dIVUbjArPaWcNDpnpRQREVEmbldKNWrUCH/++SeSkpJQoUIFbNu2DWXKlMH+/fvx1FNP5UWMRAoxB5VSrvSUylgZlbFyioiIiB66m2hGMaMWJX30KO6le+wTUsDDSimdxkmjc1ZKERERZZKj5VFq1KiBZcuWZdq+Zs0avPTSS48cFJEzctWTyp2eUi6tvmd/TKrFmoPoiIiIHg+3E1KVr3kjJ418LZHxxpnS6NyF6xEiIqLHjVuVUhaLBadOncLZs2fttv/000+oVasWevbsmavBEWWUo+l77ClFRESUa8xWEfeTHjYBN1n5OxOwbXTuuFLKwu8TERFRJi4npU6fPo3KlSujZs2aqFKlCrp06YJbt26hWbNm6NOnDyIjI3H+/Pm8jJVISTBl7NfgiNLo3JXV9zJURqWaeeFIRETkyJ0Ek91j/s5May8g3wPLuPqelo3OiYiInHJ5+t7777+PiIgIzJ49GytWrMD333+PkydPolevXti0aRN8fX3zMk4iADbT91yqlEr728Lpe0RERLnGduoeULQqpQwGA/bt24c///wTBoPB5efZfg/kyiiZXDnFRudERESZuZyUOnjwIDZv3oy6devi2Wefxffff4+RI0di4MCBeRkfkR3RjUop+aJQdGX6nryMs0qARZTYH4OIiMiJ2PSkVHEvLe4lmZFqLjo3ctRqNerVq4eYmBio1a6vwmt7AyxjpZTccoCVUkRERJm5PH0vJiYGZcuWBQD4+/vDy8sLzZo1y7PAiByRb0S61uhcfo7rPaV8DGl5WvaUIiIicux2fFpSKtjfCABILUKVUjllsbm5JQiOp++x0TkREVFmLielBEGAyqYcWaVSQavV5klQRM5Y00vfXWp0nn5R6MqNSblSykfPpBQREVFWbqf3lCqbnpQqStXFJpMJX3zxBdavXw+TyZT9E9LJVVAZm5wDbHRORESUFZen70mShMqVKyt3fxISElCnTh27RBUA3L17N3cjJLIh95RSu1Qpld7o3I1KqWJKpVTRmYpARESUm+SeUsE2SSlJkjJVCHkis9mMMWPGAABmzpzp+vPkSil15u+BvM3E6XtERESZuJyUWrJkSV7GQZQtSZKU6XtqBxd9GT2slMrB9D2uJERERJRJksmCJFPajZtg/7RG4KKU1lMpYy+lx4lFqZTK/D2Qq6esbHRORESUictJqT59+uRlHETZsi14cqdSKrseDqIoKQ1KffVpU1KL0kpCREREueV2fNqUNj+jFt66h5eRJovocOra48Isyj2lMn8P2OiciIjIucf36oE8ju1Syg6u+TJxdfqebQLKV6mU4vQ9IiKijGITUgAApXx0UKkE6OSpaUWor1ROZFkpJfeUYqUUERFRJkxKkcewvZZzpVJKnr5ncTEpJQiANxudExEROSU3OS/tqwcA6DRpl5KPe4Xxw55SDhqdK6vvudbnkoiI6HHCpBR5DNveUK6svqeRK6Wy6Skl393VqVXQp19cMylFRESUmdzkvJSPfVLqce/FKCelHK6+Z1M9ld2NMiIioscNk1LkMeTeUCoBLq3w42pPKflCUqdRQa9RA+Dqe0RERI7IPaXkpJT8e9Nkfbx/b8rJpqym76Ud93gn74iIiDJyudE5UUGTk0uuVEkBgErl2up7cqWUVi1Ar1XZbSMiIqI0kiThTqJcKaUDYFMpVUR+bxoMBmzfvh0HDhyAwWBw+XnK9D0HTS9VKgEqIW3BFjY7JyIisud2UspqtWLp0qXYuXMnYmJiIGa44/Pbb7/lWnBEttxNSsl9p7Lr36BUSqnV0KmL1sU1ERFRbrmfZIbZKkGtAop7pSelitjvTbVajWbNmiExMRFqtdrl58mNzjUOKqWAtGl9qRYx2+ptIiKix43bSakhQ4Zg6dKleO6551C9enWXplER5Qa5N5QrTc4B2+l7WR8nX0jrNCqlUspslSCKklJtRURE9LiTq6RKeOuV349Ko/MikpTKqYc3uBx3xpCvSSyPeUN4IiKijNxOSq1evRo//PADOnTokBfxEDmlVEo5uQuZkdrNRudataD0xgDSklVGnet3SYmIiIqy2PR+UqXTp+4BRS8pZTabMX/+fJw6dQqRkZHQarWuPS+bSil5u5mVUkRERHbcbnSu0+lQsWLFvIiFKEtKUsrVSinB1Ubnafv1GhXUKkFZtY/NzomIiB6KzbDyHgBl1dqikpQymUwYMmQIFi5cCJPJ5PLz5AbmjnpKAQ+bnbNSioiIyJ7bSanhw4dj1qxZkLKpPiHKbe43Ok/72yJKWf68PqyUSnuCvog1bSUiIsoNd+SklG/mpNTj/jtT7inlaPU9wKZSio3OiYiI7Lg9fW/v3r34/fffsWXLFlSrVi1TWfO6detyLTgiW/Iqeio3e0oBgCQBzp4mL2Mt95PSa1VINFmRan68L7CJiIhs3XZQKaVM37M+3tXFJnn1PSc9peQbX2x0TkREZM/tpJS/vz9efPHFvIiFKEvyhZzG1UopmyyUVZKgguPnZa6UUgMwP/YX2ERERDKLVcS9JDMAoJRNTym5F2NRmb6XU/I1irNKKflGmZnT94iIiOy4nZRasmRJXsRBlC35gs/VFfFsK6WsogStk57lpvRSennFHPmubworpYiIiAAAdxJNkKS06Xo++oeXj0Wt0XlOyckmrZNKKfmGmoWVUkRERHbc7ilFVFDc7Sll2xA9q3J5pVJKw55SRESU+0RRQtTtRFxKAKJuJ0L0wMREbHza1L3SvnoINr9f5Rs6j/vvTGX1PSfXKHLyjo3OiYiI7LldKQUAa9aswQ8//IDLly9nWpnk6NGjuRIYUUai5N7qeyqVAEFI6ydlzaLRuXx3U283fY+r7xER0aM7eS0Oa49exbmb8bgRo8LBTf+hUqAvutYNQfWyfgUdnsvuJKZd75W26ScFPEy2MCmVdaXUw+l7npeQJCIiyktuV0rNnj0br732GgICAnDs2DHUr18fJUuWxMWLF9G+ffu8iJEIwMOSd1crpYCHCays7krLlVI6VkoREVEuOnktDrN3nsOJq3Hw89KitAHw89LixNW07SevxRV0iC6TK6VK2vSTAh7+ziwq0/f0ej02bNiAjz76CHq9PvsnpHvYU8pJo3MVG50TERE54nZSat68eVi4cCHmzp0LnU6HUaNGYfv27Rg8eDDi4jzn4oo8j5iTpFT6sS5N31M/XH3PdjsREZG7RFHC2qNXcTfRhIoBPvDRayAIgI9eg4oBPribaMK6o9c8Ziqfo5X3ANvV94rG70yNRoMOHTqgXr160Ghcn1AgXzNonDQ6l7ebxaLxfSIiIsotbielLl++jEaNGgEAjEYj4uPjAQCvvvoqVq1albvREdlwt9E58HAFviyTUlb7Sin2xyAiokcVfScR52MSEORnhCAIuJtowqUEAQkpZgiCgCA/I87FxCP6TmJBh+qSOwkPe0rZUqbvPeaLg1jSk01yRVRGmvRrCzOvLYiIiOy4nZQKDAzEnTt3AABhYWE4cOAAACAqKgpSFn17iB6V3BfKWRNRR+Q7k1n1lMpcKZXeU8rMnlJERJQz8SkWpJpFGHVpv1PuJ5thFYG4FAsAwKhTI9UsIj79cWGWbLIiITXtd2Km6XvqtPdnshaN35lmsxnffvstdu7cCbPZ7PLzLHKjcyeVUloXKreJiIgeR24npVq2bImNGzcCAPr374/33nsPkZGRePnll/Hiiy/meoBEMmX1PRcbnQOuVUopjc7ZU4qIiHKJr0EDvVaFZFNaskaepif/Pko2WaHXquBryNGaM/lKnrpXzKhRFgORyZVSVrForCxnMpkwYMAAzJkzJ9NiPlmRG5hn2+icSSkiIiI7bl8JLVy4EGJ6ifIbb7yBEiVKYO/evejUqRPeeOONXA+QSJaT6XvytWFWLRzY6JyIiHJbeElvVAzwwYmrcaio91F+h1lFCZIk4UZcMmqG+CO8pHcBR5q9WLmflHfmxt/y704gbTq8xklSpqiTb3A5q+aWk1VFIXFHRESUm9xOSqlUKqhs5st3794d3bt3z9WgiBxRKqXcuN6Vq6qynL6XYRln+S4wG50TEVFOqVQCutYNwbV7yTgfk4AkkxWSBCSZLDgfk4AS3jp0qVvWrRstBeV2vON+UkBaBZBGJcAiSjBZRHjpMh1S5EmSpKwQrNU46ymVNs7yND8iIiJKk6PbWX/88Qd69eqFhg0b4tq1awCA5cuXY+/evbkaHJGth0kp139sVdn0cBBFSSm5VxqdK5VSRaM/BhERFYzqZf0wuFUl1AjxQ5LJghQRSDRZUTPEH4NbVUL1sn4FHaJL7iSmTWPLuPKeTFmB7zG9mWOxucZwVimlSb924ep7RERE9txOSq1duxZt27aF0WjEsWPHkJqadvcsPj4ekyZNyvUAiWSi5H5PKflY0UmllO0S1jo1p+8REVHuql7WDx8/VxUNIkogzEdC80ql8NFzVTwmIQUAsemVUhmbnMse99+bZgfXEhlp1Wx0TkRE5IjbSanPPvsMCxYswKJFi6DVapXtjRo1wtGjR3M1OCJb8jWfO9P3squUki8kBeHhBaNey+WtiYgo9whC2u8WXy1g1Gk8YsqeTJIk3ElwPn0PsK0wfjx/b8oV1yrBed9LpdE5p+8RERHZcTspdebMGTRt2jTT9mLFiuH+/fu5ERORQ3JfKJU7lVLZJKWUJudqFYT088o9pVItVkhZ9KIiIiJyhUWUlBsrnjY1/EGyBSarBJUAFHfSMOpx78VoydCb0hE2OiciInLM7UbnQUFBOH/+PMLDw+227927F+XLl8+tuIgysab3YZCbhbpCk11SSrmQfHhOeRqCKKX9R0LrxusRERFllGJ+mIjytGoieeW9kt465UZPRkpPqSKQcNHr9Vi5ciWOHTsGvd5xZVhGSpPzLK4XlEbnnL5HRERkx+1KqUGDBmHIkCH466+/IAgCrl+/jhUrVmDEiBF466238iJGxeTJkyEIAoYOHapskyQJ48ePR3BwMIxGI5o3b45Tp07laRxUMORrXXcqpVTZrL5nttg3OQceJqUAz/vPAxERFT4pNtPBzVbJo6plbqcnpUo5mboHFK1G5xqNBi+99BIaN24Mjca1e7fy+85qIRal0bkHjT0REVF+cLtSatSoUYiLi0OLFi2QkpKCpk2bQq/XY8SIEXjnnXfyIkYAwKFDh7Bw4ULUrFnTbvu0adMwY8YMLF26FJUrV8Znn32GyMhInDlzBr6+vnkWD+U/Mf3uosaN1ffku7qi00qptLvXtiX3giBAr1Eh1SIi1WyFj97tjwkREZEi45S9FIsIH3caJBYgJSnlZOU9ANCri05SKifk6iddFpVSbHRORETkWI6uiCZOnIjbt2/j4MGDOHDgAGJjY/Hpp5/mdmyKhIQE9OzZE4sWLULx4sWV7ZIkYebMmfjwww/RpUsXVK9eHcuWLUNSUhJWrlyZZ/FQwVB6SuVio3OTg0op28eslCIiokeVkmHhjFSz5/SVuh2ffVLq4e9Mz3lfzlgsFqxZswZ//vknLBaLa8+xyu0FnF+gsNE5ERGRYzkuAfHy8kK9evVyMxan3n77bTz33P+z9+Zxjtz1mf9Tt26p73Nmeo72+BjbeDzG+AAMxgYcWIjNEY4QICGACUfIhmNzAAsBwua3yRKyYXdDYEMWDIlNgDCAjQk2xge2xx7PeOy5+751H6VSXb8/St9SSa27pW6p5/t+vfSabnVJUy21qr71fJ7P8/kNvOIVr8DnPvc5+/7z589jaWkJt956q32fJEl46Utfiocffhjvfe97yz6foihQFMX+PpFIAABUVYWqqvb95GvnfZStQ1E1GIYB09DXvSeV3ivGNGAYBnKqVvZ9zCg5GIYBnil+LM8AhmEgnVWgqtQp1Uro56o7oO9T90Dfq84nJSsw8rmIhmEgJSsISN3hlFpJZGEYBkIutuLfGMeYMAwDGUXt+r/DdDqNt771rQCAP/7jP66rhU9WVBiGARZm5d/f0KuuRyjNQY9/3QF9n7oH+l51B93yPtW7f3Vfbb/73e+ua7t//Md/rPcp6+Kuu+7CkSNH8Pjjj6/72dLSEgBgaGio6P6hoSFMT09XfM4vfOEL+MxnPrPu/nvvvRcej2fd/ffdd1+ju01pA8cWGSxmGPxansLaifLblL5Xz64wmE0yeCQ9jcjz66uTZxPA7CoLPWzicKyQRTY9xyCqMPjZf0xhdP2fBKUF0M9Vd0Dfp+6Bvledy/kkML9iiVDzc3O49/4ZDLq3eKfqQDeBo+dZmCZwhJnCcxVWjcejDGYjDH4dm4Yx3d1OoGw2a3/985//HC6Xq+ZjZlLA7DKL7KqJw6nnym6TUoHZGRY8Cxw+fLZl+0uxoMe/7oC+T90Dfa+6g05/nzKZTF3b1S1KfeMb38CuXbtw1VVXwawQGt1qZmdn8eEPfxj33ntv1UUBUxJ8bZrmuvucfPKTn8RHP/pR+/tEIoEdO3bg1ltvRSAQsO9XVRX33XcfbrnlFgiCsIHfhNIKlh+eBr+axg0HR3HVjlDRzyq9V9mn5mHMxHHo0kG8eLJ/3XM+fDaM5ePLuGIsgNsOjdv3L/5qCufXMrj20BguHwu27Xe6EKGfq+6Avk/dA32vOp9HzoUxf3QR83NzGBsfx7Uv2oVLRjo/93I1qeAx7SxEnsXtt+2vuLYKnAkj8ewyLhkP4rarxzZ5L1tLOp22v375y1+OUChU8zFHZmKYe2oBFw35cNuLdpbdJiGreObe02AZ4LbbLm3V7l7w0ONfd0Dfp+6BvlfdQbe8T6QjrRZ1i1Lve9/7cNddd+HcuXN497vfjbe//e3o7e1tegfr4cknn8TKygquvvpq+z5d1/Hggw/iK1/5Ck6ePAnAckyNjIzY26ysrKxzTzmRJKnsmF9BEMq+qZXup2wuDMOCZVm4xMrvR+l7JQo8WJYFWLbsYwxYz+mWih/nkQSwLAsd5R9H2Tj0c9Ud0Pepe6DvVeeimYx1LgLAsiw0MF3xXsWyMliWxVDABVEUK25XOGd2x+9VDef+1/uZMvPrE0ngK27vdvwNcBxvZ15SWgM9/nUH9H3qHuh71R10+vtU777VHWjwP//n/8Ti4iI+/vGP44c//CF27NiBN73pTfjpT3/aNufUzTffjGPHjuHpp5+2b4cOHcLb3vY2PP3009izZw+Gh4eLbGu5XA4PPPAArr/++rbsE2XrsIPOq7jgSiHbVprATILM1wWd58NKFZUGnVMoFAplY5SeS7JdEnRez+Q9oHAOvWCn7+XDy4Uq0/eck4NV48J8nSgUCoVCKUdDCc6SJOEtb3kL3vKWt2B6ehrf+MY3cOedd0JVVZw4cQI+n6+lO+f3+3HgwIGi+7xeL/r6+uz7P/KRj+Dzn/88JicnMTk5ic9//vPweDx2SCVl+0Am6HENVBe5WtP38mqVUDIxRxK2zyQhCoVCoWwt2ZJzSbcUPOoVpSSeA3DhilJEZOKrjAfmHWuXSmsSCoVCoVAuRJoeK8YwDBiGgWma9kSZreBjH/sYZFnGnXfeiWg0imuvvRb33nsv/P7Oz2qgNEZTolTeKWVUcPOpFZxSF/oCu10Yhonza2lMp4Dza2nsGwrSFgYKhbLtyeZFKGL07ZaChy1K+etzSnXL79VqyFpC4CuLUizLgGMt57aqU1GKQqFQKBRCQ6KUoii455578I//+I946KGH8JrXvAZf+cpX8KpXvcruk283v/jFL4q+ZxgGn/70p/HpT396U/5/ytbRjChFBA+thlNKLHVK2QtsKkq1iuPzcdx9ZA6nl5JYXGHx639/HpPDftxxcBwHaJg8hULZxpB2PQ9Pvu/8c4thmDizkkIsoyKT02AYZsUigrSN2vdEUcQ//MM/4OjRo1VztJyQNYZQY33Csyx0w4BWKVOAQqFQKJQLkLpFqTvvvBN33XUXdu7ciXe9612466670NfX1859o1CKIG4nroFMKWKXNyqIUqpe3Sl1oVZ9W83x+Ti+fP9pRNI5DAUkaC4g6BFwbC6O+aiMD908SYUpCoWybSEilIc389939rnl+Hwc331iFg+eWoNumEgrZ3Df8HLFIsJ2EqUEQcA73vEOHD58uO6AVrKW4LnqBVqBY6BolQtlFAqFQqFciNQtSn31q1/Fzp07sXv3bjzwwAN44IEHym53zz33tGznKBQnG3FKVcyU0so7pUTqlGoZhmHi7iNziKRz2DfoRTyjAgzgk3j4BwWcWUnhniPzuHQkQFv5KBTKtoQUODw8YKCzRSlSRFiKZyHyLFwCi16vVLWIYAedX6AOIBJ0zlcJOgcAjmUB6LaIRaFQKBQKpQFR6h3veAeYBhwqFEqrIdP3msmU0itkShHRaV3QOU+n77WKqXAaZ1ZSGAm6sRDLYi4qQ9SAXbDab0eCbpxeSWIqnMaegdYOS6BQKJROgDil3ByQBpDt0IKHs4gwFJCQzunwiDx8Lh77JF/FIgIRpVTdrNrm1w1omobDhw/jiSeewK233lqXW4qITEKNKAsynU+jmVIUCoVCodjULUp94xvfaONuUCi10fXmp+813L4nXNhV31aSzGpQVAOSn8FywgrNzTlMAm6Rw3LCQDKrbdEeUigUSntxZkql0bkFD2cRIZHNAQBc+fNhtSKC022c0w24WG5zd7yFKIqC17/+9QCAP/qjP4Lb7a75GJVkStVwSpHpfNoWDgiiUCgUCqXT2Jx0cgqlBRC3E9uAY4+t4ZSq1L5nZ0p1cItFt+B38ZAEFguxrJ2j4dQI5ZwOSWDhdzU9DJRCoVA6FtM07QKHnSnVoXmFpIjgFjl7Qpzz/OgWOSjq+iICz7Egm3Wq4NZOtDozpUh7H52+R6FQKBRKASpKUboGkgvFb4ZTimZKtYyJPi/2DfhwPpyGmRcHtfzbYZomFuMyJgf9mOjzbuFeUigUSntQNAOkLuK1p+91pihFighyTnfkJBXOj9WKCCKXL+bonfm7tRPyWtVySpGfV8q5pFAoFArlQoSKUpSugSziGsmqIGvpSgtAIjqVilI06Lx1sCyDg7t6wDIMElkVqm5ANxikFA1nVlLo9Yq4/eBYV2eQUCgUSiWIc4hjASnf1WYJVZ0nTEz0ebFv0IfFuAw1Ly6R4k6tIoK4jSbwNQpxwvE1MqU4lmRvXXivEYVCoVAolaCiFKUrME3TbvlqaPpevn2v3Phl0zQL7QlVnFKdeOHQbSzFszgwFsTFw37kdANpDYhlcrhiPFR2khOFQqFsF0irnovnkI9ngml2ZtGDZRnccXAcvV4RK0kFqm6AAZDK1i4iSBewKEUKXyJfn1Oq3JqEQqFQKJQLFSpKUboCp9OJayBTilQtjTLCkjPEvNRyTzKlgM68cOgmFmIyzq2l0e8T8Te/dRWu3hnCLp+Jj73yIvzpb1xCBSkKhbKtIa16Es+CYwCi53TqueXAWBAfunkSfT4JOc3AciKLmFy7iNBOh7FhmDi3msLR2RjOraYqtuRvFWqdTimeOqUoFAqFQlkHTRamdAXOoPIaa74iWLt9b/3PnNXc0qBzgWPAMIVqtkvo3klCW81DZ9YAAJePBdHvkzDolxATgAGfRFv2KBTKtiebb9+TBA6MCrgEDlnNtAZpuIUt3rvyHBgL4uqdPVhJKnjjoXHs7PVgos9b9ZjdLqfU8fk47j4yhzMrKWuSq8Bi36APdxwc75iihmrnb9WYvkecUjTonEKhUCgUGypKUboC5/TkWpVIJ3bQeRmnVGGyEAOmxH3FMAwknkVWNS7IVoRWEZdVHJ2NAQBunOwHAHgl67CTzmmVHkahUCjbBoW07+V791w8i6ym22JVJ2IYJrKagYBbwKGJXvik2stFO1OqhS6g4/NxfPn+04ikcxgJuuEOcpBzOo7NxTEfldvS/i2KIv7H//gfePbZZyGKYl2PIdP3hBrT92jQOYVCoVAo66Hte5SuQHOoUo2Ya0irX7kFYK5CyDmBtPApHTq6uxt49FwYhglM9Hkw3uMBAHhE63VNK/R1pVAo2x8iPrny5xpyzsl28LnFuW+eOp3CrXZKGYaJu4/MIZLOYd+gDwLPQM7p8Ll47Bv0IZLO4Z4j8y1v5RMEAe9///tx2223QRDqc7KRjKhaohQNOqdQKBQKZT1UlKJ0BUST4lisczVVgzilqolSlRaREp3AtyFymoFfn48AAG7Y12/fT5xSKYU6pSgUyvaHZEqJ+UIHaQcn93cimVwhB6veNutWT9+bCqdxZiWFkaAbDMPg1FISzy0lkNMMMAyDkaAbp1eSmAqnW/L/NYtpmoXpezXa9wSWBp1TKBQKhVIKFaUoXQHJlGok5Byo3r5HFpGVnFJ2aGsHt1h0Mk/NRJHJ6ej1Crh0JGDf7yNOqVznXpBRKBRKqyDik92+J3R+wUPOH5+Js7UeRK617uJkVoOiGnCLHACrndA0gVz++d0iB0U1kMy2tsCh6zoeeOABHDt2DLpe+3fRDRNkiSHUCjrnqFOKQqFQKJRSaKYUpSsgTieukZRzAGwd7Xu1nVJUPGkU0zTxq3zA+Q17+4sq7XamFHVKUSiUC4CsRtr3OKgotIZ3slNKzu+bu4EhH612F/tdPCSBhZzT4RZZW/ghLiM5p0MSWPhdrV3KZrNZ3HLLLQCAP/iDP4DL5aq6vdP1RIPOKRQKhUJpHOqUonQFBVGqscdVa98jlUqpUqZUfjFOg84b5/mlJFZTOUg8i4O7eop+5pWs1zVDnVIUCuUCQKnglOrkoHNyfHY34pRqcfveRJ8X+wZ9WIzLdpA4YIlApmliMS5jctCPiT5vS/6/ZnG6nvgarY7ESUWDzikUCoVCKUCdUpSugLTv1ZttQaiaKVWjfY9mSjUPcUldu7vXzk8heEWaKUWhUC4ciFNKElgk0R0u3Ex+OqpHrH+ZKLb4nMmyDO44OI75qIwzq2mougGOZZCUNUTTOfR6Rdx+cKzhdUGrIa4nocwk31KIU0o16LqCQqFQKBQCdUpRugIyXafRTCm7fa9cphSZvkeDzlvKQkzG2dU0WAa4bm/fup+T9r0Mnb5HoVAuAIhTSloXdN6555amMqVa7JQCgANjQXzo5knsG/AhpxlIZTUksiquGA/hQzdP4sBYsGX/V7MQgYmvI15AoO17FAqFQqGsgzqlKF0ByWyoZY0vxQ46p5lSbcUwTEyF00hmNfzqzBpM08SB8RBCHnHdth476Jw6pSgUyvZHsTOlrHMKObd0RaZUQ0HneVGqxSHeB8aCeMf1u6BoBlTdwDUTPfjdG/dsuUOKYDul+Nr7Q4QrGnROoVAoFEoBKkpRugLSftdw+x5Tefxy7fa9/CShDq5mdwLH5+O4+8gczqykkFY0LMaz8Lt43HLpUNntfSToPKfDMMyOubCgUCiUdkDEJ0noHlHKzpRqIOicZGW1I4dRUQ0E3AIAwOcSOuq8QQSmeopmpFBWbk1CoVAo7cRZQPa7eEz0eTvqWEq5sKGiFKUrMMzm2ve4vFXeMK2JcM68B7t9r0amVKurvtuJ4/NxfPn+04ikcxgJumEYJgSORUbR8Z3HZzEUcK1rryBOKdMEMqpui1QUCoWyHclWaN/r5Nbwptr3OPJ7tV5syzgEvE4bkqHamVL1tO9Z22h0XUGhUDYRZwFZUQ1IAot9gz7ccXC8I9qgKRR6NUjpCjbqlAIsYco5rblmplS+6qt0cDV7KzEME3cfmUMkncO+QR9MEzi1koPAsdg37EU4ncM9R+Zx6Uig6H3jWAbkOiejaFSUolAo2xqSHVU6fa+Tzy1E+CkdVFGNdmRKEWSHEJVp85AMQRDwhS98Ac8//zwEQai5vWZUjwJwYged00wpCoWySZQWkN1BDnJOx7G5OOajcsfk81EubGjQOaUrIKIU16Ao5cwdLZ3Ap9bbvtfB1eytZCqcxpmVFEaCbjAMg3BagaabkHgWvV4RI0E3Tq8kMRVOr3uslH/J6QQ+CoWyndF0w27VcuXPKeTcku3gcwvJlNrqoHOC0x0lt1nME0URf/RHf4Tf/M3fhCiuz0UsxTl9rxZCflGi0el7FAplEygtIPtcPDiWgc/FY9+gD5F8Ablc9i6FsplQUYrSFejNBp07nFKlohRpy6tU3Wz1eOvtRjKrQVENOwg3IVsC04BfBBgGbpGDohpIZtcLTy7eei/SdAIfhULZxjiFJ9ISTpxSnZwpJecHUXjE+p2stiilmy2/wJEdgzE6rX2PrCW4OqbvEacUzZSiUCibQVEBGcBMJI1IOgcAYBimagGZQtlMqChF6Qr0fKYU22imlEPEIs9BqDdTartP3zMME+dWUzg6G8O51VTdFxN+Fw9JYO22CjJNj7TjyTkdksDC71p/USNSpxSFQrkAUOw8KdZuYybnFlU31xVLOgHTNAtB5w04pSTHubTVWYxOd5SiGW3NZNJ1HU888QROnz4NXa99/idOKbEOpxQprGm0fY9CoWwCzgJyUtGwFFcwEykIUNUKyBTKZkLDXChdARFK+DoWfU4YhgHLWHlSlZxSFTOl8i0W7WhF6BQ2Enw40efFvkEfjs3FsVvw2LkpHpGHaZpYjMu4YjyEiT7vuse6OCAHIE1FKQqFso0hTimSUQgU2vgAyy3l7bBcPUUzQE6XjbTv8WzhfJvTjYbyqGpR6o6SVR3+OjKcmiGbzeL6668HAPze7/0eXC5X1e2JQMbXlSlFg84pFMrm4SwgkyJJTjPt6dfVCsgUymZCnVKUroBY3Rt1SgEFt1SpA6jglCr/nNI2b98jwYfH5uIIuUVM9HsRcos4Nmfdf3w+XvXxLMvgjoPj6PWKeG4pCVU3IHAMsqqBMysp9HpF3H5wrGw4vZS/VknnqChFoVC2L0rJ5D3AOnYSV00nnl+I+1XgmLrCuwkMw7QtV0ouFaU6qIVPbSBewA46N0yYJnVLUSiU9kIKyItxucRxqtsF5MlBf9kCMoWymVBRitIVGE0GnQMFIau0fc8OOufKV3NJZVttQz7GVtOq4MMDY0F86OZJjIXcyGkGZFVHTM7hivFQ1WkervxLTtv3KBTKdqZ08h6BuIg6MVeKXLg00rpHaIco5WwnJMJPuoNEKa1GPqUTEnRumsA2W1ZQKJQOxFlAnovKUHUDhmkimsnVLCBTKJsJ9epRugLilOLa4pSqEHTuWGAqmtHUAr1TcQYf6oaJM6tJDPol9HqldcGHewZ8VZ/rwFgQL9s/CJ5lcXBXCC/bP4iJPm/VE5zEkaBzKkpRKJTtSzafSehs2QMASeCArNaRopSdJ9VE+53Etd5hrOqmvQbo9YpYSSqd5ZRqQJRyRhCougGO3T7rCgqF0pmQAvKnfvAs5qMydMNEOJXDC3f34faDYzXjOiiUzYCKUpSugLicmnFKkcc0GnTOcyx4loFmmFA0fVuJUnbwYZBDOKUgIWtQNQO9XgmAVSFfTtQffLgYzyLgFnDDvv6aIhbgaN+j0/coFMo2hohOUolTirSHEydVJ0EEn0bypAiS0PosRrI/LAP0eASsJBVkOqj1W9Xrz7x0tvjRCXwUCmWzODAWxMGdIQwHXFB1A9dM9OB3b9xDHVKUjoGKUpSuoCXte44FoGmayOUXkkKVhaTIs9ByekfmfmwEZ/AhCXyXVWuiEc+xDQUfKpqO1ZQCABgNuev6/122KNU5FxYUCoXSahTSvlfilLLb9zpwuisRfJpxSolc66fWZlRrfzwiB49onZNKg8+3Es0gTqna6xOGYexil04n8FEolE1C1Q2kFB0BtwDAKrxTQYrSSdBMKUpXQASlZg6gxFHvFKWc46orOaUAR9h5B1azN4Iz+FBxtI+kFK3h4MPluALTtISugEuo6/+X8i95RtW3XV4XhUKhEIg4UzqJjmRMdeK5pZAp1Xjdsh2ZUsQp5RZ5ePI2204SpWynFFvfkpoU11Sj8957CoWyPYmmc0XfR0q+p1C2GuqUonQFhrmRTClroegUpVRHhVKskgNhL7D1zlkAtwISfDgflXFiMQGWYcCxDFaTClaTSkPBh/MxGQAwVqdLCgBEDmB0K+w1o+rwddhIdAqFQmkFFYPO+c51Sm2kfa8dolTGsT/EvSWr7XPZCoKAP/3TP8Xp06chCLULLY1kSlnbMVA0QKNOKQqFsklEMpYI5RE5ZHI6opkcDMOkbilKx0CdUpSuQNtA+x4RsgxHphRZMAscA6aK0EXGeHdi7sdGIcGHvR4ROc1AKqthLa3UnJxXykJelKq3dQ+wskHIxQVt4aNQKNsVO1OqQvue0slB582IUqR9T2+hU0otiFKb0b4niiL+/M//HG95y1sgimLN7bU6ogCc8ByZ7Lv91hUUCqUzIc6oXX0e8CwD3QDisrrFe0WhFKD2BEpXoG9ElLLb9wr3kcVgNZcU4Gjf22aZUoRLRwI4MBZAXNag6gZcAouPv2q/HVZbD/O2KOVq6P/2SRyymoaUomGooUdSKBRKd0BEqVKnVEcHnRMRqJnpe0L7nFIugbPdW504fY+v1ymVX8fQoHMKhbJZRNOWANXrFdHjEbCayiGczqHHW1t4p1A2A+qUonQFdvteE3+xxJpalClVY/IegSywWxna2kkksxpMMAh5BEz0e+GVBCzEs3U/XtUNLCes7Rtp3wNgV7ypU4pCoWxXSEFjfaYUceF23rlFboVTqqWZUs6g8/ZnShmGgWeffRYzMzMw6sh9IuISX2fRjIhXOs2UolAom0Q0377X6xXRmxeiaK4UpZOgohSlKyAuJ67OIFEn5dr3FK2+DAjScrFdnVIx2TohBd2CHWo+FU7X/fjlRBaGaV0sBN31hZwTvPnA2hQVpSgUyjalYqaU0Lku3EyHZUo52/fcmyBKybKMq666Ch/60Icgy3LN7dU61xMEPt/mp9JMKQqFskmQoPMej4henwQAiKSVrdwlCqUIKkpRugJSUWwu6Hy9U8pu36vhlGrHAruTiGUsO2+PR8REnwcAML1Wvyi1ELNcUqMhd9VsrnJ4badU5zkFKBQKpRWQIPNKmVKd6JTK5EPES91d9UB+z3a077kF3pEp1TnFDOKUqjdTSsgX12jQOYVC2QxM00Q4XXBK9eWdUmHqlKJ0EFSUonQFxCnVhFGqIEqVCTq/0DOlYvmQw6BHwK5+yyk1HcnANOtbLC/Yk/cay5MCCk4p2r5HoVC2K0reKSV1kVOqMH2v8djRQiGndWKbcxogcW+putkxQeGNTt+znVK0fY9CoWwCsqrb55qQRyi076WoKEXpHKgoRekK2uWUKr1QKMUWpTqwmt0KYvke85BbwEjABYlnkVUNLCfqs/TONzF5j+CTrAse2r5HoVC2I6Zp2k6pUtdRYbJrZ51bVN2w28qaad8j58xcCwUj5zRAiWdBopva2cLXCHamVL3T90jQOXVKUSiUTYBkR/ldPCSeK3JK1VuEplDaDRWlKF2B3uCizwnLVA46v+AzpfLteyGPCJZlsKPXauGrJ1dKN0wsxQvte43i7cA2DAqFQmkVOd0AWe9LJa3ipCDSadP3SH4Ty6zf53oQ7UJOO9r3ODAM03ET+Mh6gq/Tyk2CzrUOcXpRKJTtjTOqAwB6vCIYxrq26RRxn0KhohSlKyAFRXYDTinDIUopep3tex3cYtEKCqKUFVJu50rVIUqtJLPQDBMugbWrLo1gB51nqShFoVC2H0RwYpn15xrbKaXpHVWplksEoEZph1MqqxYHr7s7rKBBCl611hME4pRSjc553ykUyvYlYoecW2t9gWMRcAlFP6NQthoqSlG6AiIocXWOXHZCWv7KZUoJtYLOuW0edC4X2vcAYJc9gS9T87EkT2o02HjIOQB48+17aVqloVAo2xDS9u0qI/CQTCnTbK2As1E2MnkPcDilWnTO1HTDfi6SceXZhAl89WIYZsPtewJ1SlEolE3EFqUcBWQadk7pNBpPsaRQtgCy6GvGKcWS/IZy0/dqVDYLYbRbv/htNVlVtyv5wXz1ZEevGyxjOahimRxCnsoOqPlY8617QPGFhW6YTQmOFAqF0qmQ42u5NjiRs7KRDNParnQ631ZhO6WaCDkHWl/IIe2EDFM4H7dblBIEAR/96Edx7tw5CIJQdVtnWHndmVLc+jUJhUKhtItopjB5j9DjFYG1NCLp+jJkKZR2Q51SlK7AMJvPlCK6k1EmU6pWZsZ2zpQirXsekbN/T4nnbJFpuoZbynZKNTF5DwA8AgfGDqztjDYMCoVCaRVKhZBzAGAYpnB+6aCwc1m1jsXuGkNAKiE62vda0ZZYrp3QnX895Ta9bqIo4otf/CLe+c53QhSrt6Y7syqFejOlaNA5hULZRKJ2+14ZpxSdwEfpEKgoRekKyMKvuel7bNFzAI0Enbc+tLVTKG3dI+zqqx12bhgmFvOi1FiTTimWZeDNV7zpBD4KhbLdIE4pVwWBx9WBYeeF9r0mnVJ8a9sSy7UTkn3LdMB5Q9VItEDBlV0LEoiuGZ3zvlMolO2JaZqI5ovQTqcU+ZpmSlE6BSpKUboCvQWZUoZZpn2vbqdUZ4XRtoLSkHPCRD5XqppTai2lIKebEDkG/T6p6X2wc6WUznEKUCgUSivIVnFKOe/vpPZwIgK5ms2U4ljbAduKFj57fwSnKNXe9j3DMDA1NYXl5WUYNYQj0r5X7+Q9oOD4VqlTikKh1IFhmDi3msLR2RjOraaKOj9qkZA1aIYJlikuQlNRitJp0EwpSldABKWmRKn8WtFZtFVsp1T15yOilWFa+Q+1tu8miCgVLMmNIk6ppUQWck6Hu8zFyXzeJTUSctddHS6HV+QBKEh3QMWbQqFQWolSJVMK6EynFGmX81QQ0mrBMAxEjoWiGS0RpUg7YbFTKi9Ktal9T5ZlXHTRRQCAN73pTZCkyoUX0oLXyNqABp1TKJR6OT4fx91H5nBmJQVFNSAJLPYN+nDHwXEcGAvWfHwknycVdAtF6/U+n7X2T2Q15DSjZpGeQmk39C+Q0hVsKOi8zPQ9UqGs7ZQq/Hy75UrFK7Tv+V0C+n0iTBOYiZR3Sy1sMOScUHBKUVGKQqFsL7JqfU6pbAc5pUhOU7PT94DiXKkN70+OTN4r7A8plMgdkEVIXNe1ogCc8GWGr1AoFEopx+fj+PL9p3FsLo6QW8REvxcht4hjc9b9x+fjNZ+jXMg5YLVBk3w+sg2FspVQUYrSFRgbad/LP6aZoHOWZewKaKumCXUK0QrtewCwK9/CVylXasHOk2ou5JzglWimFIVC2Z4QsanSeaYTMwsz9vS95kUp8nu1pn0vH7zuyLiyM6Xa1L7XCESU4htYm9jT96hTikKhVMAwTNx9ZA6RdA77Bn3gWSCn6fC5eOwb9CGSzuGeI/M1W/nKhZwTiFuKhp1TOgEqSlG6Am0DohSxqxYFnevWYlbkai+87QuHDqpmtwI7U8q9/kQ1kW/hmy4jSpmmabfvbdQp5SNOqQ6oeFMoFEorIW15Ui2nVAdN38uqGws6B6xcKaA17uJMmXZCj+2U2vrXjaxNGnNKkaBz6pSiUCjlmQqncWYlhZGgtc4+sZjEicUEdMMAwzAYCbpxeiVZdSgRUMiMKnVKOe+juVKUToCKUpSuoBVB5+Xa9wS+9vMVxnZvn6qmYZhIZPOilLeyU2ouKq+r5kbSOSiaAZ5lMOjfqFPKuvBJ0aBzCoWyzSCFDBdfXpQiBY9Oat+znUlNZkoBgCS0zilVrp3QGXS+1QNIbKdUI6IUDTqnUCg1SGY1KKoBt8hB1Q1ohgndKMRduEUOimogma1e1CWteT1VRKlwWmnx3lMojUNFKUpXYAedN5EpxZVzSuUXy2IdC8mCU2r7iFKJrArTtELg/dL6ini/T4RX5KDqpp0fRSDfDwddTYmETnw0U4pCoWxTiFOKBJqXItlOqc45t7Sifa8dTil3mUwpzTBbklu1EWjQOYVCaQd+Fw9JYCHndOS0wvULKeLKOR2SwMLvqu5qjaStAnRvufY96pSidBBUlKJ0BURQambSW6koZZqFhWw90yZI1Xc7te/Zk/fcApgyQh/DMNjVXz5Xat7Ok9pY6x5Ag84pFMr2pWbQOd9Z7XuGYdoC2caCzvPu4hacM0mYubOdUORYO8Npq1v4msqUokHnFAqlBhN9Xuwb9GExLiPnOJamFA2maWIxLmNy0I+JfGdDOTTdsLsiesp0RdD2PUonQUUpSsdjmibI2m1DQed5t5WqmyCO/3pyIEjVdzsFnRM7b7k8KUKlXKlW5UkBgDd/4ZOm7XsUCmWboajVg85dQme5cGWHOLaR9j2xpUHn+rr9YRjGdku1I+yc53m8733vw6tf/WrwfHUXQiEKoJlMqc543ykUSufBsgzuODiOXq+I8+E0VN2AYZqIpHM4s5JCr1fE7QfHqhbrY7LVFSFwjN2Z4ISIUtFMrmZgOoXSbqgoRel4nG13jVQjCSxT7JRy2v3rat8TSNV3+ywgY3LeKVVm8h5hwp7Al7FzO0zTtCfvjW5w8h4A+PK2Y1nVaSsDhULZVmQ10r7XHUHnROCReLYpVzKhLaJUiXPL00ZRSpIkfPnLX8Z73/teSJJUdVsiLAls45lSGs2UolAoVTgwFsSHbp7EeMiNnGYgldUg53RcNOTDh26exIGxYNXHOyfvleuKCLgE8CwD3QDi+esCCmWraH68CoWySTgt7mwLMqXU/EJZ4Ji6Ft7bcfpePN++V25ELGEk6ILAMcjkdKymFAz6XYjLKjI5HSwDDAU2Lkq5BQ4sAxgmkM7pCLqpTk6hULYHSo32Pfvc0iGiFGmF20jrHlD4vTaa96Qbpl0MKt2nTpnAR4QlvoFMKRp0TqFQ6uXAWBC3XjYMt8hD1Q0IHIvbrx6vKUgB1SfvAZYbq8cjYDWVQzidKxuGTqFsFvQKkNLxGObGnFJ8SfseWSjXO8J5O07fi5H2vSpOKZ5jsaOHtPBlABRa94YCroZGYFeCYRg7V4pMfaJQKJRuRzdM5PKiQ6Wgc9sp1SEu3HKT7ppBtMW2jf1e1doJ3WL7zhumaWJ1dRXxeLzmdL9mpu8JtH2PQqE0QErREHAL2N3vRcAtYD4q1/W4apP3CDRXitIpUFGK0vE42/eaMEo52ves7+3Je3VmQGzH6XukfS/krixKAcCufK7U1JqVK0Um77UiT4rgsXOlqChFoVC2B05nLSlslNJ57Xtk1PjGTPQS1xqnFHFBuYT17YSe/GuXacNrl8lkMDY2ht/5nd9BJpOpui1xO4lNOKU0w6wpelEoFEoiv2a/dDQAAJgJVz8uEapN3iP0+qT8tspGdpFC2TBUlKJ0PKSYyLNM2Z7oWpQGnduT9+qsbLYyH6MTME2zMH2vilMKACbyE/iIU2qhhZP3CCR8MUXDzilNYhgmzq2mcHQ2hnOrKRrYSdlyyBQ7gWMqDuggDipVN4uKL1tFq51SGz1nVmsntDOltvi8QdxOfAOZUsRlbJroiPedQqF0LqZpIpG1CgakZW8pka0rUqTglKq81u/LO6XC1ClF2WJophSl4yGLvmYm71mPyz9PvqLZvFNqe4gmWdWwXV/Vpu8BwM5eDxjGOlklsmpbRCnSvkedUpRmOD4fx91H5nBmJQVFNSAJLPYN+nDHwfoyFyiUdkDOF5XypIBiB5Wi6fBs0KG0UeQyk+6aoVWiVEa1zgnlXhcSfC5vsctMbSZTyrGW0QwTFYx0FAqFAkUrrNnHe9wIuHkkZA0LsSx25wvHlaiVKeX8WSRFRSnK1kKdUpSOR887nJoJOXc+jjyP2qBTartN34vJ1onHK3I1hTmXwGE4H2h+fD6ORFYDwwBDweoTiRrBazulqChFaYzj83F8+f7TODYXR8gtYqLfi5BbxLE56/7j8/Gt3kXKBQpxSrmqHGM5lrHbvrIdkFlYadJdoxCxbaPte5lcZWHP08ZMqUbQGsyoBIoLbCqdOkuhUKpAWvdcAguJ57Cz14rVmIlUb+HLqrp9DK021MjplKLtxJSthIpSlI6HtO81m6ttt+8ZG3VKbY/FI2ndqxZy7oTkSj16NgwAGPBJFTNSmsEn0UwpSuMYhom7j8whks5h36APksAinFLgFi2nVCSdwz1H5mkrH2VLIDlRUg3XkdRBuVKtmr4ntmiqYF3te1s8fU9tQpRiGAYCyZWiE/goFEoVEllrzR5wWWv2ekUp0rrnEbmqjt0erwiGsa5x0lt8PKVc2FBRitLxkPa90qDTeil1StmZUo2KUi26aNjq/JtCnlR9o18n+rwwTRNnV9MIpxSIHNvSffaKtH2P0jhT4TTOrKQwEnSDYRgsxGRMhTNYTSpgGAYjQTdOryQxFU5v9a5SLkBsUaqWGzX/804Qpeyg8w2279nnzBYFnZcTpdwdI0o13r4HFDKoNCqaUyiUKsRl67gccBeLUrORTFVnUz2te4AlqBPBi7bwUbaSjhalvvCFL+Caa66B3+/H4OAgXv/61+PkyZNF25imiU9/+tMYHR2F2+3GTTfdhGeffXaL9pjSDpxB581AnFIkUJQ4nuqtbIotWmADVrvRZ390Ap/6wbP4ix89h0/94Fl89kcnNrXNKJ5v36s1eY8g53QcmYniyekonpqJ4cfHF1u6z14adE5pgmRWg6Ia9sUpuagnGTNukYOiGkhmqdhJ2Xzs9r06nVKd4MSV8/u80fa91mVKVc64IkKVvNXte/kFitBA0DngnMC39e87hULpXIhTKphfs4+G3GAZaw1EiszliOYn71Vr3SMUWvjoBD7K1tHRotQDDzyAD3zgA3j00Udx3333QdM03HrrrUinC5XvL33pS/jv//2/4ytf+Qoef/xxDA8P45ZbbkEymdzCPae0EuJw4prMlOJLRCm14fa9/EXDBjM/OiX/hpzE6jlRHZ+P4x9/dR6JrAaRZ+F38RjwSy3dZzJ9b6uzQSjdhd/FQxJY201BLurJv3JOhyRYf7MUymZTT9C58+ed4JQiAs9GA9dFR8v7RjJKqu2PRyDnDb3lOSg8z+O3f/u38bKXvQw8X/21aN4pRdv3KBRKbUimFFnLCByL0fywodlo5Ra+SIY4pWoXoImbirT8UShbQUeLUj/5yU/wzne+E5dddhmuvPJKfP3rX8fMzAyefPJJAJZL6m/+5m/wJ3/yJ7j99ttx4MAB/N//+3+RyWTwrW99a4v3ntIqiJjUdPteiShlt+/VuYiU8mO7c3rzC+zS/BuXwEI3TPhc/Kbn38Tk+jKlnPu8s8cNgWPBMAwG/FJL95k4paijhdIIE31e7Bv0YTEuwzQN25WRy18IL8ZlTA76MdFXfToNhdIOCk6pWsMkSPve1jtmMq3KlMq7kE1zY+1pheD19a+hJ59FaJitd5lJkoSvfe1r+PCHPwxJqj7Uo5Ap1aAolX+NaNA5hUKpRiK/NiYtdgCwo45cqWi+fa+eAjQRpcK0fY+yhXS0KFVKPG65Mnp7ewEA58+fx9LSEm699VZ7G0mS8NKXvhQPP/zwluwjpfUQMalZpxR5nFE6fa/BTCnTbH6akDP/xjSBZxcSODYfg2YYm55/Y2dK1Wjfc+6zP38ydAksOJZt6T57pUL7ikYX6JQ6YVkGdxwcR69XxMnlFBTNgGGaSCsazqyk0OsVcfvBsabFbAplI9hOqRpDIYgTN6ttrVPKNE279bWWu6sWzsm2GxGMbFFKWO9WEjjWFoK2MleKrE8aCToHAIE4pWimFIVCqQJxSjnX7PWEnZNMqZ4amVJAQZQij6FQtoKu6WswTRMf/ehHceONN+LAgQMAgKWlJQDA0NBQ0bZDQ0OYnp6u+FyKokBRCn2ziUQCAKCqKlS10J9LvnbeR9l8lJwKwzDAwKz4XlR7rwxdg2EYUE3r57JiPR9b5fmcMKYJ0zRgmkBaVsC66stichJNZZHNaZD8ElaSst2qkZJVBNwCJJ5BNqchmspCDVWvzG4E3TARyygwTcArMFV/f+c+SzyPWIZHj0eAkc/AaHafS98rHiZgGjBMIJbO1hTLKJtDNxz/9g96cOdLd+Obj85gISZDN0xwLIOLBr1406Fx7B/0dPT+t4pueK8uNNJZ6zzDMea698f5PgmMCcMwkJZzW/r+KapuFwUExtjwvvCMiZxuIi0rkNjmhBfyGops+XO1xDFQVB2JTBZ+sXXis2maiMfjyGazyOWqX6Rlc9b6Aobe0GvGwHrfs4pKP7cbhB7/ugP6PjVHLK3AMAx4hMJrN+K31uLzkQzkrGI7LwmmaSKSysIwTATE6mt9AAhILAzDwGoyW3QtTN+rzqZb3qd6948xW92M3yY+8IEP4Ec/+hEeeughjI+PAwAefvhh3HDDDVhYWMDIyIi97Xve8x7Mzs7iJz/5Sdnn+vSnP43PfOYz6+7/1re+BY/H055fgNI0syng4WUW/S4TN481/uea1YHvT1kH7DftMfDQMoOFNIOrBwzsC9T3HHefZ6EZwKt3GAjUN7SuiBUZ+NfzLDw8sCIzIMXjPpeJkGjtY1oD3rDbwKC78eevl7QK/PsMC5ax/q9q5jOyz14ecJUpnLdyn78/xSKrA7eOG+hpnyZH2abMJIGfL7DQTIBngN/YaWCgjZ8jCqUWv1yyzjOHBgzsrXKeOR5h8GyUwd6AiUMDW7ccI+cGjgHesGfjjtV/m2Kh6MArxw00W2f53hSLnA68aoeBYJnz7k9mGcRzDF46YmC4hUu3bDaL3/qt3wIA3HXXXXC5XE3vYyV+vsBgVWZw3ZCBnb6N7jGFQtmOGCbwL+dZwAReu8uAJ28lMU3g+9PWMfbmMQP9JYeorGb9HPm1fq3uYkW3jtkAcMduA3U2klAodZHJZPDWt74V8XgcgUDlBVFXOKU++MEP4gc/+AEefPBBW5ACgOHhYQCWY8opSq2srKxzTzn55Cc/iY9+9KP294lEAjt27MCtt95a9GKpqor77rsPt9xyCwSBuje2imfm4ph9ch57B7y47fpdZbep9l7JOR1HfmxNbXzlqy7GymOz4FbTeMnBUbxgR6iufTj+01NIZDW8+KW7MRZq/GrXMEzM/eQkHjsfgQgDUl4N8vlEjPd5cHY1jevGAnjHq/a3td1oKpzG0Yem0ecV8Ruv2FfXPh+fT2B8wAvGoWCZptn0Ppd7r879x1ksJRS88EU7MTlEV+idQDcd/355Zg1zrhX7+6uuHsMV48Et3KPNpZveqwuFhYemwIUzuOnQGC4fs/4Wy71PwbNhJI4v45LxAG67erzaU7Z3f2Myjj5wHgEXj9teedGGn+/5+04jklFx44sn7FaTRjBNE4/88DmYJvCaV15UdmDB4q+mcH4tg0Mt/rw7h+m8/OUvRygUqrjt4//+HFTdxCtfsa/m6HUnq49M4/RKGtdeNYqrdlZ+fkpt6PGvO6DvU+MkZBWP3nsaLAPc/ppLitbascdm8NxSCnsPDOGGvX1Fj5uJZHDEnELQzeO1t9Z3PD9x+HnIqoFrX7IHvW6OvlddQLd8pkhHWi06WpQyTRMf/OAH8b3vfQ+/+MUvsHv37qKf7969G8PDw7jvvvtw1VVXAQByuRweeOAB/OVf/mXF55UkqWx4pSAIZd/USvdTNgmWA8uyEHiu5vtQ7r0yGBZsflwzy/PQTQYsy8ItiXW/r26RRypnwADb9N/CGw/twK/OhJFUNPT5JOQ0A/GshnNrGfT5JLzh0E5IUhM2rAZI5UywLIten1TX7/HGQzuxGD+Nc2sZjATdcIsc5JyOxbi84X12vld+t4iVlArFAP2sdRjdcPwjf9eERM7o+H1uB93wXl0oqAbAsix87vXHWuf75HOJYFkWmsls6Xunmllrf131nxer4ZYEsFkdJtPcOVPO6WAYFgwDBDzSuvYUgLx2WeSM1r52zueq9pkyTRMGGLAsA0+Dr5skCmBZtunXh7IeevzrDuj7VD+ZpAqWZRFw8+vW2rsG/Di5ksFiIrfu9Uzm10T9flfdr/VAwI25qIyEYmIoYD2GvlfdQae/T/XuW0eLUh/4wAfwrW99C9///vfh9/vtDKlgMAi32w2GYfCRj3wEn//85zE5OYnJyUl8/vOfh8fjwVvf+tYt3ntKqyAB5VyTDiJnQLphFILOpQb8qZJQCONuFpfAYe+gHzPhNHwuHlNrGfAcg+v39OGOq8dxYKz9zo7C5L36hKQDY0F86OZJ3H1kDmdWUlhOGJAEFleMh3D7wbGW7bMvP4EvrWz9WHRK9xHLjzH2SRxSim5/T6FsFQ0Hnatbe+yrNumuGcggkWanCmZy1sQpiWfLClJAYUiGrG7N5FbdMEECMGjQOYVCaTWJrLVmD5TJst1VJey8kcl7hF6viLmobIWd0/wDyhbQ0aLU3//93wMAbrrppqL7v/71r+Od73wnAOBjH/sYZFnGnXfeiWg0imuvvRb33nsv/H7/Ju8tpV1ourVoY5udvucQs3TTtEfH1zt9DygIWBsRpX5xchW9XhG3XT6My0YD+O/3nQIDBu+7aS+GApUzK1oJuVgPNRAmfmAsiEtHApgKp5HMavC7eEz0eVvaZujNi1IpZWsuLijdDZkoOdHvxfH5hP09hbJVEDHGJVQ/z5CfNyvetArZFqVasywkE/ianVhLRLJqkwDd+Z9t1fQ9VS8ISnyD50MitFFRikKhVCIhW2viQJk1+1iPGwxjrX/islo0JIhM0WukpZhsG04rNbakUNpDR4tS9WSwMwyDT3/60/j0pz/d/h2ibAnEKdXooo/AMAxYxgoM1A3TXiQ3JUo1Wc2ei2ZwdjUNlgFeMjmAHq+Iy0aDmIvKWE0qmyhKEadUYzZPlmWwZ6B9WU8FpxQVpSiNE82LrbvzolSUjjWmbCGmadrOJ6mKqAJ0kFMq//97auxvvZDza67JQo5M9kesvD+evIC2VaKUlp9EyzCNO7nJekZrUrSjUCjbn4JTav3lusRzGA64sBjPYjaSQdDRuUDWRD0NiFJ9+W0jdP1E2SJovj6l49HzlcSNOHPIgtEwCk6pRuz25MKhWafUg6fWAABXjofskwQRopbi2aaesxmaFaXaDXFKpXNUlKI0RlbVbZfJ7n4vAKtNtUsGy1K2IapughhgarWJd5pTqpoI1AgbFqXq2B/ys8wWFTOIU0pgmaJBIPXA58dhOd1WFAqF4iSRj9wo55QCYA+RmC1p4Ys02b7nfCyFstlQUYrS8egkU6rJ9j2g0PqnNemU2sgCey2l4PhCHADwkosG7PuHiSiV2BxRyjRNxPMnuKC7vYHqjUKyQWj7HqVRSEXQI3IY8ElgGOtCj/4tUbYKkifFMLVFqUJeob6lQioRgVwtEqWkDYpShYyrKu17IsmUaq2gx3Ecbr/9dlx//fXguMr/P3E5Vcq8qgbPkvY96pSiUCjlSWStdUywgii1o0yulGEU1vq9DYhSfV5rAFgknYNB24opW0BHt+9RKADsg2OzQeeAZZVXQBb+1n1iQ06p5jOlHjq9BtME9g/5MBwstOkNB60TwPImiVJZ1bD3v9OcUrR9j9IstvvPLYDnWARcAuKyilhGhb9MOCiF0m6I60ni2ZoOGuKUMkwrf0mqEYzeLlrdvrfRHEYSXl5f+15rzxsulwt33XUXDh8+DJercmt9zhalGl+bCBxp36MXfxQKpTy2U6pM+x5QcErNx2TohgmOZRCTVRimdd0TcNd/me938eBZBpph2kORKJTNhDqlKB2P1sL2PWeLREOilEAW2I1lVySzKp6cjgIodkkBhfa9cDrXdDW5EaKOCWWNTgpqN146fY/SJOTvOpS3nhPBNUon8FG2CJIPVS2kmyByLIhutZUtfHKOiECtqVUScS2nN3dMt51SVV5Du31vizKlSLRAI2sJAg06p1Aotag2fQ8A+n0i3AIHVTexGJcBOFv3hIbailmWQY+9fqKiFGXz6awrUwqlDMQp1WzQOVAQtMjFAs8yDYlczWZKPXw2DM0wsaPXbefdEPwuAT6Jg2lujluqkCfVWa17AODNXwgpmgGVBr9SGiCe/7smiyliV6eLKspWQYoXtVr3AGsQh4svtPBtFfW0yzWCaA8H2Wj7XmWRrNC+tzWtj+RcRVrxGkGgQecUCqUKilbIy6yUKcUwDHb2ugEAsxFLlIo1EXJOoLlSlK2Etu9tIwzDxMKDj8F89FGIg/0Y2DkCdqAf6O0F+voAtxvYQC7TVkEypdgN7DvJoyITfRrJkwKam76XVXU8di4CwJq4V65iMRRwIbWaxnIia/eGt4uYbJ1kKvWmbyUugQXHAroBZBQdQQ/Vyyn1EbXb96zFVDAvTsWoU4qyRZALiXqcUtZ2LGRVb1rAaQXk3NhqUSrXpOiSrWf6Xv71NU3rNW/VvqfTafh81rTZaDSKUChUdjsSUt5M+x5xSqnUKUWhUMqQkC33qsSzVc8lO3o9OLmcwmwkg+v29iHcRMg5odcnAcspKkpRtgQqSm0Tjs/HcfeROVz0zW/hLf/yt+U3kiTgxz8GXvYy6/v77we+/W1LtCK3np7C15OTQH5htpXoLXBKkfY9EubaaPtaMwvsJ6aikFUdAz4Rl40Gym4zHHTh7Gp6U8LO4x06eQ+wqj1eiUdC1pBUVFtYoFBqYbfv5f9myEIsShdVlC2COJ5cdRY/rAsOdUudUva0uxZlSpGWtg0HnVfZH55jIfEsFM1AJqe1TJSqF5IHJTQlSlGnFIVCqUwyWz1PirCzJOycrH2acUr1EadUJodgw4+mUDYGFaW2Acfn4/jy/acRSecwNroTzx66Ca5kDGI8hoCcQCCdAKNpgKIAfn/hgU89BXzta5Wf+Kc/BW691fr6rruAz3ymWMAit74+4Dd+A9i1y9pWlgFNswStFjiz9BYEnXMl7XuNO6Xy7RU1KtmGYWIqnEYsk8MPn5kHAwYvvqi8SwpwTOCLb0L7nlzsKOk0fKIlSm1VPgilOyFTZogo1evNO6VoUCdli2jUKUWcuFuVKaXqhu36aXn73gZFqWpOKcDaX0uU0tHX1P/UPGp+cl4zGY08S4POKRRKZcjaplLrHmFHrwcMY+XTphTNdo/3bah9T6WiFGXToaJUl2MYJu4+ModIOoc9A14cDd6MZ697BVwCB9M0cWYlhSvGgvjTl+4AG40AIyOFB994I/C5zwGRCBCNWv+SWzgM9PcXtp2bA55/vvKO7NtXEKW+9S3g934P4PnyItZHPgJcdVXheU+cKBa4AoEiMUtvYdA5aVGoJ+vDST2ThIhb7cxKCuGUgkhaRb9fxFteuKPiY0jY+eZmSnWmC8mTDztP0Ql8lDpRdQPJ/Mhk4pAK5kXXWEaFaZoNBX1SKK2gkaBz53bZBtrDWwkRgFim8XNjJcjzNOuUIsHrtUQyj8AhBtU+t28mqkYypZpwSrGkfY86pSgUynoS+bVNpZBzgkvgMOCTsJJUMBvJrHOPN0KfI1NqouFHUygbg4pSXc5UOI0zKymMBN2YjchYSSoYCbryyjmDkaAbp1dTmMqx2DMxUfzgF73IutXD294GHDq0XrgiXxNBCrAELsByS62sWLfS5yL89KeWgOWE4woi1Ve+Aj1wEQDAf/IE8N1flhe6/H6gyuKO5FHZTqkGK5u1pu853WojQRfCKQVivq3g7/7jLD508yQOjK2vOwwGJDAMkFJ0pBQNPql9H0mSKdWpopTfnsBHRSlKfRChVeQY21FB/r4VzYCs6i2bJkah1AtxPNUr8Gy1U0p2uJJaJeJuxCllmmbBKSVU//wS0Worzhtkcl4zTimBo04pCoVSmYTtlKq9htnZ68FKUsHZ1ZRdqOttwinV4xXBMNZxW6HRrpRNhq7Wu5xkVoOiGnAHOfgkDitJIJ0rLM7cIoflRMFN0DQjI8Uuq2r80R8Bd95ZEK5KXViXXFLY1usFrriisK0sA7oOrK5aN9O0g86Dx54C/uzPyv6XAoDR//yfgde8xrrjF78A/vIvbdHqhUlgnPUgOD6MjOmC+4VXA9hT969fbfqe0622b9CHuKxC0Uy4BBZXjgdxfi2De47M49KRwDq3l8Rz6PWICKdzWIpnsW+wPRlemsNR0onT9wDAI23dxQWlO4nbQqtoX0wLHAu/i0cyqyGSzlFRirLp2JlSXeKUskPOW5QnBThyGJsQpRTNAMn/rumUyn++t8QpRabvNdO+l38MzZSiUCjlSNiZUrULyTt6PXhiOopjc3EAVqGjmeO5wLEIuARE0wpSNAGBssnQ1XqX43fxkAQWck6325+si3oTAAM5p0MSrIu0TYNhAI/Huo2PV9/2t37LuhFkudiJdeWVMJ63DrK53XssV5VT4CKCVzoN1RnKfuYM8JOf2N9enb8BwEsAPPTZrwCvfaF1xw9/CLzvfVbrYH//ehfWa18Lcc+k9aul0jCmZ6yphvlphk63GsMwWMznQw35XeA5znKrrSQxFU5jz8B60Wko6EI4ncNyon2iVCKrwTStNgPvJofB1ovXbt+jmVKU+ohWaEnt8YhIZjXEMirGe7ZizygXMrZTSqg36Hxj+UsbJWO3yrVunSBx1nkmpzd+PCfOLZ5laoaIE4dkZgvOGxsKOieZUnT6HoVCKQOZvlcrUwoAdvVZYecJh0uqWddrr0fATDiFMxng/Foa+4aCG4pPoVDqhYpSXc5Enxf7Bn04NhfH3gEvWAbQDWtRLPEsFuMyrhgPYaLPu9W7Wh9uNzA2Zt3y6EYMAJC5/kbgLf+p7MPUVAprP/1p4Y6XvhT4+tdt8eq5Z6egrqwhKCfBxyJQR0YL2y4tAQsL1q0cExOQLtoPAJh8+ldg3/5H1v2SBPT1YcQfxCcMF4yeHjzwyrfg18FJMAwwqSew59GHkfIGcEYVkT3vA7y7bTGLMBJw4cRCoq1h5zFHj3mnZuz4aPsepUGiFUYfhzwCZiKF9j4KZTNpNFNK2mqnVJ2h4o1AnFK6YbmBGnETEddTPe2ExEmVaeFrx3EcXv3qV2NlZQUcV/k1sZ1SbDNOKSpKUSiUyjTilBrwSRA5BmupHFTdwFBAgmGYDYtJx+fj+OWZNZxaToJROcz++/OYHPbjjoPjZSNIKJRWQkWpLodlGdxxcBzzURlnV9PgWAaaZmAtmYWsGuj1irj94FhXq9xk0cZWW5xKEkze8ec8OWnd8jz+yBSeW0zabT0v2tNb2PaNb7TyslZXgbW19e2GF10EnmXAMoCoZGHyfGGa4cIC3FjAZfmn+sXBm4HgJEJuAXvOPIm3/9UfFf6fvyjsK3p7gb/+a+DNb8Zw0IWBufPY/d0vA1ftKwS+9/UV3Fr9/dbjmsSevNehrXsA4BVp0DmlMcjfdXCdU8r6PpIXYymUzYQ4nlx8ne17+e2yFTIL2w3Jb2rV5D2gOE8r16Ao1cj+kPOGnGvdecPlcuH73/8+Dh8+DJfLVXE71diIUyofdE7b9ygUSgmmadqZUsE6nFInFhN4bjGBmYgM3TAxE8kgLqsNiUkkG3cpnoXIsRBME0GPgGNzccxH5YrZuBRKq6Ci1DbgwFgQH7p5EncfmcPDZ9eQympYTeVw/d5+3H5wrOsPIkY+U6qZaiSBTN/Llpu+FwoVpgFWgIGV//TUTa/FTZ/9Qwyymt1iaKyu4Z77n0F0ZglP9Vs5Vb1eETmXB1P7XwA+FkUom4I3HS+IWYuLQP73GQq4MDR7Gge/8w/AdyrswFe/Crz3vdbXv/418Md/DPT0lA99v+YaYPdua1vDABgGcdLmVMfJbaugTilKoxAH4HqnlPV9nIpSlC3APs/U2b5HttuyoPM2ZEqx+dY7VTeR0ww0Ug9pxLnlFq3XjghZmwnJg6JB5xQKpZWkFA2GaTVV+GrEr9iDljI5iDwLnmXQ4xEbEpOc2bi7+zw4s5qGrlrrcv+ggDMrqYrZuBRKq6Ci1DbhwFgQl44E8KNjC/jh0UXs6ffiY6+6eFscPEghcQOaFLi8y0rVm5+WIwksZFWHoptA0G9N/Nu1CyyAiw9ci7/88fN4bikBn2Eg4OLx1CXX4vCHr0CvV7ROCqMBIJUqZGHt3AnAGsEaH92JX772HbjGb8CVjK+fbtjrcHbNzAAPPlh5R50C1i9+Adx6K673B3GZJwBxaAAYGyq4sN7wBuD6661tEwkri8s50XATW/1I0PlWXFxQuhPSntdTJlMKKGROUSibSbbRoHN++7XvAdaUW1XXG87KsjOu6nj9SND51ohS+YJZM04pEnRumDBNs2Pb6ikUyuZDBhP5JN4uqpejaNDSgA+nV9IArAiDoLt+Mak4G9e6L6sxOLuawo5e77psXMMwMRVOI5nV4HfxmOjzbovrTcrWQkWpbQTLMrhmog8Pn40gqxmbqSe0FeKUqnZgrkXpwVKsc1S3E8kecb1+8XtgLIjr9/VhOZmFaQIzERmSwOKK8VCxW81fELOc+2Zc+QIc3nUxeq7dub6iYZqW44nwohcB3/1u5emGexxTBcNhQNchxSIYjEWAhSngKcdzX3xxQZR6/HHgFa8o/Izjih1Yf/iHVqsjYDm9vve9Qquh898mxSzilFI0AznNaOo9olw4GIaJOGlLdRfbMIhIFaVOKcoWoJCg8zqPYVsfdN769j3AOs+mc3rDE/hIPlQ9wetESJNbKEql02kMDg5C13UsLS0hFAqV3U41NpAp5ViTaIbZVAsghULZnpC1TaCGS8opJrkczlwXz4JhmJqDlgjOSe4cA/T7RKRTQCStIibH0esRoBkmklkNx+fjuPvIHM6spKCoBiSBxb5BH82domwYKkptMwb9EgSOgaIZWEvlMOBvPoeoU9AN0r7X/KKNKxFJxGacUnY1e/0C2zBMRDMqDu7swc2XDGLQ72qoejAUcGE+lsVyIrv+oM4wlkBEGB8viEO1eN3rgLk5/NOPjiC7vIbXjosY0WVLrAqHgYMHC9tqGjA6at2vKICuWzlbq6vWz9/5zsK2J04AH/hA+f+T54G/+ivgwx+2vj9zBvjsZ21xiw0EMDY9DYbjgMFBq9VwYABS3nasGSbSigaR79z8K8rWk8iqMEyAZbBuuijJmMqqBuSc3vKLbQqlEoZhFjKl6nVK5bdTtsgp1YgzqRFEvjmxLdtQ+157HLaZTKbmNmr+9xL55qfvAZbjqsUvPYVC6WJInlStyXtFYhLLYDTogmoYcOWPi26Rw3LCsJ1XlXBOcve5eOzu90JPhWF6BMRlDUsJBapu4IfPLODcahoJWcVI0A13kIOc02nuFKUlUFFqm8GyljI+E8lgLprZFqJUXUHnNSh1WQlNuHDIAjtXJpj03JplY/WIPF62f7ChUFcAGA5aYapLiRZP4BNFmKOjODsURa5/AsKtFwG+Cn8Tr3wlMD9vfS3Lxe6rSAR4wQsK2waDwO23F7cYhsNANmuJWx5PYdupKeCf/sn+lgNwCLCC3gHgi18EPv5xMAyDiYWzeP1//QN4RoeAwX4r4J2Evvf1ATfcUNgPTbP20+drW5shtSh3LqR1L+QR1r0nEs/BJ3FIKTpicg5u0b0Vu0i5AHEKMK66nVKd0r7X2iUhKeQ07JRqwLlF9llW9aamTW0EzWg+75JjGTCMZYTWDAPWmZFCoVCARF5EqjV5r1RMGu/1FP1czumQBHZd4a4U5yT3fZLlqJJYYMegD2lFx9G5GPwShwdPriImq7hoyAe3yIJjGfhcPPZJPpo7RdkwVJTahoz1WKLUQiyLq3Zu9d5sHMNoQ/teU06pfNW3jFPq6GwMAHD5eKBhQQoAhgOWKLUcb7EoBWuxnstnX9QzxQMA4HYDY2PWrRyHDgF3373+/kzGaif0OWzCe/cCX/qSLV4Zq6sInzmDfpYFE40Cw8P2pn2pKPqW54DlufL/7xe/WBClnnkGuPpqy5lVLvD9jW8EXvMaa9tUCnjkkYbaDKlFubMhrXmlrXuEkEdESpERTVsVPQplMyDt3TzL1H0ukOyCh7npwgpQCDpveaZUlZb3apD2PU89mVKObWRVh1favGWtuoFMKYZhwLNWEDwNO6dQKE4KTqnGxCRnNp1pmliMy7hiPISJPm/V53FOcj+zksJQQIJhWoHrywkFl4wEcP2+PnzjV1NwCRwW4wpWkzns7HOj3+dqqFWQQqkEFaW2IWMh6wJsPlbbft4NtMQpxZQ6KVqXKaXpBo4vxAEAV46Hmtq/obxTai2dg6obTQWxV4I4SvwuvqXPWxaPp9glBVjteX/8x/a3uqri4cOHcdttt0EQikWy1JVX4e//4p9wyxCHfVyu0GZIbpdfXtg4ErH+1TRgZcW6ObnssoIodeoUcOutxT8nYlZfH/DBDwLvf791/9oalr/0Nzi6rCLAe/GigX7oPT2ISD6cjcfx5UgGH3rFRVSY2mJicsEpVY6QR8BcVLYn9FEomwFp73bVOXnP2rYgrGQ1veWOpVq0M1MKaNwpJZN2wjr2h2UZSDwLRTOQyW2uKLWR6XuA5bBSdd3OpqJQKBTAiicAajulSsWkkaAbbtFqqVuMy+j1irj94FhdhQ7nJPfTS0msZgE+o9rZuLph4rB/CSGPgIW4DDln4PxaBn6XAInn6m4VpFAqQUWpbch4jyVKLcSyW1J1bTUk6LyZaiShdM3YVNA5yf0oWWCfXE4iqxoIuK0+7GbwSzw8IodMTsdyIovxHk/tB9UJEaXqdkltIWJvL2YufgEWDgxj30UD1Te++WYgnV7fZkhuN91U2NYwLEGLiFuKUixmJZOFTc9PYei//QXeVuG//d6tb8c9vZ+0LMpLi8D73lfeqdXbC1x0ETAxsdGXhVIGIjb1VJg1TyfwUbYC0oJXb54UYLmABc5yzSiqgQp/0m1Bd2RgtdopJXGVW96rkWmwndArcVA0o6Vh5/Wg6iTovLm1icAxkFVQpxSlYWi0wPYmIefb9+pYtzvFpDMrKSwnjPKDluqATHI/sxzHT36+jFe9/GLsGwqCZRmcW01BElhIPIfLR4N4fjmJhKxhKZ7Frj5v3a2CFEol6F/ONmTAJ0G0w84VDOZbw7oVEnRe6nZqBK4k86GZyqZUIbT1mbmCS6rZsc4Mw2A44MK5tXQbRKl8m1MFR0knQarcaaWOSgvDFJxZ4+NFP7IXbLMxa8F28GqwzzxT2CCTKZ5e6JiGOGfwOPuS1yEkp+DLJMBGIgjICQTlJEQlC6a3t2BRXlwEfvjDyvv4iU8AX/iC9fX581bbYyUB6yUvsYQ2AFBVa/u+PiAUKg66pwAozpQqR4hO4KNsAY2GnBNcAgdV15BtsNVto8iOHCsX3yFOqQbbCT0ij0haRUbdvAq9YZhYTSkIp3JYjGebEgVIoY2KUpRGoNEC2596nVIEIia1QqhkWQa7+73Y5QN29xeeo6hVcNCH0aAbCTmJ1aSC0aCr7lZBCqUSVJTahrAsg5GQG9PhDOZi8rYRpTZSBWqFU6rcAjur6nhuMQEAuHJHqOn9A6wWvnNraSzFlQ09jxPDMHFyOYlwSoGiGh3vnPNK1kVIqh5RqgJ1LdgqiFkAEB2dwN+/5eOY6Pcimsnh3GoaAHD5WAB+U7OcBUndsijv2AH87/9dELdK3VpOl5QzFL4cn/hEQZSanQX277e+ZhhLmCKthn19wBveALz73dbPZdnK9+rpKRa5enqsFsVtStQWpao7pWj7HmUzIU6pRlvEXTyLJMpPd20nxF3kEtiWnxukptv38u2EdQp7ZLtWTeBjWRYveclLEA6HwZYJMSfnmAdPrSKnmYjJp/Gz55YbFgVIQLpG2/codXJ8Po4v338akXSOTj/bpqi6YR/LamVKOWFZpq1ZTutbBV1wiywSsoajc3Hs7vfW3SpIoZRj+16xXOCM91ii1EJMxsGdPVu9OxuiFU6p0jyqjTmlCgvfE4sJqLqJAZ+I0eDGxL+RFk/gIwvnX55eQzyjYjEuYzaa6ehqmq8Rp1QZWrFgc04zce5HStHg9ruQymqQBNOyKA+EgPe8p76dO3AAePbZyu2G119f2DaVAgIBIJGwxjNFo9bt7Fnr585JiAsLwG//dvn/MxgEPvAB4C/+wvo+mQT+y38pDnx33oaGrMd0OKZp1nQA0vY9ylZgi1INOqWkLZrAV5i813o3pljBXVwNVTfsAPF6M67Ivreqfc/tduNnP/sZDh8+DLe7eEiC8xwjcCwknkHILTQlCgjEKWVQpxSlNoZh4u4jc4ikc9g3WAi1ptPPthckk0ngmLqF+c2itFXQNK2iQ8DF486b9nbstQWlO6Ci1DZlNB92PheVt3hPNoZhmIhmcshpBmajGQTd68e/10Pp5L7mgs7zmVKOSvYz+al7V+5ovnWPYE/ga4Eo5Vw48ywDv4tHj0fs+Gqa3b7XxMVF8YLNi6RiiUoBt9DQgs1pUXYKkClFQ7+v/mkm63C5gEsvrW/bK64A4nGrjY+IVsRpFQ5bAheBYYBbbikWuOJWSynicUvUIqysAF/5SuX/973vBb76VevrWAwv/ehHwf3t35YXsA4cAK65xtrWNK2MLmFzWkTTOd2+cA1VyFwgYlUmpyOr6g23U1EozZDNCzCNnmPI9pstSpGWt3aEqzfTvkccAixT/2tIxKtmixn1UioKxGQVpgn4XQJ6vWLDogCZzqg2mLlFuTCZCqftMGsAmI9m0OMV4RF5Ov1sGxGXC617G72uaAfOVsG4rOJfn5yFbgDyJrt8KdsPKkptU8bzotRiTO74lq1KHJ+P4+4n5/DEVBS6YeKLP34e+4f9TTl9SkUpsQWZUilFw+mVFADgiian7jkZDEgArCpJStFs11CjlC6cj87GwDAMQh4RoyGuo6tp5Hdupn2PLNhCHgEnl1NIyBoYBrh8LAiXwNW9YCMW5dlIBkdn03CLPDiWQSSdg6abDU0z2TCCYLmXhoYqb7NnD3DvvcX3aZrlrAqHLccVwecD/uRPCoHvpS2Hvb2FbcNhhM6dA86dK///vu99BVEqFrMeGwoB/f2WiNXfX7jdeCPw+tdb2xoG8PTTBXHL77eEtQYgLqmAi7cv7EpxCRzcAgdZ1RGXVSpKUTYFpYmgc+f2jbiKWkEm19z+1gM5zyoNiC5O51a9F2REUJPbLOiVigJE72dZNCUKkID0aplSNNCaQkhmNSiqAXeQw2pSwXwsi0RWwyUj1jmeTj/bHiSIKNVA695m42wVzKoGfnB0AQ+dWcW1u3vp8YnSNJ37F0/ZEP0+yR6TvJpSMNRluVLE6RNOKxB5FjzLIORpziIPFLf+8SzT1EGTjPgm7pljc3EYptUqOeCXGn6+UiSeQ69XQCStYimexb7B5ipdpQvnXH7BK/Fsx1fT6gk6r7RIX05kMR+V8/qG9f6aJpDKWoJEIwu2A2NBvPnQDsxEMkhlNcg5ExzL4Lo9fXjjoR0d6TIrgueBgQHr5mRoCPjc5yo/zumqGhrCo3/6pzi0dy/4RGJ9u+GVVxa2DYetf2Mx63bmTPHzptMFUSoWA66+uvAzjit2YL3udcDHP279zDCszK6SdsOEzAKmWTFPitDjESDHdUQzua47BlK6E+KUcjWaKbUN2/fsibUN/E6ZnHV8djfg3CL73qpMqXQ6jYmJCeRyOUxPTyMUCgEoFgVI4UTgGFtcalQUsEWpCplSNNCa4sQZLUD+xtKKBtM0wTAMnX5WJ50u9JL3tt6Q863m6l09uP+5ZUTSKo4vxFtSpKdcmNAj1zaFZRmMhlw4v5bBXFTuqgsyp9NnT78XsYx1gPa7BARcQlNOH6dTqpk8KQAQ8xPQSCvCM3MxANbUvVYxHHAhklaxnGhelHIunEnlmGMLk346uZpGLi5U3YSi6XbLJKHcIn3PgBe7+7w4OhdHJqdD5FkMBySwLLCazCGpaOj3o+EFm0vkcHBnD4aCLsxHM5BzBt72op24aChQ+8HditOZ4PNh+dAhmLfdVrstb88eYHUVWFsrvoXD1v3OvKx0Ghgbs4QtWQZ03dpmddX6uVPsikaB979/3X93KYDPcTzOv+YNwE3ftu7UdeBd7yoSsF4QMeBVReS4FeCFlwIjI829LhRKnWSbdkqR9r2tCTpvS6ZU/lyba8AplWlif9wtFqUAYG1tbd19TlEgns+qC7oFkCJIo+eYQvveeqcUDbSmlOKMFkgp1t+fYQJpRYdX4uj0szroBqHXbt+rEE3QaYg8i+v29uFnz63gwVOruHws2JFth5TOh4pS25ixkAfn1zKYj8m4elf3hJ0XOX0cBzYWAJp0+jjFq2Ym7wGAJBTa96LpHKbCGas9bLx1J7KhgAsnFpNYijefK+VcOBePlW1u4byZSDwLgWOg6ibSSrEotX6RzmIhJuOnx5fBsQwuGw1grMeFnGZiz4AXCVm1RKmsVUlsdME2E8mAYRhctSOEkYALR+fimI9mt7co1SwsW2jVq8WOHcDcnPW1LBfaDEle1s6dhW113XJYlbYbKgo4XYPocjgU43Hgm98s+q9enL8BAN78ZuCuuwrPe/nl6ycWktullwIve1nhiaJRqw2Soy2AlOo0275nZxZqm+uUyqiNTbprhGYypYio14goVQg6b2+hpShvML+fwfyFYzPnGDvovESUKm3BV3UDsUwOfT6RBlpfwJBogfNraUyF0/DkowVWk1ksxjc5WqAL6Raht3jd3h1ct7cPD55axXwsi7OrKewb9G/1LlG6kM67KqW0jLEeEobYXWHnRU6ffOWTZxlboGrG6eNs3xO55k7Yzul7z8xbQdK7+7z2orQVDLdgAp9z4azlK9QbWThvJgzDwCPyiMsq0oqGXq/VnlUaYp5SdDy/lERa0eEROWRUHS6exadecym+8h9ncWYlhcGABMM0kcxqOLmcxIBPamjBNhuxPjc7ez0QORZH5+KYiWTa9rtfkLjd1m10tPzPBweB731v3d3f+sXzmD49h1uuGIMtYQkC8Jd/WRQMH59fRmZxBUE5CY/TJRWLAc89V3m/3vzmgiil65ZQxTBWZlapgHXjjcCddxYee9991gRD8vNgkIpZFxDE6dRo0LndHr7JTqms7Uxq/XJQ2kDQeSOinkfgix7bLogoMB1O45m8KOCTeKSyGhbjcsOiAM/mnVIl7XvOwhzDMDi3mkYiq4FlGPR4xY5uwae0lwNjQbz2ihEsxGQkZA26YYJjGLzkogHcfnCsI0SVTqRU6AUsB2cnTi7shkypUjwij0MTvXj4bBgPnFqjohSlKbrnL57SMGMk7DzeXWHnRRZ5OR9o7BB+mnH6ONv3Gh3VbT8uX8nWDeDIdBSANXWvlZAJfKtJxc4JaBSycJ4JZ/DMXAxukYfP1fzCebPxSRxmIxk8OR2FqhuY6PNai/TlFDwihxOLlhgFWG2JI0EPfBKPqKzC5xKKxtXmNB26Aezs8eA9L9lT94ItLquIyyoYxhJ3ScvnTCTT9PtCaR1hnUOibwi+XWOFO/1+4GMfK9pubiGOf350BuM9bnzgZfsKP/D5gP/4j/U5WcSJdd11hW1jMetf07R+Ho0CZ88Wfm6aBVFK14Fbby3eWaeY9epXA3/7t4Wffe5z1r6QvCznv6EQFbO6EOJ0ajboPLvZTik7w6k5B3E1SoeD1Lc/ndG+V4kDY0G8+sAI5qIycpqBuagMSWBxxXioYVGAtNTrJU4pZ2HOMAwk8/lVcTmHHq/Y0S34lPbDsSwO7uyBV+IxG8kg5BHwJ7ddDK7JaIoLgVKhdyaSxlJcwf4hH4KezhJ6u9EpBQA37uvHo+fCOLOSwkJMtqfAUyj1QkWpbUy/T7TDzpeTWTv4utNxOn1IhTWUrxg06/RhHSKC0KRTSmAZJGQVqm4gIasIeXgcGGttK1e/TwLPMlA0A5F0Dn2+5gLUD4wF8boXjGI2moGs6liMZZteOG8mx+fj+OXpNUyHMzi/lkKfT8Kefi+8Eo+zqym4BGsiE8sAA34JoyFLMNINE6tJBcmshit3hOxxtf9+dBEnl5O45bKhhn7n2bwjasjvgsRzGAm6wLMMMjkd4XQO/U2+L5TWEM3nufTUDDq3fk6m9dlIEnDTTfX9Z319gKKsn1RIWg73OcSudBq46qpCu2E6XSxmOXNqNA34sz+r/P/eeivw058Wvr/jDsDlKt9uSLOyOgbilCLOp3ohAs5mB50X2vdavxxspn1PVi2hpRFRyisVJhfqhrlu2m6r0Q0TB3f24NLRAK4cDzUdlkycUqVB587CnGEa9vwJIkJ1cgs+pf1MhdNgGAavvXIU//LELAwTiGcLznLKepxCL2AikrLWBOF0DkFP5wi9pmkiIeeDzrskU4rQ4xVxxXgQT8/G8eCpVfzWC3fWfhCF4oCe0bYxDMNgLOTGubU0FmJy14hSTqfP0bkYPCIPryRsyOnjXKSKTVSTSDjikZkoNN1a9E70e3BuNd1SgYdlGQz6JSzEs1hKZJsWpQArAPPgzh7sG/Thmonejpwy4oT0+y8nshB5FiGPiKyq4yfHl2DCuhCQBA5jIReGAq6iwPrSRToZV/vii/qxmMjarXj1Mhe1RKmdfdZnhudYjIbcmIlkMBvJUFFqC8mquh3gX6t1lohSKUVHTjOazpODKFqTC4eGqm8XCABHjhS+z+WKxayAQ8RWVcth5czTIv8mk5bYRNA04J57Kv633K23FrcQ7thhTWAszczq6wMuuwx461sL287MWK4sv7846J7SFBt2Sm1R0Lm7HUHnPBFdzLrFIuJ2akQkc/EcGMbSfzM5Df42OgwMw8TplRQYhsGLJwewu7/5NnjilCoNOncW5pzipqwaUFS9o1vwKe1FzulYTigAgH2DPoyG3JiLypgOp6koVQWn0MtzjD2V2goVNztG6JVVHZph7VugC0XnF08O4OnZOJ6Zi2H/sB8sw3T8tQelc+i+v3hKQ4z1WKLUXFTG1bu2em/q58BYEK+90nL6ZFUDC7HmLfJAyfS9Bi9MneGIboEDBGuRHc+o+PL9p1sejjgUdGEhnsVyIovLRpt7XtM0cWo5CYZhcONkPy4e7uxwbme//1jIjaWEYi+8vBKPdE6Dz8WhxyNgLOQuap+r5p7blf9+PpaBqht1T14kItaOHo99385eD2YiGcxEMrhqZ/cMDthukMk0boGreeHvEljbLRqTcxj0b/IU0mpiltsN/N3flX+cqlruLIJpAl//evlWw0gE5sUXFz+WBMlPTa1/7le+sliUuvxyIJGwMrlI+yC5XXMN8Kd/Wtj2hz8suLXItlTMKqLpTKmtCjrfhOl7gOWWqkf4amYaIMsycPHWtFk5p29YlGJZFldffTXi8ThYtvh9nI/JyOR0SDyLnb2eCs9QH3bQeYlTihTm5qMyTiwkwLIMOJaBbph4bjGBHb2ejm7Bp7SP6UgaADDgE+GTeOzq82AuKtN1SQ2cQq9T7LGG6mhYjGc7Qugl6xuPyNnTObuJ0ZAbPonDA6dWcWQmhoBL6MgJh1uNYZiYCqeRzGpUtHNARaltDsmVmuuysHOgYJG/aMiPq3f1bOiD26xTqjQc8Zn5OBTVgEtgcPlYAOfWMi0PRyS5UktxpcaWlVlL5RDNqOBZZkOV3M3C2e+fdkxQ8kpW65zIspiLyXCLnL2dW7SC8Ku553o8AgIuHomshrmoXNdrYRgm5mN5Ucpx0bGjt3s/S9uJaL4Vr8dT+8KTYRj0eEQsJbKIZdTNF6WaRRCsm/P7d76z4uaGqgKHD1vfcBxw6tT6vCzixNq/v/BATbNugCVmLS1ZN0KupO3x7W+3BCwnPG+JUzfdBHznO4X7P/tZaypjuXbDvr5i19g2QdMNu8rduFOKtO9tnlPKNE3bddgOpxTPseBYK4exUVGq0f3xiJYo1YpcKbfbjUceeQSHDx+G213sMD+9kgRguVQ22iZoB52XOKUAqzD3/pv24hN3P4O4rMElsMhpBib6PR0zJYyy+UytWS5uUnCb6PPiV2fCmAnTISzVcAq9J5eTME3YQu/zS0lM9Hk7Qugl7YOtHKC0mRyfj+PEQgKRdA4+icelIwGoutlxEw63EtJ5c2YlBUU1qGjngIpS2xwygW8pnt2UrIVWYZoFi/yNk/24aGhjkxyKpu81UMEuDUckz9PjEcFxXFvCEVsxge/0srVw3tXnsQPaOxlnv79btDKifC4+X9GyFg4Cx+INB8fxzHwcZ1ZSWE4YNd1zDMNgV58Xx+bjmAqn6xKllpNZKJoBiWcx4GjTI1XxhZi8sVYwyoaIpq1KYqgOUQoAerwClhJZRNO52htvB1gWmJysb1uet3KvMhlLtFpbK243HB4ubGsYhbwsInIpiiVqraysF6v+23+z2hDL8cIXAo89Vvj+jjssAcwZ9E6+Hh8Hrr++sK1pdqwzK+vITmrUKSXZ7Xv6pg1TULRCXpG7yQEgtRA5DrKhQ9F1ALU/s4WMq8b2xy1yQLr9YeenllMAsOE1CeAIOjfWi1KA1XJ01c4eMAxw7e4+3HdiGbv7PRf8hcuFzFTYckpN9FvrEVI4W0xkkVX1hsXwC4kDY0F86OZ9+MTdx7CWytlC70jQ1TFiiT15rwtb90gRX8m/ppmcgXBawWjI03ETDrcKZ+fNSNBtT5mnop1F9/3VUxqizyvCJbDIqgaWE9mumYawmlQQl1UIXGucPk4HfiNOqeJwxEI1dsBviRXtCEccCrhgmibOrabw5HQEPR6xYYfYqbwo1YqF82bg7Pf3ufh1f6ek3//KHSG87gVjDdled/V5cGw+jum1NLC/4mY2pHVvvMdd9LxBd8F1tRCTMdEFDrTtCJnIGaoRck4g25FwdEoZPB7rtmNH5W1YFvjFL4rvk+WCSMU7lhOmCbzvfZWnGzrzsgDgvvsqC1jXXgs8+mjh+8lJIB4v78C66CLgAx8obHvsWKHdcBOmGZKQcolnGxaViIhlmJZzRuTbv2gnAo7IMXW3NjeKyLOQVb3usPNm2vec25Og9A2j6xATCetvzesF3G7IOd0egjE5uPEilB10rpd/bc6uWoHWV+0M4VUHhvHrqQjCaWsybLc6KSjNo+oG5vNObeKUCroF9HgERDMq5qIy9rXg73I70+9z4cBYEFlVx2uuHMX3jsyj1ytg/3BnrJVJ+163hZwDxUX8nK7jzEoaSwkFQwEXOJbtqAmHW0Fp5w1ZI/hcPBXt8lBRaptDws7PrqYx30UjOkk1cqLP25LFclH7XgMV7FKxZFefB6MhF1z5ENZ2hCPORtI4Nh9HOJXDbFRGwMU3ZO1UdQPn1qxqWreIUs5+/32Sr2pmFAkxr5ddfVYlcSYi1+VAmMlfdOwoyQthGAY7ej14diGBmUiGilJbRL2T9wikzW/dBD7KxnG7LSfT+Hjx/QwDfOlLlR+nlQgH3/hG+cD3SMQKZneysmIJWM5JhoRrry0WpV7zGivInexTKFRwYF1xBfB//k9h229/23KDlQpdDYhZRJRqxq1gCVmWnpfV9E1xYmbyrdJusX1LQamBCXyabkDJb+dpcJ+8+e1b4ZTKHDuGSw8eBDQNJwAIHg+QTuPsagqGCbz7r/8zen7vcautVhSLb4IAPP54wc33xS9arkDSkku2EUXsyurgbnlPoX3vRz8Cnn/e/rl0LoqrsiYOXTQE10wPdnj3YyZt4PxaGi/I5Z2Njucr2p/e3raLsJTNZT4qQzNM+F08+hyh5jt7PYhm4piJpKkoVYNzq1YHxhXjIdx88SAen4ogIWuYDqexb3Dr18uJLHFKdZ8o5Szi+xgObkGGrBqYDmewZ8DXtgmH3ZLP5BTtVN3ATCSDgEvAYEACwzAXvGgHUFHqgmC8Jy9KRWVcM7HVe1MfJLehVaKKs32vEZGrVCzhWBZcvrpZLWC7WSxr5xkkZBUiz6LfK8Ij8g1ZO6fDaai6iYCbx1CgO6bEOfv9G8mMqofRoBtSvlq/krSqNtUglfByIbZElJqN0vyGrSKWabB9jzqlOg++ZOlx++31P/a55wqOq3C4uN1wZKR4W78f8PmAVMpSe6JR6wasbwH8xCcKAlYpV18NPPFE4fv/8l8sYczZbtjbC5Nxo2dZhrBvb/2/D8juMJB4y9WsqAawCfFnRERrR8g5gYhrSh2iFMm3YhgUTZ2rB1f+d9iwKDUzA/O22zCdF05NwM53Iw7kkJop/B2VwvPFf1uPPgp8//tlNx0CgFveUwg6//a3gf/3/+yf31Sy/eQvn8VMGji/lsIL/v4vi0XVMr+H7Xz8z/8Z+Nu/XS9gka9//GNg925r2699DfjWt9YLXWT7T30K2Jkf9f7AA8DPf14shjm/fvWrgcFBa9vpaeD06fL/vygCo6OWyA1YorWuWz9n2y/OdgukdW9Xn6eouLazz4Ojc3FM01ypmpzNF2z3DHjBMAwmB/14cjqK08upjhCliGDTjU6p0iL+7n4fnltKYC2VQ8ijQOS4lhfxuymfySnaTYfTiKRVRNIqwukcdvd72ybadRNUlLoAIO4oEt7c6ai6VQkEgMmh1qjFbJNOqXaKJaU4rZ3jPW6sJHNQNAODgcasnSeXLJfZ5KB/U3JJWoXV7z9pn2DqyYyqB5ZlbGF2ai1dVZTKqjpWU1bAfKlTCigIVcRNRdl8iOOpXlEqRJ1S24uxMetWD8ePW//mcsVCViRSuAAmvOxl1tRCZ7shaSkUSv7W/vmfgdnZdf/dDgDvGt2Ff/1/9xfufO1rrfD43l5woRCuSCbBPvYYMDBg/R5vepO9aVBJQ9V5WyxqN0TAaVeeFNCYU0p27E+j5y4vad/biCi1tAS84hWF6ZUA1KUlwOvNT7S1zq2J//U1DEiGNRwglyv8m8utdwF+4APAbbcVb5f/dy2agsFxBafUjTdaIkwuh1RKxuxSDC5Tx0RQBKMo2DHaA6ys4dxq2hJDd+1avw+qau2D829WUQr7Vw7n9L/Tpy2hqRIf/nBBlHrwQeC//tfK2z78cEGUuuce4KMfrbztvfcCt9xiff2P/wi8973W1xy3XiD7+tetKaIAcPgwuD/7M7w4nQb3//1/gCQVb/vhDwM33GBt+8wzliuz1N1Gbi9/OUAmma6sAI88Ut6JJgjWZ7evz9pWVa1jhVNsa4OYRkSnSlOGZyKZTcuj60YMw8T5VevaYm/eiTI56LNEqZUUXr2VO5cnniHte913eV5axPe5eIwEXViIZXF+NQ2vZGXktbaI3z35TES0S2ZVRPIZpyxjiVXH5+Po8YgQeUu06xb3V6vpvr96SsOM58faL8Wz0HSj48eMTq0VnD6D/tY4fZoNOgfaJ5aU4rR2yqoGJHN2e0Uj1s6Cy6z77J8HxoK4dCTQ8oPxrj4vzq6mMR3J4No9fRW3m4tmYJpAr1eAT1p/eBwNucAyQELWEM+oCNYpjHQS3Xyy03QDiXwVqdFMqURW64rjH6UNiCIwNGTdKvGNb6y/T1WBWMy6qHfy8Y8DCwvr2g2V1TXEe4eLXT5Hj9oCFgtgN2A5UwArA8shSr3t47+DgXPPw/B4gYF+oN9x273bmmhIeOopy5FD2gy93oYD4DNNTrprBKkJp1Qzzi3yOzintzaEaQK/+ZuWKLNzZ8E1J0mAz4fVRNbOudx5YB9Q73GECC1liK0kYT40VQg6f9/7rBuAXz27hF+cXMVVO0PYfchyPO1SdTBH17CWyiH+qc8i+MUvVv5dnHzuc8DHPlZewMrlikXet7wFuPLK4m2cXzvdiIcOWaIbEbyc4lwuV5wd19sLXH55+X1QVet1JjjFM123blnH4Ben8LeyAvbIEfQCwMmT61+LN7yh8PXJk8Bf/3X51wywxDAiSj35JPD611fe9stfBj74QevrRx4BXvrS4p9zXEGk+q//1RLHAODZZ4E3vrFy2+Vb32rdAGBxEfjMZwBBgCkI2DeTwDjD48ATg0DAA7zoRcBNN2Ek4IJXlTH5wP2Irx1BKOgtL6IRN5yuW8ckp7uNbMtxHTtIYqMsJrKQVR0Sz9qTyffm2x0X41kksyr8W9w2183te+WK+MMBF1aSCsIpBWCA2w+OtryI3y35TES0e+j0mj3MaXLQh/NracQyKs6tpbCz14Nzayl889HprnB/tRoqSl0A9HgEuAUroHs5qdgH42Zp90WtPd2mhU4fjnOIUk1clLZLLHHitHaSXzulaNAMAzzL1mXtjGdULCcUMAy6Nlug0cyoepjI50pN5+3vlSAh5zt61rukAEDiOQwHXFiIZzEbzSDo6a4TRDdZncsRy4eAChxjOyNq4RU5iByDnG4iJqvo93VHSyulAxAEYGDAOuetpgrH/vffWfbY/+TZNfzw6CKucDqP/uVfgNVVIBKBvrqKM489hn39/eBisYKLJI8rbU0wZDNpYDpttTwR9u8vFqXe+U7L+eHcVyJQ7dsH/OAHhZ/98z8X2g0dLYe5qA6YZludUmIDTikikjWTyUUyqJp2SjEM8Fd/BbznPcB3vmNljjlodc4lADsKoFzQOXGL7x0ouApcAoexkBtzUdnKldoRKv/EpeumYNC61cOVV1q3enj1q61bCUVrxNWUtVb6nd8Bfud36nve974X+O3fLha7nOIYEVcA4BWvgPa97+HIY4/h4OWXgzfN4sccPFjYdv9+S1B2CmfOm/N5/X5L9CkV5sj3fkerVzkHmq5bgyBk2fqakExaLciVeOELC1+vrQH/638BABgAN5Ru+7GPATfdBJZlMIkM3vzl/1L5ee+8E/i7v7O+DoeLf9dSfvd3gX/4B+vrTAbmRRdB4wToggBWFCC4XWCIiPWKVwB//ufWtqYJvO1t5Z1logjmoousjD7C//7f1t9qOYFucNBqmyY8/3xB6Cu3fR35aedWrc/w7v7C2t0n8RgLuTAfy+LMSgpX7eyp+TztQtMNpBTrb6Ub2/eA9UV8JWEg5BbAAtjZ50Vcbk1rmrOIbwI4t5KCwDPY2evt2HwmlmVw+1Vj+I/nV5BSNIyErAD4sZA1RMMr8pB4Dp/6wQm4eBYXDwfgDfId7f5qNVSUugBgGAZjPW6cWUlhPipvSJTajIta4vRpVeseADCmNWpV1Q2sJLMwjMbV83aIJU6K+7E5eEQOmZyO5XgWYz2eukLVyWu3o8fTcFDsdmZHrwcMA0TSKhJZtWIVimRFlWvdI+zs82AhnsVMJNNVJ4duszqXw86Tcgt1C9YMwyDkEbGSVBDL5KgoRWmIRs55imqJC0VOqWuvtb80VBXPHz6MPbfdBq60JRDAPd99ANPnl/Cbu1y43K1bF6Xk5ik5JvX3A8PDlkuLXCgvL1s3o0Tk+NKXrEmEJbwEwJW9g3j4F08X7vyTPwHm58tPNxwYAK66qq7XjWCLUnptsSjT5OQ952M2lCl1ww3W6+R05eRpx0RbIV8sU41iZ5OiFab87e4vXnPs6fdiLirj3Gqqsii1hbRkjSgI9Yto4+Mwh4awyDAwb7ttfautkyuuWCc2VuTGGy0HVD3cfPN60crpBOtzuLMvucRqjyzdnvzr/HwNDFhOKVXFwmoC5+aj6BGBy/rd1raHDtmbDvcHcOrK6xHiTQy62PXPOzxceF5VtdqXc7liwYzgaD08cX4Vl87PQwBQ9pV1DrnQNCsXrQLsa14D/N7vFe74gz+w9qUcN98M/Oxnhe+vu85yrZbjRS8qfq8uv9wqBJRknV2SM+Eb2oX0P3yjsO0HPoA3nJ7BkqwjFPQAIz2Fxw0PW1mDhG9+0zrelhPF/H7g1lsL2544YR1HKuW49awXv1KKJdhwLOouunUi5Yr4S4ksfnh0ET85voR9g76a2a61cBbx56IZhPPtcL0eET6X0LH5TAzDYP9wAPPRDBgwmFpLQxJYXLe3H688MIQv/+w0FFWHxLGYCqcx0edFwC10rPur1dCr1guEsVBelIplAPRW3K6aC2ozLmrb4fQ5Ph/Hd5+YxZPTUeiGiUg6h1+eXus4d0hpP/ZYyI3TKyksJbIYCkh1haqftBfOnVEZ6BRcguVwWoxnMRMuLyaZpmlfCFRySgGWYPXouUhX5Up1o9W5HHGZ5EnV17pH6PEIWEkqNOyc0hCNnvNIFpTEN3dBIblEyP4gEuMjwL7+6hvfn8+tMk0gkynkYEWj69u3XvUqYO/eQrshuakqFLe3uH3v3/+92IHlZHDQEr0Id9xhuT7KCViDg8Dv/z5E4mBYXgGGJSAQqJi3I29AlCJur4ZEqWwWeNe7LPfMC15g3VfGcZHTCjmXrTy3EsdVqVNqJpyBYVp5eD0lLeK7B7x48PSavT+dxHYofDQFw1gB9zy/XjwuJRi0I/zRUwAAWetJREFU8uvqYXjYdiE99PgsnpqN4eaLB3HZpevbkIcvm8TX//yrGPCJ+Oit+6s/79gYkMnkHaBJpBIZ+HkTu/wiWK3QSnl8Po6v/HoRgU98DUMuDm7WhC5nkUikEeKB114ygJ1XTBa/Dn/91xVbNA3SGkl43evKC3O5nNXaXPq6ERecohQL76VC5MqKdSuhH4Ci6mAc7kP85CcYPncOw+u2huWsc4pS/+2/lRX37dfUkUWH3/1da8hBOUKh4kEJr3418MADCAgi/gQsTEEA8xG3JV55PMDTTxe2/bM/A3796/JClyAA//N/Fo5h//qvRdM81/1L2kgB6zi+vFz4GQD/7KzVzuz1WuIjeV5Vtd7rKq2epUX83f1enFpK4uRyCt95fBZ33rR3Q1EKpIi/kpSxFC+01y8msph0CW2ZjN4KHjkbRq9XxOuvGsX+IX/RtfZUOA2WZXDZaAAryRyyqoHnl5IY73FjNOTuSPdXq+msd4vSNsZ78mHn0cph59UqXJeOBDblopY4fcZC7pY4fcgiKZxWIPIseJZByCN05CJpfT+2C5LAIJXV8dRsHHv6vVVD1Q3DxJmVQsg5pZhdfR4sxrOYCqfLvufRjIqUooNjgZFQ5SoOEawWYnLXZBQ5rc4MA8znK0uTgz64Rb5rTnbRtCUq9Xgbs7b35MdnR9M07JxSH6VCrmmaYNnq57ysRtrPmjsmkLa1hoLOGca6aPB6C9PWSvnSl9bfZ5r49s9P4Nz0Km5xnms/+UkrT8mRlYVo1PrX6fgAgFOnKrci5UUpkil1xR/9PvDko5Yg1dNTLGANDQFf/7qdoTh27Alg7dni6YbBYNXwaI8ddF5nZVxVrTyvH/4Q+NWvrIuv/MU4wzC45JJLkEpZ4+OnwmlohomgW8BAi3IuAYDL/93YQed5zuUFp9393nWO0Ik+LxgGVq5UB+UabpfCR6dyPh89MNFfvii5o9da46+mckgrGrxlMjGdVHW09QTt93MtayB06BBWAfs9NU0TT62kEB4N4U+vvQT2p5LngY98pOL/aaoqcPhw4Y5/+Zd6fnWLqani73W94EgrFeEfeshqm3QIXStrCfzoyAwYrw+/E3Ss7z77WejhCO57egamquLGXUH4OViPKz3e3XYbcNll5TPUBgaKtyWDLMo56EpFtEwGkGWwsgx79RXO/1sqcj75pDUUoBJf/Wrh6+9+t/pr/LrXFUSpL32pKFdRAPBy57ZLS4VMxj/8Q6sVlLRelt4efLAwDOHv/g646y4wgoC38wLOxRTkWB6rPV6MDASBz3++4La7777K0zwFwcr8y7/OE9FF3HLuCTy+kMIVLgkerxtRFdAEHoF0CNN8Dy7eM2wV8RXFupHnqaPVsx0sJ7I4t5YGywDX7+lfd+wm7q+Jfi+GAi7MRWWsJBW7s6NT3V+thIpSFwikZW8pUT7svFaF67bLh3FsPg5JYLEQk6FoBkzTCn5u5UXt6byo0gqLvHORNDngQywTAwD4XQL6fVJHLpJK+7EZMMhpBnwSh/e+dG9VAW02mkFWNeAWOFuEpBSY6PPi0XORimOTiUtqJOiumhnS7xPtjLalRNYeJNDJOK3Oi/Es5mNWe8p0OIOLR/xdc7IjmVIhd2NOKeKsilGnFKVOnEJuNJ3D2bU0hgOufCtw+cyKbL59r1mnFBGzyPO0FYZBUnAhFeordib91m/V/xzf+Y5VXSfClXNyYX66IWnfY+R8QcwwCk4tQv5ihwSdX/bVvwIeL2mdYhhLzBoZKUxVBKzsm7k5+AMhvGBahuwLQutPgB/oL4hapdV8XQfe/nZLkHK5gH/6p6KgbY/Hg6NHj+Lw4cPweDw4Nb0KwHJJtXKymZAX2fSS9r1z9oSw9QKEM1fq3NrWZuA4KS18zEYyYBirINqpGS/dQiyTQyyjgmUK4lMpHtEaDLSSVDATyeCSkUDF56u13v/9l+xBIqvi1+cjYAEcnYshp5nY3e/BgN/VGe8nx1k3V5kC4uTkuruePbmCU9iFy0YDxZ/ht74VHIClX53HyeUUfJcP48WTA+seDwCoNFigHM5MPyemuX5C57/+K5BO48nTy3jw2QXs75Nw2/5+S8gqbcX+4z+2hhGUc5dpWvGx7hWvsI6Z5YYQqGpBkAIsV94ll9g/M3M55NJpiACYUiGNtFyaZkHwceLch7NnLZEQluBQ4n/D7Ps+jIjpg9/FY/cDD4Cp9hpffbUtSjH33I13fP6TeEeFTf/qk1/Fq95wtXVt93/+T2EoAWAVN5yC1113Wa8VAPzbvwGf+lR5F5ooWjlu119vbfvEE9Yk0NJhBeTrV78auPRSa9v5ecx88/u4LJzF+FAQwQfni7fdswd+lxuSwEJNZTCophDgeYwPecC5CnmJnej+aiXb9zejFBHyCHZGUemFdGmFK5PTsRiTIWs6sqqOp2djeH4pATmnw+8qznKJyyr2D7fmotbp9GmFRb5okcQyYBjrGMqxTGecVCvg7MdOyCq+99Q8FM2oOdKeBLFODvk6RmTrJHb1FRxOiqavu3Ak7XjV8qQAq1q4s9eNk8spzEQyXSFKEavzfDSDpYS1gGAYayJdNK1C4NiuONmRz0CoQXcAaYGJ1vgMUSgEIuQyHsulYJqwW6lFnit7zlPUjTmlyDGpIafUBthIsDgAa8FNFt0VIKLUj7/2fbz9BUPrxatIZN3+KHv2AVq24NZKpayTdySy3mXwzW8CDz4IEcCbyX2fz/8rCMUXTJ/4hNWamExaF0qCANxzD3DTTVV/h1NLrc+TAgA+nymlGSYMwwTLMlA0HXPR8nlSBJIrdX4t3TGilLPwEcuoWIxbhQ/DMLGzz9M1hY9OhBTSRkPuqoL3rj4PVpIKpsPpiqJUOUdbMqvm84xMPDMfx5987xh29nqwklDgd/H2mn8+JqPfJ4FhmK57P4nQu6eM0AsAk0N+nFxO4fRyqrIo1QoYZv0xbGAAGBjAakrCStKDyX19wBWj5R9fb+snAPz+79e/7Re+YN3yaKqKnxw+jNtuuw1C6f7+zd9Y25YTu3K54im373qXldXnENCOnFnBqbkI0ikZj/46jPgzOUgCi1fx43jle+5Ev8isf25VLZrmeZbxwLX3MkDNQTB0QMmB0VSwqgrB1PEb1+7BJaSIXzqIwDCs1m2SHeh02q2sVG5fB6wBI4STJ612yUoMD9vnR+Xhx3DNJz+Aaypt+9WvYuI9v2/F1vzsZ/jklz8CAPjuBz+Hp276TzBNs64Il26ns69AKC2DYRiMBl04MhPDAydXcd3ePjsviog3PR4BZ1ZTdosMwSVwyCgaeI6BS2QRdAmQBA7RdA4pRcfzSwmMhdwbvqidj8nI5HS4BLZqpk+9OBdJAIPxHjc03bRdMJ18UnX2Y7Msg2//eha/OhPGDfv6K15AnKJ5UlUJeUQE3QLisorZiLwus4yEnO+sIUoBlnB1cjmFuYgM7G3L7raUiT4ver0ifnVmDQGXgLFQvo0vlsV0JA2vyOGK8Z6OP9lFbVGqQadU3llFM6Uo9eJ38RB5FieXktAN2EWN+ZiM3f2+slXLrLYxp5REnFLa5ohSG8lwqpei6Xsul+V0Ghkpuy0RpRa+9D8w7BRbFKXQQiiXRBC88Y1WS004jHOnZiElYhjW0uCiUaul0Vm1f+QRq7UEsFwWd91Vdnqck2gmh9VUDiwD7G1x8Yp3TAXWDBMiy1TNkyJ0Yq6Uc1DLUrwQFL+UUMBzDAIusSsKH53IVL51jxTWKrGrz4PHp6JV8y6LHW0MFmIy5hyxHiLHIi5r8IgcQh6rq6DPJ+Lsago5zUQ4bQ0L6SbXhqYb9uTlSp/hyfx6cCqchqobLZuw2QiJ/GS6SoN4Oga323bC1uTyy62bk+ko/te/PI1kVsMY78b+fi/knI7vDV2BB/YcqhmrshTP4v/ufzm0L70Mr71yBC/a3WcX8b//9AJkVccr9w/jEvKAD38YeP/7y+acIZcrtBoCwG/8htUeWSEXzc4eBIADB6zMt3JDC3I5YM8ee9NTKg/PgWvgMnWMennLfebcPhSyI1zuf4CBxnLgDR0ayyOV1bAYl9HrFatGuGwHOv9oQmkJx+fjePhsGM8uJPDcYgI/eXbJ7h8PpxXMRTNgGQaA5SiyJhjwkHgWAmeduEaCbizGs9jVZ+UcDPolnF5JIZ5R8exCHDfsG9jQRS3Jk9o70BqnT/E0O6vF0Em3nFQPjAYx6F/BSlLBI2fDeNnFg+u2SSsa5mPWwmIfzZOqyESfB0fn4piJpItEKU03sJhvadtRR+sjEa66Jex8OpKBZpiQeA5m/oJH4lnMx2SsJnNw91bPK+sETNNEPN++V+lirRIkgyqRVaEbpp3lQtneVBvcUQtrWyCcVtDnFbG734fTKymspXIYDmhlq5ZZdWPOI/I4ZRPa90zTtEUgd7NOqToQOTJ9r/bvRF6/dSKZJFlV5+EyccR/8Af2l/9270mspnJ4z4t3W0Wd0jaZz3wGOH/ecly95CXANeXr1plMBocOHUIqlcLX//0BAFYhwt1i8U5wZGRphgERbNU8KUIn5kqRQS1PTkWRUjSwLIORoAsLsSxmIzIELrvhNeKFCnFK1XrtdvZaP5+LVs67dBZrNd3AYtxaN/Z4hPyamcNaUsE7b9iNHxxdwLG5OHwSj6GAC7MRGUvxLPq8Qle5NmajMlTdhE/iMFghE27ALyHg5pGQNUyH01uyjibrm4B76z/P7cIwTPzwmQX4JAEswyCl6IjLOfR6pbqy51TdwF2Pz0AzTFw87Md1e/rAMI5QdYbBdx6fxSPnwrhxst8SFzmufiFtbMy6Vdn/qdWUtaYY3YuJT3265prCNE3c278fa5/5Gl7/glGM7emruO2BsSDwh7+Dv3jpzTiznEQup0OUc7hiPITbD451TAZyu+jsq3FKSyD944txGSLPWhUQt4inZ2J4/HwEvV4Rcs6AyLMYCogYD7nhcgSfprIa3CKP11wxgruPzNtVFrfIYTToRjilQOA5KJqOE4uJpj80pP2sVRb50ml2zgVeN1khWZbByy8exF2Pz+KXp9dw3d6+dRc9Z1ZSME1gOOBCcBuf0DbKzrwoVZortRjPQjNMeEUOvd7aLhzSshdO55BSNPhqhIpuJYtxGf/0yBQCLgH/6coRZFUDZ1atcFOfxINnGYyFXDWrsFtNIqtBNwCWabySSH5PzbCErXre4+3ERsSZbmWjo+mPL8ThFqyLNIFjIXAsAi4O4bSKp2djuGjIv07IJVlQTQedk/a9TXBKqboJLZ9l1GqxxQl5LXJabVGKBJ17hOaOp26RB5ArTODjS57npptqtuoB1vrguXyA+9kVSyTa3+LWPcA6t7MMYJiFsPNqeVKETsyVIlX+x86FkciqGMtPi8qqOs6vpeESOFw+1lh+54V43CpFzkduALWdUv0+0Y7pWIxny0YROIu1sUwOugF4JQ6TQz4ATH69zyHoFooG7wwGJDCMdf48Pp/AaMjd8YUswrlV69piz0DlTDiGYTA56MeT01GcXk5tiSiVyOZFqQ4vlG8E4tSb6PMiJuewEMvi/Foamm5iwC/VjFX58fElLOfbSu+4enzd+3n5WBA/Ob6EuKzi6GwMhyYqT5svpdbxptk1BSlmSTyLF+wM1dwPZ4TLhXbs275/+RQAxf3jFw/7cXQugaxmIJxWkMyqiMkqcpqOXX0eZFUdewcqize3XDqMsR6P/aFcTlgfypsvHoLPxWMtlcO3fj2DOw6O46odoYY+UFlVt4OmJwdbY5FfP83OEtLknN51VsjLx4K4//kVrCYVPHIujJftL3ZLnaSte3VBBMjpcMbO8AAKIeckxLgWbpGzQ0Vna4SKbjbOE6thmvjxsUVkVQMTfR68+8bd4PLTpJJZDT6Jw4+PL2E+lsXPnlvGb141vtW7XxGSJxV0Cw1/ZhmGQY9HwGoqh2gmd0GJUhsVZ7qRjY6mX0spuOfIPHq9It51/QTmY9Y5xDAtcSXo5vHb1+1a9xxKXkzqhqBz0rrHMrAn5LUDMT/pSKlDaLOdW02KZPYEvhZmcp1bS4MRPevavVuFwLFQNAO6YdaVJ0XoxFypsZAbu/q8litX4DC1loaL53BgLAiJ5/DEdBSXjgaxf7j2Bf+FeNwqx0wkA9O0BCd/jWIMwzDY1efBc4tJzEQyZUUpUqx9ejaGVFYFwOQHITHrirUsyxQN3kH++OeV+I6aXF0LO0+qwuRCwuSgzxKlVlKo3tTbekzTROICcEo5nXpeyY1kVkMyq2EqnMFyIouRkBuKWohVca5nV5JZPHxmDQzD4A1Xj5ctBnMsgxv29eHwsSX88vQart7VU9eavtbxZiNrikfPWUM9Dk301L02cEa4XEhQUWqb4+wftyq+DFTdxHJCAcBgMCDBI/J49w0T61xQ5cSbSgouAHzvqXk8MR3F/37wHBjGaimrdzFBFvwDPtEe394KSqfZESGt26yQLMvgZfsH8N0n5vDQ6TVct6fgljLNQkD8ZBuquduJ4YALEm9dBCwns3ZLJ2nDqydPirCj19NxopTzxJrJ6VhJZuEWOFyzqxfvuG7CzklwnuxeeyWLrz5wDo9PRfGiPX3r2lw7BTI5r9GQc0LQI2I1las5MGA7sVFxxkm3uBY2Oppe1Q18+7EZKJqB3f0evPuG3QBg/+4/f34ZywkF89HibCPTNKFoG3NKSUL9As5Gyah5V5LItXSiXClFmVJVMAzTFuOaFaXI42ynVAtQNBN9vvZNtCWtxJpuYCas1MyTInRirtSvzq4h5BHxpkM78PKLB+1jxa5eD/7lyTkcnYvj/z02jd+9cTd29HgqHk8aOW4Zhonza2lMp4Dza2nsGwp25HGpWQp5UvU5+nf0WqLUdDiDG/at/zkp1h6ZjiKaUdHnk+B3Vc6tca75F2Iy7np8Fj6JR5+vOwo7qm7Y67taF/l788LzYjyLZFatKQK2EkUzkMu7JTs+U2oDlMaqXDzsx0pSwUJMhqwaOLmUBMsCyaxatJ5NKxqWE1l4JR63XzVetaPmmole3P+cFXlyajlVUwSvdbz5g5fvw/eemm9qTRFJ5/B8flDGtbsrt+1RLKgotc0pDfsOeQSsJnPwSTzGe9zwSjym1tIYDrrrFm8qKbi3HxzDSjKLB0+tQtF07O33YSIfYFfrIojkSbVDVNkuVsgrx0P4j+dXsJrK4dFzYdyUd0tZJ1ANIsdgosNbsLYalmWwo9eDMyspTK1lbAGGhJxXGrdcjp29Hjw5XT1UdDNxnlgHAxKimRwYMEhkNSzEZZxdTZX97O3q8+KK8SCemYvjR88s4ndv3N3Wi9RmaTbknNCbz5UqHeSwXdmoOOOkm1wLzkJMVtWxnFDgElgM+l35nJvq7QGHjy1iIZ6FV+Tw5mt22q8N2XYwIOHL95/BM/NxvCyRxVDAGkmuaIY9xKfpTKm8gNPuTCnDMHFqKYlwSgHPuYpco62GuLCUGqKU093kafL1s51SudYOL5kcqtz2s1FI2LlqmHXlSRE6LVdKzul4YioKAHjpRQPrPltvPLQDWVXHyeUU/vu9p+ASWCzEs+uOJ5eOBOo+bp1YTODuI3M4vZTE4gqLX//785gc9nfkcalZSEB3vWu7XfnC2nQkDdM0y/4dTfR7sbPXA1nV4RU5TK1lqhZryZp/z4APs1EZz8zF8cvTa3jToR0b/O2KaUfhYzps5WkG3Dz6awhpPonHWMiF+VgWZ1Y2ty2WuKTcAmcL+duRcrEqQwEX+n0iFmIyTi0n0eMR8c1HpjETzUDircFXyawKjmWRUjQ8MR3B8fn+ip9xl8Dhhbt78cvTa/jl6dWqolTpOknRdCSyKjTdgEfkcGYlhU9//1lE5RxEjsPTszHohomgW8Dufi94jq26pnj0XBimaXWxDFTIM6MUoKLUNqdUld7V68VI0J2v5Fr94yTse8+Ab0PijWkCKwkFbpGFxLNYS+cg8izGe91VL4JM08TpFudJlbIdrJAsy+CmiwfxL0/M2dlSEs/hdN4ltXfQVzbYklLMRJ8lSk2H07hubx9SioZIWgXDFLKi6oEIWHNRueZFXbtdJs4T654BL06vpCDnrJPqweEQ5qJyVQHiVZcN48RCAmdX03huMYlLRzfP+VXva2M7pZq0thMxK3qBOKWc4oxumFiIy+BZ2FOXaokzhFa6rYDWfRYqPU8yqyEhq0grGuJyQZxYimcx1uNGr0csmrrqfJ6leBaPnA2DYRi88dCOsvl8I0E3DowFcHw+gZ89t4y3XbsLQEFI4liAb/KzXXBKGW0TiojA+PRMDHNRGV4Xh8/+6ETbLuTJBZaqm1V/J+Jukni26d/bm8/C3KhTyjAKI8JTWdV2ULQDEnau6UZdeVKETsuV+vVUBIpmYDjgKtvqyLEM3nrtLnz+8HN48FQUmmHgBeMhjATdRceT268aw7PzCYgci/NraWRyOniWgUfi7Ilwp5aTuO/EEu4+YrkXhgISNBcQ9AhNH5c6EU03MBuxHJn1OqXGezxgGWuSW1xWyxZxHjq9Cp9LwH+6chSvPjCMlKLXfSy+cV8/npmL45m5GF552XDLMkzrLXw0ev4geVJ7++sTlvcN+jEfy+L0JopShmHixGIc4ZSC4WB7iwRbTeVYFQNZ1cCBsSCuHA/iR8eWEEnnEHQLmIvKSGZ1SDyLgyNBzMWyNQtqN+ztx6/OrOHsahrzMTnforqewjrJhYWYjPlYtujnmmFiOt9CG/Iw9t9QNKMivRDH3gEfPCJfdpJ7TjNsof66vdQlVQ9UlNrmlKrSLMvAxRbavkrDvjci3kyF0zizmsIlw0EkFRWzERkL8SxSiobdA96KF0FrqRyiGRU8y2Cinzp9qvGCvFtqLZXDo+cieOlFAziVt4a2K/Niu0EWd9N5hxPJkxrwSQ05HIb8hVbA1ZRiOyZKaddiywk5sQZcPE4sJCCrBjjWEnndIl9TgOjxirhxsh+/OLmKw8cW68oma4W40IgDh7TdNdveG3DxSMgqTiwkcG41teVuyXYLlcQlm2FVnFrJQsu3BsQyGvYNeuEWubILqdJ9XO9aMCu6rdoVFFpKpee5bk8fnpmLYTmhQORZiDyLkFtAJqdD0QycX8tghsnAK/HwSdy69oCleBY+F483XD1etbr6ikuG8OxCAsfnE1iIyRgNue1wcolvvh3O5aiQK5pRto1tI383ToHRI3Lwu3h4Ra6tF/LOqn9ON+z1Rykk42rd5L0GIMfvjYhSx+fj+PbDp+zvn5qL4d+OzNvZSK2GOKXSSv15UgSSK3VudWtzpTTdwMNn1gAAL76ov+LfP58/RpimCa/IYzYmwyVyUHUDLp7FicUETi0nkVY0+F1C0fMkiIhsmkhlVfx/951CTjOwu98L0zQBxnK6+AeFhlygncx8TIZmWFPjarl8CCLPYjQvVk6HM+tEqZSi4ZGzVsbNzZcMYW+Dgd47ej3Y3e/B+bUMHjm7hlcdGGno8eWot/DRzPmDuA/31CH0ApYr8oFTq/nhQeWdZq2E/E5HpqNYiGXhd/FtLRJ0ArViVTwih4fPRiBxLDKqYX/2d/Z64JYEjASZmgW1oEfAleMhPDUbw0OnV/Hma3aW3S6Z1ZBVdSypOiJ5F71X4iByLHiOAQcGC4msNRCox4WgW4RhAOfXUpBVA88vJdHnFSHx6ye5H52LQVZ19HoFXESnotcFFaW2OZsZ9u1sFfS5eAgcg+lwBomshuPzcYyVBNgRTudDunf1eZoOiL1QYFkGN+0fxL8+OYdfnlpBr0fAk9MRcCxLRak62dHrBstYzpt4Ri0KOW8ElmUw3uPG2dU0ZiOZsqJUqxdblS5Iw6kcFmIyTNMKOxU4BvsGffDmgyDrESBeetEAnpyOIpzO4ZFzYbxoIlRx21YIbY06cKJ5p1StrJVK+/vtx2bw5HQUDAOcWU21tf1ss8QZ8n+Vy1TJaQZWU1ksJawwZbfIQtVMpBQNxxcSGAlItku2Es4qYjSTw3xMhqIa8Lt4hDwi+ryivTjM5PS2BYU6Wf88LFaTWdx3Yhn3PruMy0YDCLp5KJqBAyN+uCUBpmliJalgPppBOJ0DwwDfeHgK59bSUFTL3RHN5MBzLDKKjsenorh+b7zi/gwFXLhiLIijc3H87LllvOO6Cdsp1WyeFADwHGvnPmZVfZ0otZG/m1KBcTmRBcMw8Eg8dvd723YhzzsmzCmaUVH4J+17GxGlvBJp36ssStVzTFqJJuHtG4aqGQi6RDy/lMSX7z/dFtGO5PydW0vVnSdF2DPg64hcqaNzcSSyGgJuHldUeX2mwmmcD6dxxY4QZiIZKKqB4/MJ++csY7n3WZaBW2TR55XglXhoholMTkNG0RHLqDBN5IVVHjMRGaZpIiczGNEMuOoowmwljYjKU/kpwbv6ardzOtnZ67FEqUgGV+4IFf3sodOryOkmxnvcuLiOwPlyvHhyAOfXpvHY+Qhu2j/YdLsyUH+buWma+Nufn2no/KFohQFK9f4d7Or1QOQsx+1ywnIutQvnuUzirSJBwM1vK7dfJarFqhydjcE0gUtGg8jkNCzEZGu4UMBqf6tnPQsAN07246nZGJ6Zi+OVl+XKugYFjsFaSoFuWILu7j4P+v2F9zyV1cCwVhF6OpLBcMAqOl02GsBUOIPVpIKzqylMDvowmG/Ps9ZkKXzvyDwSsopXHRjqanF8M6Gi1AXAZoV9l7YK9vtc8LsEnFu1DjpnV9NWW0P+s0lOzv9xcgUJWaWiSp1ctSOE7z4+iwdPr+H+51eRympwCSz+/hdnt3V1pVVIPGdXEqfCaczmA4t3NBFku6PXg7OracxEMutGz7Z6sVXpgvTgzh48dGYVaUXPV0kl7Oz1gGMLF8dyTq8pQLgEDq+8bAj/+uQ8fnZiGX6RLRse2wqhrSg3ZMCLnG6CZSrnHZmmNYoaAILuxpxSZH9Xk5Z7RuAYhFzta/No5xSXSv+XM1Nl76AP470eTK2lIHAs0koOe/oDGA66kdMNnFlJIZW1hKkrx0PY2eOpeKGUkFVE0jmE0woUtdDSFJet1jjDtCaGfePh8zi7koai6RgJeloaFOqk9DMVl1U8v5S2/vY5FqmsCkUz8GevvRT/9PA05mJZjAQZuEUOXpGHT+LhEa3ckMfORxBJ57Cjx43VlAI5Z8AtcLhs1I+ZSPV2V8ByGTwzH8dzi0nMRjK2U8q1wcKKxLNQdd1+Pud73Uj4c+n76WznVPVC9ZnnmIbaORvFNC3BKZXVcHo5iYM7e9a9poZh4vSKlXHV4xGabl8hgla6QqZUvceki8f7IXzubswsRTDcF8CO3vaJdiTonDie68mTIuzq84BhgHC6NblSzbjwTNPEL0+vAgCu39tfNUKAFC5HggIuHvbjucUkVN363PkkDm6RQziVw85eDxbjWYyG3I7XQrKHuuwftrKNej0CZM2w2nV14PmlJC4eCdR90bqR37sZGhWVC3lS9bl8CLv6PHj4bNgWZAjFLqnBpl1AFw/7MeCzBoc8OR3FDfv6az6mUvHEWfhIZlUsJxQksipYhgHHMtAME//x/DKOzEQRz6gYCUlIZFWIPFvz/DEdzsAwrUJWvRN3eY7F7n4vTi6ncHol2TZRqvRcNhPJgGEY+F0CxkLubeP2q0alzhzntaTfJWD/cPFxrZ71LACMhtzYO+DF2dU0fnUmjN+4otjVt5ZScPjYYn6dpOLAqB9BTyH3ydlN9JtXjeJvf36myNwx6JcQl1W4BQ4Bt4i/+8VZHNrVg4fOrOHYXBzT4Qx4jkGvV0CfV6LXZnVARakLhM0I+y4XYCfxHC4Z9mMxnsWJhTh6vCL+7cg8ziwncXw+gdMrKZxaToJlmLxFmX5wa3FiMYHzaym7Umi5FrZXlkK7IZXEqXC6aacUAOzo8cA0TRydjWHvgK/oc2UvtgIuJGQ17/QxwbMsOI6ByLM4Mh3BbDSNlYSCvQNWaCLH1idcJbIqfv78iu0OGQpI0A0zH4BbnNlW2qZbiYM7e/C9I/N45FwYj5wNQ5eLw2PLhtCa69u5qgltM+EMXnxRPx47FwEYE0dmY9ANQBJYjIVc6PdKRRfIE31ePL9ktUkJHItAjYWIE+fCb/+QD0/MWBU4nrcuBlq98Gv1FJd6nWYkU4VjgZ+eWAbHWJNSb754EM8tJZHKakgrlvNmZ48HJxYTkAQOPMfiiz95HllVx3TeuSAJLPYN+PCCnSE8PRNDOGVlA7oEFiNBF0IeEQlZRSwvWOm6iQdPrSGlaHmHhwx3vnLu4lmcXU3h8z96DjFZhVfibRHYK3Lo84kNCSLkMzUUcGEqnMZq0mrpZBlgMOiCT/Ijk9MwWmFwx5U7enD7wTGYpok//tdn4JV4JLI6AEsA2jPghSTU57QY8Et4wY4QnpqJ4WfPLePqXVb71EZcA+TxKUW3J9EBjYXWk/Dn0gvfi4f9CKcUJGQVKaUgeBF3cqMX8vVALsJ/fT4CRTXwpZ+cxGVjgaKLcLLNE1MRLMUVnFtLQ1b1pgosksAhIavIqtq6Ft1an83/dOUojs3HIXIspsNpO8Mu6BbaKtqR/LHVlPW3XE+eFKGVuVLNuvBOr6SwnFAg8SxeWFKYKaW0cHnFeBAwTbD5Akoqq0F1m3jNFSNVp0H/5lXj+OfHphFyixh38cjmVDyeTELRDDy3mMB4j7uui9aN/N6N0mgxwjRNTNtOqcbWJrt6rb+hhZgMRdPtz/gvTxVcUvs3kN/KMAxunBzA956ax6/OWNOgq50/yxVPyJoik9OxmlSwllKQ0wqFDx2mlUNnmojJKhJZDV6Jx3LC+pwsJ7KYHPTD56p8vCZ5Uo1+XieH/JYotZzCiycHGnpsKZXO4c4igYmCu1Pk2LYeb7qBcteShEbWs4Dl6ju7msavz4exZ8CLnGbYx4X/99gMMjkdV46HsJJUsJLMgWO5st1Elcwd1+/tx4sn+3FkJopTyyn88OgCPCIHl8P59txi+5y22w0qSl1AtDvsu1qrYErRcMloELv6PFiIZ3HviWXwHIOdPW74JB7/f3t3HiVXWecN/HvrVtWtvXpL752dLYEYIA4koCyvCWHgBQUdDjDRKMwIGZQZHR1m9EwAD8rACIwoAooBnVHQAYRXY2KUEIEkLEknISvZk9732rdb93n/uFWV6i1d3V1109X9/ZzT5yTV1VXPU797b937u8/zeyQAhztD3HFHkL440YR+UZT+Eq/y2OC1T55aCoU2s9yJdw514S8fdSEQTcBhlVHpGv3KGMFYAttP9MIfUbGnxQ+7VU6d0NahuTeKlr4I2k1RJIdYeEoTAn3hOE72RuBUzNjZpE9jMMsS7Ba9qKtVNmHnyV50h2JZF6T6HZ4T3RHIkoSwqiKpCaz+v/Px9KbD45qmu6fFj6PdoUzNmXJr/+Kxn76wFrubfbDIJhzpDCEY1+98m2UJimyCEALvHOrEgXY/eoLxVKJNgi8S1++Sq0nsbArhYEcgcxdMkiRIkl4o+khnGK19UVR7bIglNOw82YdfbD2OPc1+HO0KQTHrSZRcLxiyT/wkkwk2swmRhIY9LT5Mcysod1n7nfjlcsd8uOcMTBwkhUAknoQmNJQ7rTjeE8aj6/ejJ5SA3SrjRE8YmhCQTSaUOixwK/1Prk83HW5gcjCeSKIjCqA3CodFRjiehM1iwj8vOwf72gKDTqSuPKcS59d5sHF/B/6wuw2qpmFejQc1FQ60+6P44942/HGvnuwsdVqQSAp8rN4Ls6xf3DisZlR5bDjYHkDJNCtO9oThtVmgCQm94QR6cWqFw0RS4GBHMFUoVOt3gtkVjGFWhSvnhEggqiIYVdETiiOa0CBJQLXHhhqvDWbZhKQm0BvSt7WPNZScdnqAx2bBOakbJvqFoi0zvD/X9vyfcyux82QfDrQFEE1o6A7GUOm2jnmkj6YJhONJdAdjONwZwIwyR78LmGluBW2+KHrCcZgkCTaLST9WKDL2tPiwfk9bJulZ47XD7jGhMxjV4/xhKxJJAadihjVV/6LCpaAiNYIg17vPuepXv8qiH8ucSv/6VQAyz7FZ9JN4tzK26Su7m3345bvHse14LzQh0BWIY27VUPuLE4mkgD+SQDiuQkBgV7MP+9v8/Y5JQgj9BsEopkCPhWXAyKJc60ml5aOu1HhGb751UK8l9fGZZUPWQMs2qMapJAGp40H2xebSedWoK3UMO7p/Xo0HW492Z17HapZR6xQIp459u5p8+ORZFSNetOZ7EYfhjGUl1M5ADOF4EhZZQu0wRZqH43VY4LVb4Isk0NwbwexpLgRjKrYeGf8oqbQLp5dgw9429IYT2N3iw4L6kiGfN9TNE6/Dgu3He/H+sR5UuBT0hROZGx/TXAoq3AokAElNwB9NwGICVA2ZYtW9kTgicQ372/yYVeFEyYDFK9IOd46unlTaWalZG8e6Q0gktUH7aNp4puknNYFAVEUiqaG3PQE1tbiCkpr+XajjTTHIZ9kZvT6qwJYjPdh2vBcOqxnxpIZoIplZefrzi2fgeHd4xNlEpxvcccmsMtz58w8QTeijtkNaEpIkYUa5E3aLzGuzHDEpRXk10lTBc6rcuPt/tiGuJqGYLWjxxSBJEipc1oLWtZgs+i93ruJIVxiSpBdxnup3V0YjFFczyaSkJuCyyXjoD/tGdXd0d7MPP33rKPxRFVazCRUuBZoQeOtgF9480IG6EkdmSp3DKqPMaYVF1oejq0mBcExFLGGCqgF2iwma0OuuqEmBQFJFIKpmElfHusPw2i040B6AEMicqLgUM2ZPcyKmaihxWMY1TTd94hxLaJhZ7kBPKIHeKFCharCaJexp8eNAewDhIYrQqkkBNZnMtLcrGIdTMWNfW3DQ+zisZsTUJJyKGdUeGyo9ChSzCR2BGFp8UUQSGj7qCCKuanh+8zFYZFOmKLPFbBrVBUN2nTsAmF7uwInuMCIJDW2+GASi0DSBj9oDCMfVzN354e6Yn+4k02GVcbA9CKci41BnMFP7JC2R1LC3JZBKzvT//Np8UX1aod2CUDyJxhO9WL+nfdiLpRsvrMWuJh8kCdjXGkAwlkAoIcFlA6q9dnjtZvSE4jjeEx72RAoA3jvac6rocG8EnYEYIgkNillGMKbvGw/eMB8/eesojnaFB50clrsULD+/Gi++dxIzKxyIqRp84QQSWauXCU2gqS8CkwSUOa2wW2QkhUBnIAZfRMXuFh+q3CPXtwL0O//t/ijMqW1iTqULHtupof0DEysjTQ/QNGDONBdmV/QvaJtrgqbcpaDSreD3H7bizQOdSKamqPij6piLt2860IlAVEVHIIZ3DnXj5ovq0NIXRVNvWL+Ix6l2BlILBWlCIBBJ4Ht/2Id4UqDGo6AzGIUvoiKuajBJEhJJDRIkKGYTLqjzQLGc6tto7z6PZOBF+N5WP0IxfcTGtAorDneF8L8fnAQkDJq+4rKZRz19JX3h2x3Up+iaTRI8dj2Z3tQbwdXnVOKDYz2QIGFnk6/fiAxAH50QjqmQZQkOqwlui8CfHl2FZCKBSx/+b8h2R96TdmnpQufA6OpJpY23rtRYEiZpLX36RaNJApbksLrUaC42Rxrdn/06VR4FJgD1pTbsbgnAYjYhomr4qCOAc6s9QyYPAPTrdyKpQdW0UU8pHvhZnm5UTLXHhs5ADJ2BGGSThEqPDWUOy6DzNk0T2Hy4G91BfQT1WM6EZ5Q7sKvJh+M9Ycye5srbKKk0i2zCpbPL8ad9HXjrYBcuqPMOSnQN3LY0TcPJJBDpjeirpEYTiCeSqPYqiMaTWFDvhSyfSmzqx6UoLqgvRYc/ihKHFS6bGbWaDYc7Q+gNJ3C4M4RSZxz21MiUtGgiieY+fVTunFEmeqe5FbhtMpp7o1i/pw3zajyjrgs5XMJzV1OfXmO31I42XzRVUkBfsbzSrcCb+j4r1PGmWOSr7MyeFj8OdQRTK/lZUeqQ0OaPIxxXIUnA3VfMhttmyXk20XDnFG3+KGRJwlmVbnQGYxBCP89wpFaE5bVZbqbm1k4Fdbqd+0hnEDFVw8caStDmjyIS14eQlBR4iPxkkX2R7VRkhBMa7BZTZvj7VL67kqvdzT787O1TySSzSUKZ0zqqZEf2yVZDiR094QSOdoegJvWVhQJRFV2BKOZMc6IvksD5tZ5MjABkamMsnjOt38mWpmmIqhoi8SQi8ST6wnGEYqbMymn+1BL3JgmoK7WjxmNDUgDHukIjjg4ZSXbC02KW0BfuQ0iVcLgzBClV3yEU1S/cnFYZZS4FLkWGzWJGUtMyCQkIPQFT4rBATYpTNUNs+lRTu8WM5t4wqr02tPqisFn0wpHVXntmJMi+Vj8SSYHmvghqvDaYJL3ujcdmwZxpuSevB04X8dqtuKBev4Pc5ouiKxhHXNXw31tPoM2vJ4bmVrhg95oH3TEHMOxd9SOdIcwsd+BQRxAO66nV19Lbl143RiCpaZBNJpQ4LHAqZsiShKiaRG8ogURSoMUXRSyRxNObDkPVgPOqXZnizYCAw2rCvtbByUEhAMUsMK/GDZfNiqSmjxRJHweGOpE60hlEc18EF04vQVdQTySmV22s8tjhspkRjKoodykjrpTzamMzInH9gi59EpamFy4+VSi0rlSvE1PlVnC4K4RgVMXeVj/Oq/agOrVYwMCLu/oSO9btbcfmQ11w2fTYzK/1wGoeeAGTW2IlX9MDdjf70HiyD72hOOxWM8wmCW7b6Ef6ZF/ApONtM8vYcrgbbx3sQrXHhkhc01cTdJhR5VYgSRKiCX2any+agEnSV51zWFNTEqP6dBCzSUK5S8GsCqc+akmRcaInUtBFT/qNUJQkyKnP92CHnqROJDX8blcrIAFOxQxfJIF0mshsGl2Nq+xj8VlVLvii+hTdvkgCSU1LjWTzDxqZ6bDKqR995FinX69h1OqLosZjQtfRfQD07SHfSbtssqTXbkskNdSX2lOLVeT+97nWlRopYVLlsaEnFEdnMIakJlDlsaHCaT1tHN5OjZK6oM6b86qoo7nYPN3o/uzXOdgWQGcUMEeT+NR5VbBbTOiLqPjFluO4aHoJdqcuTLOTB5fOKseh1E2Ej9oD8KW+W712c2b05WhG0Z4uSRFJ6FPUOgJRZJeK80eDUMwmVLisiMSTCKQWBXp5exPePtiFvnACTb0RRNXRr8Y2o8yBnSf7sP14L2wWGev3tMFmkfMySirtktnl2PRRJ072hPGXjzpR4rAOm4zrCsbQ3BtBT1iC06QXs6/x2qGYTbjzE7PwyvZmHBnixkeZ04ovLJmBVxubs1YRN+GsKheaeiJo7ovgWFcIZ1e5UeM99f3x1sFOdAViqPQoo07s7GnxY19rAAfbgzjQHkCN1zaqupBDTdMPxVR0BaOZUg6+cBye1IIcZ1c5UWK3DjlqMN/Hm2Iy3rIz6e8GIQTKnVaoGtDii8Eim3BOlRsxVcPrO1uxoL4EJpM0rtlE+kwADTMrnJjmVtARiKIma/EjXpvlhkkpKojhdu5ThS7tqHApaOmLIJrQRj1tYqoaeJE9fUAdpKl+d2Uk2Rcw00vt6E4tAVvhUlDiyH36Y/ZFVyiuoiecgJoUkCQ9wVpbYoMA8PlLZ+Dl7c043BnK+WTLYTXBYTVDCIFIIolPnDUNbb4IbBYZJhMQVwXKnNZM7ZpITM1pdMhIshOesklCXakdB4JBuBQZLpsFdouMnlAM9aV6Edoary3r5NYEm0Uf/XLRjP53NQcKRlXYrPIwdUM0hOPJ1OpgMWhCIBzXEI7rdSQU8+jqLQydfJDgtVvhsVmQ1Pxw28zoDETRF47DY7PgcFcIFS59id9ypxXNfRG8sPkoZJMJXcEYzq50QTKZoGoawnF9SuKx7hBOdIeA1KV1jdeGCpcV9qwETTCqwqWYTyVnsor4ziwX8EXi2N8ahNtuRiCqwmqWcaA9BLMpDAGRmQIqSaeSg26bGZUeBS6rjPaWQCYhlMtxIB1vp9cCj12f7hFTNVS6lcxUuO7AyFPhNE3klOAZqlDo9FIH9rf7YbPIcChm/HDjISxsKMGWI92Zizt9m9cwzW1DmdOKzy1qwHtHe3C8e+gLmFwSK/mYHpA+lgSjKuZMc2VqAjmV0Y30GTia4GhXCOG4lrrbqk9fkSVgTqUT/oiKeTWeQZ/xoY4gzq50obkvgmqPDYmkniR2WM2ZUXnp6SKfvageu5p9BV30ZOAIxRKHFYGYmhk5mC5gDElPyqQH1kmpJBWQ+7lA/wSYngROJAXafPowMptFRjSuwmbRa1ZWehQ4reZ+MQlGVTgUc+aYdKTTl/ldKKbiZCCY16Rd2u5mH37/YSsOtgeR1ATa/VF0BWOjSkDYLDJqvTbsaw1g/Z5WLJpZNqqESYc/hpa+CNqkKLIGOOJIZwgtfRFUeZR+qyankzOtfVG8dbATTsWMy88audB1tnzVOE2/zqF2H9a90Y7lV5+LuVVeCAAvb2vCn/d34LENB+FUZJxT5ckkD3ae7MOWQ93oCsXgyUpUAulFHIJQzPp22ROKIxwf46iYk33YcbIP5Q5rZoqaU5FR47VDTWroCMQQUzUc6w5DTWr4fztb9IUoYioE9PO9Ks/obpilRdUktp/oRTCm4g+72xCMqqgtsUEdqpbAGLkUMyrdCl7f2YL3j/Wg3Kn0+2ziajoZF8vcsJNNQF2JDZUeO2SThGNd+rnRSIlKkyQNOl6XOKzoCsZgt8hw2yz46VtH8fGZZVi/tw1bDnejOxhHmdOC7/w+96ReOpZdqYVRFLMJJXbrkAmn+lI7YqqG7qAKVRNQzCYcbA/i31/bA18kDotsws6mPqhJkbVvSSh1WGFNJePWftiGzkAc5mFqGU31GSPjSRSd+m5wwKnoCV5J0hO2lR6bvvhGngZBDLw2m6X0fz1em+WGnw4ZauCOW1fKpMpo5LMA4FSUfQETjquZpJRTGd30x+yLLodVRkLVYEklMdIX9Me6Qqge48nWsImrysLFfOC+We2xIeEWaKjRR3kFoyriSW3EIrQDE23Dtfd0dUPOr/PgxfdOor7Ujo5ADO1+/YLJbh3diMCRkg9VHhtuvqgez28+Bo/dkrnb1dwXzbxGIqlh4/5OQNKnHvoifZBN+lTL9EW2UzHDJEmYX+uFL5JAfal9VMmZSDyJzoA+0uOa+VV4fvOx1HvpI6gAfelij80Ch1VGXziOhtQKVeVOfVWq0W4TA+NdPqCmWq5T4XJN8Aw3QuKTZ1Vi8Zwy7DjZh0MdIby2swV2swnnVnsQt2g42B5EIJZAVzCOf1l+Lq5bUIPFs8vHPax/vNMDso8lVrMJ3aE4NKEnXMZSvD09qii7fonNImOaW4FskjIJ7uE+4+sX1OK/3z0O2WSCd4ilr9Px/FhDCW5cWFfQRU8GHUu8NlR7FAjo22cgpqI1tY9VeW1wWs0QEJBNEsypEaW5ngsMTIBVuhV0h+JwphYBsVtltPuimZGZrtSxPm2oY9KvNn+E11K/90cSWDCjKq9JO+DUxW9zbyQzorLaaxvTKLvtx3uxtzWAfa1+1O1qHTFhEool+o3CS08zdylmTHNbIUFCmz+KaELDoY5Qquh2CJIEvNqob4OtviiCURXTy+zoCydQP8pyVvmqcWoySZhV4cQMl75y4anpfXX4/YetqULfJvijccgmPYkRiCTQFYojpmpwKfpIzPRy8+3+GDoDUQRjScRVDT/edBhdwRhMkFBfmtvKoqqmwReJwx/VR8WEoglUuq2IJwUW1Hkzo6ZrSuzoDkSxtzUAlyLjLx91oiesrwgqhN63aW4bqjzSqKez/uq9k+gLx2Gz6De4rGYTYmoSP/jzobzVytrd7MP7x3ozi+5UexUIIWFXUx92NflQ7rScSsZZZVR5bIiaAqgtsWfOKdL7+OxprtMmKoc7Xi+eU4HFc8rw3tEe7G7RF3lw2yyZ6VOVHiXnfSr7BsH8Wg8aT/oQTejTOr12Mw53BvHvv92N7lAcJpOUWrymv6QQONkTzkzTT+cATRJQ6rBimkeB0yrjWFcYsypchqyMPlUNnFkC6NuEOzVFMp+DIHhtlh+88idDcccdn3wWAJyKsr+k9KXQ9RXCrObRJTsGXnTVn2bE2lhPtkaTuMpHzAfum9lGU4R2NO0d7o75se4QXm1sRiIp0FDmQLXXhlBMhdc++noLI32+SU0gqQnMqtCLyPcEY/BH9XpKqiYQTyYRiqkQ2qmVstInmg6rnjQosVvQ1BvB9Qtq8IfdbWNKzmRPh/vtjhaU2K2YXeFEMJaEbEKqiLCEYFRFUvRfoarKo0AT+pLf7f5YTttEPo/FuSZ4TjdC4rI5Ffjyf29DLJGEIptwpCuEmKqlpnnYIJtM2Ha8F9eeX533kRZjeZ2BIwtrS2xo7ovCPcqRPoOSKh69Np3XbkGJ3ZKZnjtSgntg8efTxbPQi54MuW1JUqY2Toc/hoUNJQCAD5t9KKm0QJL6T2/Odfsb6kZX9s2u04/MHPqYdN/y8/Dd1N/fd+3ZWDCrNq/fqdkXvzVeG9pSq9eVOa0odVhzTkCkE07t6VEdliFGdWxvRlcwhvpSOxJJDSd7ougOxaEm9VF4JgjMqnAgEFVxfq0HUiphUuVR0O6PYl9rACUOC36/qxUfdQSgmGXMqnAC0BMdoXhyQi5Qc7wnDE0TmF3hRHcogabeKJp60zcb9DqmPSF9RdDsmwjTyxyo9SrY2eSD127Gye4wesJxeO0WdASiKHVYISBQ4rDgRHcIj67bj+5wAk6rjKbeCFRN01cl1fT3KXNY4bDK+PKVc/DqEKOm+yIqzq/zYsnccvx88/FT028BOCwy5HQCa5TTWXtDcdR67QikVtoscZgxr8aNQx2hvNRtTb9PKKZiRrkDfWH9u8epmFNT1eIIRCyocFsRVzVcUO+FBOBkn/73Q+3jIx2XTne8vmh6Ke584QMEY6r+PZ2a8l/lscNsyi2p17+EgZ7ICMWS+Kg9Pe1Y4GRvJJNwMkn6jQPFbIJZ1hPqEoDmvgjMsoS6Ehs8ditkkwSLyZR539Ek42jsBn43DFwwIJ+DIHhtlh+TJin11FNP4dFHH0Vrayvmz5+PJ554Ap/4xCfOdLNoAO6445evAoBT0cAvqfm1nsydeSD3L6nRXtCP52Qr/ftCx3zgvnm6REc+2zvUZzPw87XIpswU37Ekr0eqc5e9TVS4bajIqgMbjKqwW/QVQqs9NtitMlRNQIK+BH36OelRKGdVuceVnOk3Ha7S1W8K5HDJwUxNlXAi520i38fifBQKBYB5NR60p6a1AEC1R0FDmb5oQPZFWT5HWozldQaf8Doyo52A3I8lA19HMcuYkbVdZ0/PHekCZqJ8t+aybd18cT0A/QJuPO3N9VicSzI9u/3Zr5/vzyz74tcf1Udb6NuJfqMklwREdmJrXo0bjSf1Au59kTjMJmBfmx/f/N9d8EcSMMumTM2kNLtVRmWqNtnKJalReP0SJhqCsSQuqPfiE2dV4DcfNCEQVSHZJBzuDEIICR6bjPPrPHlLdOTTqRovLrhsMRzv1heFKXFYUO2xwWGVsafZD5diHnL7m1nuxE0X1uGZvxyBLEuIqyJTey8tkdSwt3XoxSucVhm1JXZ47GYc6wqjdoSkclITWOtqQ7VXr7/UG06gyntq9OpYprP2ReKZpFR9iT69NV91W7PfBxDoCwf6fT5euxWKRcbdV87BbxubcaQzNKabJwMNd7z2RRJwKDJqPDaEExqQGlmdHnmaS78H3iCocis4Hg/DbDbBnrp52e7XE8Azyp0oc1qAAWXog1F9Rc/0NP0qjzzu80MaG6MHQfDabPwmRVLqpZdewj/+4z/iqaeewmWXXYZnnnkG1157Lfbu3Yvp06ef6ebRANxxxy9fIwWmmoFfUtlFmUfzJVWI5Op4E1f5MGTx2GESHYVsr5Gfb04nLql+f9isJ4rM8vCjOkwmaVzJmbGsUDWwpspoEkn5PBaPu1BoQi8UWunRp1p5bOYJW29wqO0mve0Usuh6rsWfz/R3a65tGW9787mim1GyL34ViwmaEKh0jy4BkZ0UMMv6NPJQLImW1LRICRI6A/oqUCUWGRZZyqwEW+HSiz8ntdxG4TmsMv64twMuxYK+SDwzpVivK5i/REc+ZSd7qzw2eG1mmEynRkUHoyrKXFb87SXTsfVoz7CJIotswllVbkQTSbT7o4gkkpCzalBlL17hsJohSfp766N6pZxHxaRvjggBzCh3YsaAxQzHMp1VwIKWvqjeHkd+pyz1Hymq9zkQVaGYTajx6vX/jnfrtRNzPacYb3sSqsB5tR609kXR6tentqfl0u+BNwj0G1QK0omnYFSFbJIyCadSh6XfogS5TNPnzXfjnIlBEBPlO6ZYTYqk1GOPPYY77rgDd955JwDgiSeewPr16/HjH/8Y3/ve985w62go3HHHj3dXRi+fX1Jn4gLQiJiPN9GRbTztNerzLcSojvHGabQjzYaqqTKa95oIx+JiW8QhX8eSMzVizQi5tCUf7c3Xim7ZKioqEI/HR3zeWAya/j2G2poDR3U0lDrQ7o/CLEuwynqNqqbeMCxmGTPLnfDYB6/KF4nnljDZebIPidTKUjMkB7oCMaiahopUHbqJljAGBid7bcPcgFo6rxpL51XnNIp24HYTjKpwWOVMkuJ0tQRHSirna1RH9rbltlkwv9YDm8WEdGIlX8fRgdvwWVWufivCDpWMy8c5xcjt0VBf5kBNiS0z9RHIrd/DLYwCjD7hNJFuEExlk/U8fbKaGGd34xCPx7Ft2zbcd999/R5ftmwZNm/efIZaRbngjktnQj6/pCbSBWA+jTfRkS9Gfb5GjeoYbZuM2rYmwrG4GOsN5utYMpFGrOVbLm3JR3vzub84nU60tLRg7dq1cDrzv73lY1sfmBTw2C39Ek/BqIpKjy2TMHHbhi/wPlLCZOB7VWYtdQ5MvIQxMPpk73gSRfkYFZOv5PTANqdXsxzY5vEeRwe+j9lkgsc+/BT7Qp9TDGxPdkIq136Pd9GO0UzTJ+MwDsVj4nyDjFFXVxeSySSqqqr6PV5VVYW2trYh/yYWiyEWi2X+7/f7AQCJRAKJxKnVFNL/zn6MJibGqnhMhFidU+nAfcvOwvGecOZLakaZAyaTNKZ2NZQoAPS7xsmkimQyzw0+AyZCnNKM+Hxz2Sbyvd3kIpe+T6RYjceNC6pxsjuEg+0BVHttsFtkRBJJtPmiKHVacMOCqgm3f41mmzhdnM7EtjUZ5etYUeh9arzbep3HitkVDuxu9mPONOeghElLXxgX1Hlw48dq8KM3j4xrn8r1veo81jOyrQ4Xq3MqHVh1xSy8uqMFhztCaPMloZhlnF/rxqcX1uKcSseI7c0lTudWOcf9Pvlqb65tzsdxdLTvM9H3KSD3GIzmeF2M54eT5ZwiWzHGYSTFEqdc2yeJ7LWki1BLSwvq6uqwefNmLF68OPP4Qw89hF/84hfYv3//oL+5//778cADDwx6/Je//CUcDsegx4mIiKjwmkLAex0mtEUAVQPMJqDaDvxVpYb6iTNIimjcxrutN4WAdSdNCKlAiRWwmoC4BvTFAacZWN6gv04+9qlc32si0gTQFQUiScAuAxU2YDSDJHL9/Mb7Pvl8HaOOoxPteJ2v9uQrlkQEhMNh3HbbbfD5fPB4PMM+r+iTUvF4HA6HA7/5zW/wmc98JvP4vffeix07dmDTpk2D/maokVINDQ3o6urq92ElEgls2LABS5cuhcUyeD4+TRyMVfFgrIoD41Q8JlusNE0MeQe62E22OE1WkUgE119/PXp6erBp06bTnkSP13i39T0t/syojpiqj+qYW+nEpxfWYn7tqXbnY5/K9b2MZsR+VYzHJKPanOv7GHX8K8ZYTTT8rioOxRInv9+PioqKEZNSRT99z2q14uKLL8aGDRv6JaU2bNiAG2+8cci/URQFiqIMetxisQwZ1OEep4mHsSoejFVxYJyKx2SK1dk11jPdhIKZTHGajOLxON566y0AgCzLBY/VeLb1hTPKsaChLKd6KePdp0bzXmdCoferYjwmGdXm0byPEce/YozVRMTvquIw0eOUa9uKPikFAF/72tewYsUKLFq0CIsXL8azzz6LEydO4K677jrTTSMiIiIiKggjC9tPpCL6REQ0eUyKpNQtt9yC7u5uPPjgg2htbcX555+PtWvXYsaMGWe6aURERERERERENIRJkZQCgFWrVmHVqlVnuhlERERERERERJQD05luABERERERERERTT1MShERERERERERkeEmzfQ9IiIiIqLxcjgcSCaTZ7oZREREUwKTUkREREREAJxOJ/r6+rB27Vo4nc4z3RwiIqJJj9P3iIiIiIiIiIjIcExKERERERERERGR4Th9j4iIiIgIQDQaxU033YSOjg5cffXVsFgsZ7pJREREkxqTUkREREREAJLJJP7whz9k/k1ERESFxel7RERERERERERkOCaliIiIiIiIiIjIcExKERERERERERGR4ZiUIiIiIiIiIiIiwzEpRUREREREREREhuPqewCEEAAAv9/f7/FEIoFwOAy/388lgSc4xqp4MFbFgXEqHoxVcWCcikMoFMr82+/3w2Ti/duJjPtVcWCcigdjVRyKJU7p/Eo63zIcJqUABAIBAEBDQ8MZbgkRERERTQQzZsw4000gIiIqeoFAAF6vd9jfS2KktNUUoGkaWlpa4Ha7IUlS5nG/34+GhgacPHkSHo/nDLaQRsJYFQ/GqjgwTsWDsSoOjFPxYKyKB2NVHBin4sFYFYdiiZMQAoFAALW1tacdecyRUgBMJhPq6+uH/b3H45nQwaZTGKviwVgVB8apeDBWxYFxKh6MVfFgrIoD41Q8GKviUAxxOt0IqTROlCciIiIiIiIiIsMxKUVERERERERERIZjUuo0FEXB6tWroSjKmW4KjYCxKh6MVXFgnIoHY1UcGKfiwVgVD8aqODBOxYOxKg6TLU4sdE5ERERERERERIbjSCkiIiIiIiIiIjIck1JERERERERERGQ4JqWIiIiIiIiIiMhwkz4p9b3vfQ8f//jH4Xa7UVlZiU9/+tM4cOBAv+cIIXD//fejtrYWdrsdV155Jfbs2dPvOc8++yyuvPJKeDweSJKEvr6+Yd8zFoth4cKFkCQJO3bsKECvJh8j4zRz5kxIktTv57777itk9yYVo/ep3//+97jkkktgt9tRUVGBm266qVBdm3SMitWbb745aJ9K/7z//vuF7mbRM3Kf+uijj3DjjTeioqICHo8Hl112GTZu3FjI7k0qRsZq+/btWLp0KUpKSlBeXo6///u/RzAYLGT3Jo18xKmnpwdf+cpXcM4558DhcGD69On46le/Cp/P1+91ent7sWLFCni9Xni9XqxYseK054jUn5Gxeuihh7BkyRI4HA6UlJQY0b1JxahYHTt2DHfccQdmzZoFu92OOXPmYPXq1YjH44b1tZgZuU/dcMMNmD59Omw2G2pqarBixQq0tLQY0s/JwMhYpU3UPMWkT0pt2rQJ//AP/4CtW7diw4YNUFUVy5YtQygUyjznkUcewWOPPYYf/vCHeP/991FdXY2lS5ciEAhknhMOh7F8+XL827/924jv+c1vfhO1tbUF6c9kZXScHnzwQbS2tmZ+vv3tbxesb5ONkbF6+eWXsWLFCnzxi1/Ezp078c477+C2224raP8mE6NitWTJkn77U2trK+68807MnDkTixYtKng/i52R+9R1110HVVXxxhtvYNu2bVi4cCGuv/56tLW1FbSPk4VRsWppacGnPvUpzJ07F++++y7WrVuHPXv2YOXKlYXu4qSQjzi1tLSgpaUF//mf/4kPP/wQzz//PNatW4c77rij33vddttt2LFjB9atW4d169Zhx44dWLFihaH9LWZGxioej+Nzn/sc7r77bkP7OFkYFav9+/dD0zQ888wz2LNnDx5//HE8/fTTOV2DkbH71FVXXYVf//rXOHDgAF5++WUcPnwYn/3sZw3tbzEzMlZpEzZPIaaYjo4OAUBs2rRJCCGEpmmiurpaPPzww5nnRKNR4fV6xdNPPz3o7zdu3CgAiN7e3iFff+3ateLcc88Ve/bsEQBEY2NjIbox6RUyTjNmzBCPP/54oZo+5RQqVolEQtTV1Ymf/vSnBW3/VFLo419aPB4XlZWV4sEHH8xr+6eKQsWps7NTABB/+ctfMo/5/X4BQPzpT38qTGcmuULF6plnnhGVlZUimUxmHmtsbBQAxMGDBwvTmUlsvHFK+/Wvfy2sVqtIJBJCCCH27t0rAIitW7dmnrNlyxYBQOzfv79AvZncChWrbGvWrBFerzfvbZ9qjIhV2iOPPCJmzZqVv8ZPIUbG6bXXXhOSJIl4PJ6/DkwhhY7VRM5TTPqRUgOlh7KVlZUBAI4ePYq2tjYsW7Ys8xxFUXDFFVdg8+bNo3rt9vZ2/N3f/R1+8YtfwOFw5K/RU1Ah4wQA//Ef/4Hy8nIsXLgQDz30EIcEj0OhYrV9+3Y0NzfDZDLhwgsvRE1NDa699tpB02Aod4Xer9Jef/11dHV1cVTHGBUqTuXl5TjvvPPw85//HKFQCKqq4plnnkFVVRUuvvji/HZiiihUrGKxGKxWK0ymU6dpdrsdAPD222/no+lTSr7i5PP54PF4YDabAQBbtmyB1+vFJZdcknnOpZdeCq/XO65j6FRWqFhR/hkZK5/Pl3kfGh2j4tTT04P/+Z//wZIlS2CxWPLYg6mjkLGa6HmKKZWUEkLga1/7Gi6//HKcf/75AJCZslBVVdXvuVVVVaOaziCEwMqVK3HXXXdxuso4FTJOAHDvvffixRdfxMaNG3HPPffgiSeewKpVq/LT+CmmkLE6cuQIAOD+++/Ht7/9bfzud79DaWkprrjiCvT09OSpB1NHoferbM899xyuueYaNDQ0jL3BU1Qh4yRJEjZs2IDGxka43W7YbDY8/vjjWLduHeurjEEhY3X11Vejra0Njz76KOLxOHp7ezNTV1pbW/PUg6khX3Hq7u7Gd77zHXz5y1/OPNbW1obKyspBz62srOSU2DEoZKwov4yM1eHDh/Hkk0/irrvuylPrpw4j4vQv//IvcDqdKC8vx4kTJ/Daa6/luRdTQyFjVQx5iimVlLrnnnuwa9cu/OpXvxr0O0mS+v1fCDHosdN58skn4ff78a//+q/jbudUV8g4AcA//dM/4YorrsCCBQtw55134umnn8Zzzz2H7u7ucbV7KipkrDRNAwB861vfws0334yLL74Ya9asgSRJ+M1vfjO+hk9Bhd6v0pqamrB+/fph57LT6RUyTkIIrFq1CpWVlXjrrbfw3nvv4cYbb8T111/PRMcYFDJW8+fPxwsvvIDvf//7cDgcqK6uxuzZs1FVVQVZlsfd9qkkH3Hy+/247rrrMG/ePKxevfq0r3G616HTK3SsKH+MilVLSwuWL1+Oz33uc7jzzjvz0/gpxIg4feMb30BjYyP++Mc/QpZlfP7zn4cQIn+dmCIKGatiyFNMmaTUV77yFbz++uvYuHEj6uvrM49XV1cDwKBsY0dHx6Cs5Om88cYb2Lp1KxRFgdlsxty5cwEAixYtwhe+8IU89GBqKHSchnLppZcCAA4dOjSu15lqCh2rmpoaAMC8efMyjymKgtmzZ+PEiRPjafqUY+R+tWbNGpSXl+OGG24Ye4OnKCO+p373u9/hxRdfxGWXXYaLLroITz31FOx2O1544YX8dGKKMGKfuu2229DW1obm5mZ0d3fj/vvvR2dnJ2bNmjX+DkwR+YhTIBDA8uXL4XK58Oqrr/abllJdXY329vZB79vZ2Tnuc5OpptCxovwxKlYtLS246qqrsHjxYjz77LMF6MnkZlScKioqcPbZZ2Pp0qV48cUXsXbtWmzdurUAPZq8Ch2rYshTTPqklBAC99xzD1555RW88cYbg07mZs2aherqamzYsCHzWDwex6ZNm7BkyZKc3+cHP/gBdu7ciR07dmDHjh1Yu3YtAOCll17CQw89lJ/OTGJGxWkojY2NAE4lQej0jIrVxRdfDEVR+i2NmkgkcOzYMcyYMWP8HZkCjN6vhBBYs2YNPv/5z/NiYBSMilM4HAaAfnWK0v9Pj0yk0zsT31VVVVVwuVx46aWXYLPZsHTp0nH1YSrIV5z8fj+WLVsGq9WK119/HTabrd/rLF68GD6fD++9917msXfffRc+n2/c5yZThVGxovEzMlbNzc248sorcdFFF2HNmjWDvrdoeGdyn0qPkIrFYnnqzeRmVKyKIk9RuBrqE8Pdd98tvF6vePPNN0Vra2vmJxwOZ57z8MMPC6/XK1555RXx4YcfiltvvVXU1NQIv9+feU5ra6tobGwUP/nJTzKrFzU2Noru7u4h3/fo0aMTrqr9RGZUnDZv3iwee+wx0djYKI4cOSJeeuklUVtbK2644QbD+1ysjNyn7r33XlFXVyfWr18v9u/fL+644w5RWVkpenp6DO1zsTL6+PenP/1JABB79+41rI+TgVFx6uzsFOXl5eKmm24SO3bsEAcOHBD//M//LCwWi9ixY4fh/S5GRu5TTz75pNi2bZs4cOCA+OEPfyjsdrv4r//6L0P7W6zyESe/3y8uueQSccEFF4hDhw71ex1VVTOvs3z5crFgwQKxZcsWsWXLFnHBBReI66+/3vA+FysjY3X8+HHR2NgoHnjgAeFyuURjY6NobGwUgUDA8H4XI6Ni1dzcLObOnSuuvvpq0dTU1O85NDKj4vTuu++KJ598UjQ2Nopjx46JN954Q1x++eVizpw5IhqNnpG+Fxsjj3/ZJmKeYtInpQAM+bNmzZrMczRNE6tXrxbV1dVCURTxyU9+Unz44Yf9Xmf16tUjvk62iRjsicyoOG3btk1ccsklwuv1CpvNJs455xyxevVqEQqFDOxtcTNyn4rH4+LrX/+6qKysFG63W3zqU58Su3fvNqinxc/o49+tt94qlixZYkDPJhcj4/T++++LZcuWibKyMuF2u8Wll14q1q5da1BPi5+RsVqxYoUoKysTVqtVLFiwQPz85z83qJfFLx9x2rhx47Cvc/To0czzuru7xe233y7cbrdwu93i9ttvF729vcZ1tsgZGasvfOELQz5n48aNxnW4iBkVqzVr1gz7HBqZUXHatWuXuOqqq0RZWZlQFEXMnDlT3HXXXaKpqcngHhcvI49/2SZinkISgpXIiIiIiIiIiIjIWJygS0REREREREREhmNSioiIiIiIiIiIDMekFBERERERERERGY5JKSIiIiIiIiIiMhyTUkREREREREREZDgmpYiIiIiIiIiIyHBMShERERERERERkeGYlCIiIiIiIiIiIsMxKUVEREQ0BUiShN/+9rdnuhlEREREGUxKERERERXQypUrIUkS7rrrrkG/W7VqFSRJwsqVK/P2fvfffz8WLlyYt9cjIiIiKhQmpYiIiIgKrKGhAS+++CIikUjmsWg0il/96leYPn36GWwZERER0ZnDpBQRERFRgV100UWYPn06Xnnllcxjr7zyChoaGnDhhRdmHovFYvjqV7+KyspK2Gw2XH755Xj//fczv3/zzTchSRL+/Oc/Y9GiRXA4HFiyZAkOHDgAAHj++efxwAMPYOfOnZAkCZIk4fnnn8/8fVdXFz7zmc/A4XDgrLPOwuuvv174zhMRERENg0kpIiIiIgN88YtfxJo1azL//9nPfoYvfelL/Z7zzW9+Ey+//DJeeOEFbN++HXPnzsU111yDnp6efs/71re+he9///v44IMPYDabM69zyy234Otf/zrmz5+P1tZWtLa24pZbbsn83QMPPIC/+Zu/wa5du/DXf/3XuP322we9NhEREZFRmJQiIiIiMsCKFSvw9ttv49ixYzh+/Djeeecd/O3f/m3m96FQCD/+8Y/x6KOP4tprr8W8efPwk5/8BHa7Hc8991y/13rooYdwxRVXYN68ebjvvvuwefNmRKNR2O12uFwumM1mVFdXo7q6Gna7PfN3K1euxK233oq5c+fiu9/9LkKhEN577z3DPgMiIiKibOYz3QAiIiKiqaCiogLXXXcdXnjhBQghcN1116GioiLz+8OHDyORSOCyyy7LPGaxWPBXf/VX2LdvX7/XWrBgQebfNTU1AICOjo4R61Nl/53T6YTb7UZHR8e4+kVEREQ0VkxKERERERnkS1/6Eu655x4AwI9+9KN+vxNCAAAkSRr0+MDHLBZL5t/p32maNuL7Z/9d+m9z+TsiIiKiQuD0PSIiIiKDLF++HPF4HPF4HNdcc02/382dOxdWqxVvv/125rFEIoEPPvgA5513Xs7vYbVakUwm89ZmIiIiokLhSCkiIiIig8iynJmKJ8tyv985nU7cfffd+MY3voGysjJMnz4djzzyCMLhMO64446c32PmzJk4evQoduzYgfr6erjdbiiKktd+EBEREeUDk1JEREREBvJ4PMP+7uGHH4amaVixYgUCgQAWLVqE9evXo7S0NOfXv/nmm/HKK6/gqquuQl9fH9asWYOVK1fmoeVERERE+SWJdAEDIiIiIiIiIiIig7CmFBERERERERERGY5JKSIiIiIiIiIiMhyTUkREREREREREZDgmpYiIiIiIiIiIyHBMShERERERERERkeGYlCIiIiIiIiIiIsMxKUVERERERERERIZjUoqIiIiIiIiIiAzHpBQRERERERERERmOSSkiIiIiIiIiIjIck1JERERERERERGQ4JqWIiIiIiIiIiMhw/x9/XBk213weogAAAABJRU5ErkJggg==",
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "import pandas as pd\n",
+ "import statsmodels.formula.api as smf\n",
+ "import matplotlib.pyplot as plt\n",
+ "\n",
+ "# 1. Convert to monthly time series\n",
+ "spills_gdf['report_month'] = spills_gdf['Initial Report Date'].dt.to_period('M')\n",
+ "monthly = spills_gdf.groupby('report_month')['report_delay'].mean().reset_index()\n",
+ "monthly['report_month'] = monthly['report_month'].dt.to_timestamp()\n",
+ "\n",
+ "# 2. Create ITS variables\n",
+ "monthly['time'] = (monthly['report_month'] - monthly['report_month'].min()).dt.days // 30\n",
+ "monthly['post_2020'] = (monthly['report_month'] >= pd.Timestamp('2020-01-01')).astype(int)\n",
+ "monthly['time_post'] = monthly['time'] * monthly['post_2020']\n",
+ "\n",
+ "# 3. Run the ITS model\n",
+ "its_model = smf.ols(\"report_delay ~ time + post_2020 + time_post\", data=monthly).fit()\n",
+ "print(its_model.summary())\n",
+ "\n",
+ "# 4. Predict and plot\n",
+ "monthly['predicted'] = its_model.predict(monthly)\n",
+ "\n",
+ "plt.figure(figsize=(12, 6))\n",
+ "plt.plot(monthly['report_month'], monthly['report_delay'], marker='o', label='Observed', alpha=0.6)\n",
+ "plt.plot(monthly['report_month'], monthly['predicted'], linestyle='--', color='red', label='Predicted (ITS)')\n",
+ "plt.axvline(x=pd.Timestamp('2020-01-01'), color='black', linestyle='--', label='Policy Change (2020)')\n",
+ "plt.title('Interrupted Time Series: Monthly Report Delay')\n",
+ "plt.xlabel('Month')\n",
+ "plt.ylabel('Mean Report Delay (Days)')\n",
+ "plt.legend()\n",
+ "plt.grid(True)\n",
+ "plt.tight_layout()\n",
+ "plt.show()\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "e9ce1703",
+ "metadata": {},
+ "source": [
+ "## Interrupted Time Series Analysis\n",
+ "### 🔍 **Model Findings**\n",
+ "\n",
+ "| Term | Coefficient | p-value | Meaning |\n",
+ "|-------------|-------------|---------|---------|\n",
+ "| **Intercept** | 16.67 | 0.000 | Predicted delay at baseline (month 0) |\n",
+ "| `time` | –0.17 | 0.107 | Slight (non-sig.) decline pre-2020 |\n",
+ "| `post_2020` | –5.09 | 0.753 | No significant level change |\n",
+ "| `time_post` | +0.12 | 0.558 | No significant trend change after 2020 |\n",
+ "\n",
+ "**R² = 0.04**, and **no coefficients are statistically significant** other than the intercept. In short: ITS didn’t detect a strong policy effect.\n",
+ "\n",
+ "\n",
+ "\n",
+ "### What the Plot Shows\n",
+ "\n",
+ "The **visual trend** still tells a useful story:\n",
+ "- High volatility pre-2020 with outliers and erratic reporting\n",
+ "- A more stable pattern post-2020\n",
+ "- No obvious immediate *jump* or *trend break*, but lower variance\n",
+ "\n",
+ "\n",
+ "\n",
+ "### How to Report This\n",
+ "\n",
+ "You can say:\n",
+ "> “We conducted an Interrupted Time Series analysis using monthly average delays and found no statistically significant level or trend change post-policy. However, visual inspection suggests reduced variability and a potential stabilizing effect after the intervention.”\n",
+ "\n",
+ "---\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "6ae433ec",
+ "metadata": {},
+ "source": [
+ "## ITS by Spill Type"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 32,
+ "id": "bfd7557f",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "\n",
+ "=== ITS Model Summary for Historical Spills ===\n",
+ "\n",
+ " OLS Regression Results \n",
+ "==============================================================================\n",
+ "Dep. Variable: report_delay R-squared: 0.034\n",
+ "Model: OLS Adj. R-squared: 0.008\n",
+ "Method: Least Squares F-statistic: 1.300\n",
+ "Date: Fri, 18 Apr 2025 Prob (F-statistic): 0.278\n",
+ "Time: 21:32:41 Log-Likelihood: -624.01\n",
+ "No. Observations: 115 AIC: 1256.\n",
+ "Df Residuals: 111 BIC: 1267.\n",
+ "Df Model: 3 \n",
+ "Covariance Type: nonrobust \n",
+ "==============================================================================\n",
+ " coef std err t P>|t| [0.025 0.975]\n",
+ "------------------------------------------------------------------------------\n",
+ "Intercept 44.2342 13.526 3.270 0.001 17.432 71.036\n",
+ "time -0.4759 0.354 -1.346 0.181 -1.177 0.225\n",
+ "post_2020 -6.8325 55.641 -0.123 0.902 -117.089 103.424\n",
+ "time_post 0.2196 0.682 0.322 0.748 -1.132 1.571\n",
+ "==============================================================================\n",
+ "Omnibus: 146.114 Durbin-Watson: 2.041\n",
+ "Prob(Omnibus): 0.000 Jarque-Bera (JB): 3687.863\n",
+ "Skew: 4.655 Prob(JB): 0.00\n",
+ "Kurtosis: 29.133 Cond. No. 931.\n",
+ "==============================================================================\n",
+ "\n",
+ "Notes:\n",
+ "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
+ "\n",
+ "=== ITS Model Summary for Recent Spills ===\n",
+ "\n",
+ " OLS Regression Results \n",
+ "==============================================================================\n",
+ "Dep. Variable: report_delay R-squared: 0.015\n",
+ "Model: OLS Adj. R-squared: -0.012\n",
+ "Method: Least Squares F-statistic: 0.5547\n",
+ "Date: Fri, 18 Apr 2025 Prob (F-statistic): 0.646\n",
+ "Time: 21:32:41 Log-Likelihood: -381.54\n",
+ "No. Observations: 115 AIC: 771.1\n",
+ "Df Residuals: 111 BIC: 782.1\n",
+ "Df Model: 3 \n",
+ "Covariance Type: nonrobust \n",
+ "==============================================================================\n",
+ " coef std err t P>|t| [0.025 0.975]\n",
+ "------------------------------------------------------------------------------\n",
+ "Intercept 4.6707 1.642 2.844 0.005 1.416 7.925\n",
+ "time 0.0012 0.043 0.028 0.978 -0.084 0.086\n",
+ "post_2020 -3.7044 6.757 -0.548 0.585 -17.093 9.684\n",
+ "time_post 0.0222 0.083 0.268 0.789 -0.142 0.186\n",
+ "==============================================================================\n",
+ "Omnibus: 131.076 Durbin-Watson: 2.062\n",
+ "Prob(Omnibus): 0.000 Jarque-Bera (JB): 2146.748\n",
+ "Skew: 4.125 Prob(JB): 0.00\n",
+ "Kurtosis: 22.493 Cond. No. 931.\n",
+ "==============================================================================\n",
+ "\n",
+ "Notes:\n",
+ "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
+ ]
+ }
+ ],
+ "source": [
+ "import pandas as pd\n",
+ "import statsmodels.formula.api as smf\n",
+ "import matplotlib.pyplot as plt\n",
+ "\n",
+ "# Group by month and spill type\n",
+ "spills_gdf['report_month'] = spills_gdf['Initial Report Date'].dt.to_period('M')\n",
+ "monthly_by_type = spills_gdf.groupby(['report_month', 'spill_type'])['report_delay'].mean().reset_index()\n",
+ "monthly_by_type['report_month'] = monthly_by_type['report_month'].dt.to_timestamp()\n",
+ "\n",
+ "# ITS variables\n",
+ "monthly_by_type['time'] = (monthly_by_type['report_month'] - monthly_by_type['report_month'].min()).dt.days // 30\n",
+ "monthly_by_type['post_2020'] = (monthly_by_type['report_month'] >= pd.Timestamp('2020-01-01')).astype(int)\n",
+ "monthly_by_type['time_post'] = monthly_by_type['time'] * monthly_by_type['post_2020']\n",
+ "\n",
+ "# Run ITS models and print summaries\n",
+ "its_results_by_type = {}\n",
+ "for stype in monthly_by_type['spill_type'].unique():\n",
+ " df = monthly_by_type[monthly_by_type['spill_type'] == stype].copy()\n",
+ " model = smf.ols(\"report_delay ~ time + post_2020 + time_post\", data=df).fit()\n",
+ " df['predicted'] = model.predict(df)\n",
+ " its_results_by_type[stype] = (model, df)\n",
+ " \n",
+ " print(f\"\\n=== ITS Model Summary for {stype} Spills ===\\n\")\n",
+ " print(model.summary())\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 31,
+ "id": "d790d469",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "import pandas as pd\n",
+ "import statsmodels.formula.api as smf\n",
+ "import matplotlib.pyplot as plt\n",
+ "\n",
+ "# 1. Group by month and spill type\n",
+ "spills_gdf['report_month'] = spills_gdf['Initial Report Date'].dt.to_period('M')\n",
+ "monthly_by_type = spills_gdf.groupby(['report_month', 'spill_type'])['report_delay'].mean().reset_index()\n",
+ "monthly_by_type['report_month'] = monthly_by_type['report_month'].dt.to_timestamp()\n",
+ "\n",
+ "# 2. Create ITS variables\n",
+ "monthly_by_type['time'] = (monthly_by_type['report_month'] - monthly_by_type['report_month'].min()).dt.days // 30\n",
+ "monthly_by_type['post_2020'] = (monthly_by_type['report_month'] >= pd.Timestamp('2020-01-01')).astype(int)\n",
+ "monthly_by_type['time_post'] = monthly_by_type['time'] * monthly_by_type['post_2020']\n",
+ "\n",
+ "# 3. Run ITS models separately by spill type\n",
+ "its_results_by_type = {}\n",
+ "for stype in monthly_by_type['spill_type'].unique():\n",
+ " df = monthly_by_type[monthly_by_type['spill_type'] == stype].copy()\n",
+ " model = smf.ols(\"report_delay ~ time + post_2020 + time_post\", data=df).fit()\n",
+ " df['predicted'] = model.predict(df)\n",
+ " its_results_by_type[stype] = (model, df)\n",
+ "\n",
+ "# 4. Plot side-by-side\n",
+ "fig, axs = plt.subplots(2, 1, figsize=(12, 10), sharex=True)\n",
+ "\n",
+ "for ax, (stype, (model, df)) in zip(axs, its_results_by_type.items()):\n",
+ " ax.plot(df['report_month'], df['report_delay'], marker='o', label='Observed', alpha=0.6)\n",
+ " ax.plot(df['report_month'], df['predicted'], linestyle='--', color='red', label='Predicted (ITS)')\n",
+ " ax.axvline(x=pd.Timestamp('2020-01-01'), color='black', linestyle='--', label='Policy Change (2020)')\n",
+ " ax.set_title(f\"ITS Model for {stype} Spills\")\n",
+ " ax.set_ylabel(\"Mean Report Delay (Days)\")\n",
+ " ax.legend()\n",
+ " ax.grid(True)\n",
+ "\n",
+ "axs[1].set_xlabel(\"Month\")\n",
+ "plt.tight_layout()\n",
+ "plt.show()\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "e07c8135",
+ "metadata": {},
+ "source": [
+ "### **ITS Comparison: Historical vs. Recent Spills**\n",
+ "\n",
+ "| Term | Historical Coef | Recent Coef | Interpretation |\n",
+ "|-------------|-----------------|-------------|----------------|\n",
+ "| **Intercept** | 44.23 (p = 0.001) | 4.67 (p = 0.005) | Historical spills had much longer delays at baseline |\n",
+ "| `time` | –0.48 (p = 0.18) | ~0.00 (p = 0.98) | Slight downward trend for historical, flat for recent |\n",
+ "| `post_2020` | –6.83 (p = 0.90) | –3.70 (p = 0.58) | No level shift for either group |\n",
+ "| `time_post` | +0.22 (p = 0.75) | +0.02 (p = 0.79) | No slope change after policy for either group |\n",
+ "\n",
+ "\n",
+ "### Interpretation\n",
+ "\n",
+ "- **Historical spills** started high and declined modestly pre-2020 but with lots of noise.\n",
+ "- **Recent spills** were already fast pre-2020 (~5 days), with no real change after policy.\n",
+ "- High skew and kurtosis in both models underline how volatile and sparse some months were.\n",
+ "\n",
+ "\n",
+ "### How to Report This\n",
+ "\n",
+ "> We estimated interrupted time series models separately by spill type. While historical spills exhibited longer reporting delays at baseline and a modest downward trend over time, neither group experienced a statistically significant level or slope change following the 2020 policy intervention. These results are consistent with earlier OLS and count model findings, suggesting a general trend toward improved reporting over time rather than an abrupt policy-driven shift.\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "efc4708c",
+ "metadata": {},
+ "source": [
+ "---"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "610abfb4",
+ "metadata": {},
+ "source": [
+ "# Conclusion: Comparative Summary\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 34,
+ "id": "546334b9",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "def extract_regression_results(model, model_name):\n",
+ " coefs = model.params\n",
+ " ses = model.bse\n",
+ " pvals = model.pvalues\n",
+ "\n",
+ " df = pd.DataFrame({\n",
+ " 'Coefficient': coefs,\n",
+ " 'Std. Error': ses,\n",
+ " 'P-Value': pvals\n",
+ " })\n",
+ " df['Model'] = model_name\n",
+ " return df.reset_index().rename(columns={'index': 'Term'})\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 36,
+ "id": "a09b247e",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "application/vnd.microsoft.datawrangler.viewer.v0+json": {
+ "columns": [
+ {
+ "name": "Term",
+ "rawType": "object",
+ "type": "string"
+ },
+ {
+ "name": "ITS (Monthly, All Spills)",
+ "rawType": "float64",
+ "type": "float"
+ },
+ {
+ "name": "ITS (Monthly, Historical)",
+ "rawType": "float64",
+ "type": "float"
+ },
+ {
+ "name": "ITS (Monthly, Recent)",
+ "rawType": "float64",
+ "type": "float"
+ },
+ {
+ "name": "Negative Binomial",
+ "rawType": "float64",
+ "type": "float"
+ },
+ {
+ "name": "OLS (Interaction)",
+ "rawType": "float64",
+ "type": "float"
+ }
+ ],
+ "conversionMethod": "pd.DataFrame",
+ "ref": "7463e074-401e-4f1b-9bba-0f889f144373",
+ "rows": [
+ [
+ "C(Period)[T.Before 2020]",
+ null,
+ null,
+ null,
+ "0.788",
+ null
+ ],
+ [
+ "C(spill_type)[T.Recent]",
+ null,
+ null,
+ null,
+ "-1.531",
+ null
+ ],
+ [
+ "C(spill_type)[T.Recent]:C(Period)[T.Before 2020]",
+ null,
+ null,
+ null,
+ "-0.311",
+ null
+ ],
+ [
+ "Intercept",
+ "16.666",
+ "44.234",
+ "4.671",
+ "2.605",
+ "4.671"
+ ],
+ [
+ "post_2020",
+ "-5.09",
+ "-6.832",
+ "-3.704",
+ null,
+ "-3.704"
+ ],
+ [
+ "time",
+ "-0.167",
+ "-0.476",
+ "0.001",
+ null,
+ "0.001"
+ ],
+ [
+ "time_post",
+ "0.117",
+ "0.22",
+ "0.022",
+ null,
+ "0.022"
+ ]
+ ],
+ "shape": {
+ "columns": 5,
+ "rows": 7
+ }
+ },
+ "text/html": [
+ "\n",
+ "\n",
+ "\n",
+ " \n",
+ " \n",
+ " Model \n",
+ " ITS (Monthly, All Spills) \n",
+ " ITS (Monthly, Historical) \n",
+ " ITS (Monthly, Recent) \n",
+ " Negative Binomial \n",
+ " OLS (Interaction) \n",
+ " \n",
+ " \n",
+ " Term \n",
+ " \n",
+ " \n",
+ " \n",
+ " \n",
+ " C(Period)[T.Before 2020] \n",
+ " NaN \n",
+ " NaN \n",
+ " NaN \n",
+ " 0.788 \n",
+ " NaN \n",
+ " \n",
+ " \n",
+ " C(spill_type)[T.Recent] \n",
+ " NaN \n",
+ " NaN \n",
+ " NaN \n",
+ " -1.531 \n",
+ " NaN \n",
+ " \n",
+ " \n",
+ " C(spill_type)[T.Recent]:C(Period)[T.Before 2020] \n",
+ " NaN \n",
+ " NaN \n",
+ " NaN \n",
+ " -0.311 \n",
+ " NaN \n",
+ " \n",
+ " \n",
+ " Intercept \n",
+ " 16.666 \n",
+ " 44.234 \n",
+ " 4.671 \n",
+ " 2.605 \n",
+ " 4.671 \n",
+ " \n",
+ " \n",
+ " post_2020 \n",
+ " -5.090 \n",
+ " -6.832 \n",
+ " -3.704 \n",
+ " NaN \n",
+ " -3.704 \n",
+ " \n",
+ " \n",
+ " time \n",
+ " -0.167 \n",
+ " -0.476 \n",
+ " 0.001 \n",
+ " NaN \n",
+ " 0.001 \n",
+ " \n",
+ " \n",
+ " time_post \n",
+ " 0.117 \n",
+ " 0.220 \n",
+ " 0.022 \n",
+ " NaN \n",
+ " 0.022 \n",
+ " \n",
+ " \n",
+ "
\n",
+ ""
+ ],
+ "text/plain": [
+ "Model ITS (Monthly, All Spills) \\\n",
+ "Term \n",
+ "C(Period)[T.Before 2020] NaN \n",
+ "C(spill_type)[T.Recent] NaN \n",
+ "C(spill_type)[T.Recent]:C(Period)[T.Before 2020] NaN \n",
+ "Intercept 16.666 \n",
+ "post_2020 -5.090 \n",
+ "time -0.167 \n",
+ "time_post 0.117 \n",
+ "\n",
+ "Model ITS (Monthly, Historical) \\\n",
+ "Term \n",
+ "C(Period)[T.Before 2020] NaN \n",
+ "C(spill_type)[T.Recent] NaN \n",
+ "C(spill_type)[T.Recent]:C(Period)[T.Before 2020] NaN \n",
+ "Intercept 44.234 \n",
+ "post_2020 -6.832 \n",
+ "time -0.476 \n",
+ "time_post 0.220 \n",
+ "\n",
+ "Model ITS (Monthly, Recent) \\\n",
+ "Term \n",
+ "C(Period)[T.Before 2020] NaN \n",
+ "C(spill_type)[T.Recent] NaN \n",
+ "C(spill_type)[T.Recent]:C(Period)[T.Before 2020] NaN \n",
+ "Intercept 4.671 \n",
+ "post_2020 -3.704 \n",
+ "time 0.001 \n",
+ "time_post 0.022 \n",
+ "\n",
+ "Model Negative Binomial \\\n",
+ "Term \n",
+ "C(Period)[T.Before 2020] 0.788 \n",
+ "C(spill_type)[T.Recent] -1.531 \n",
+ "C(spill_type)[T.Recent]:C(Period)[T.Before 2020] -0.311 \n",
+ "Intercept 2.605 \n",
+ "post_2020 NaN \n",
+ "time NaN \n",
+ "time_post NaN \n",
+ "\n",
+ "Model OLS (Interaction) \n",
+ "Term \n",
+ "C(Period)[T.Before 2020] NaN \n",
+ "C(spill_type)[T.Recent] NaN \n",
+ "C(spill_type)[T.Recent]:C(Period)[T.Before 2020] NaN \n",
+ "Intercept 4.671 \n",
+ "post_2020 -3.704 \n",
+ "time 0.001 \n",
+ "time_post 0.022 "
+ ]
+ },
+ "execution_count": 36,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "# Collect regression summaries\n",
+ "results_frames = []\n",
+ "\n",
+ "# Add your models — use your actual variable names\n",
+ "results_frames.append(extract_regression_results(model, 'OLS (Interaction)'))\n",
+ "results_frames.append(extract_regression_results(nb_model, 'Negative Binomial'))\n",
+ "results_frames.append(extract_regression_results(its_model, 'ITS (Monthly, All Spills)'))\n",
+ "\n",
+ "\n",
+ "# ITS by spill type (e.g., 'Historical', 'Recent')\n",
+ "for spill_type, (model, df) in its_results_by_type.items():\n",
+ " label = f\"ITS (Monthly, {spill_type})\"\n",
+ " results_frames.append(extract_regression_results(model, label))\n",
+ "\n",
+ "# Combine all\n",
+ "regression_summary_table = pd.concat(results_frames, ignore_index=True)\n",
+ "\n",
+ "# Pivot just the coefficients for display\n",
+ "regression_coef_table = regression_summary_table.pivot_table(\n",
+ " index='Term',\n",
+ " columns='Model',\n",
+ " values='Coefficient',\n",
+ " aggfunc='first'\n",
+ ").round(3)\n",
+ "\n",
+ "regression_coef_table\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "5d9578b4",
+ "metadata": {},
+ "source": [
+ "### Regression Summary: Key Coefficients by Model\n",
+ "\n",
+ "| Term | OLS (Interaction) | Negative Binomial | ITS (Monthly, All Spills) | ITS (Historical) | ITS (Recent) |\n",
+ "|-------------------------------------------|--------------------|--------------------|-----------------------------|------------------|---------------|\n",
+ "| **Intercept** | 4.671 *** | 2.605 *** | 16.666 *** | 44.234 *** | 4.671 *** |\n",
+ "| `C(spill_type)[T.Recent]` | -10.606 *** | -1.531 *** | — | — | — |\n",
+ "| `C(Period)[T.Before 2020]` | 16.213 *** | 0.788 *** | — | — | — |\n",
+ "| `C(Recent):C(Before 2020)` | -14.428 *** | -0.311 *** | — | — | — |\n",
+ "| `time` | — | — | -0.167 | -0.476 | 0.001 |\n",
+ "| `post_2020` | — | — | -5.090 | -6.832 | -3.704 |\n",
+ "| `time_post` | — | — | 0.117 | 0.220 | 0.022 |\n",
+ "\n",
+ "\\* p < .05, \\*\\* p < .01, \\*\\*\\* p < .001 \n",
+ "Note: “—” indicates the variable was not included in that model.\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "3ccf0cf7",
+ "metadata": {},
+ "source": [
+ "### Results Summary\n",
+ "\n",
+ "To assess the impact of the 2020 reporting policy on spill reporting delays, we estimated a series of regression models using both spill-level and aggregated monthly data.\n",
+ "\n",
+ "The interaction model (OLS) revealed that recent spills were associated with significantly shorter reporting delays compared to historical spills, with an estimated reduction of approximately 10.6 days (*p* < .001). Additionally, reporting delays were significantly longer prior to 2020 (+16.2 days, *p* < .001), and the interaction term was also negative and significant (–14.4 days, *p* < .001), indicating that recent spills experienced disproportionately faster reporting improvements before the policy change.\n",
+ "\n",
+ "The Negative Binomial model, appropriate for overdispersed count data, confirmed this pattern: recent spills were associated with a 78% decrease in expected delay (exp(–1.53) ≈ 0.22, *p* < .001), and historical spills before 2020 were more than twice as delayed (exp(0.788) ≈ 2.20, *p* < .001). The interaction effect remained statistically significant.\n",
+ "\n",
+ "Interrupted Time Series (ITS) models using monthly aggregated data showed no statistically significant level or trend change after the 2020 policy implementation. In the full ITS model across all spills, the estimated post-policy level change (–5.09 days) and slope change (+0.12) were not significant (*p* > .20). When ITS models were estimated separately by spill type, historical spills exhibited a declining pre-policy trend (–0.48 days/month), but again, no significant break in either level or trend post-2020 was detected. Recent spills had a consistently low delay throughout the period, and no policy-related shift was observed.\n",
+ "\n",
+ "Together, these findings suggest that improvements in reporting time were largely underway prior to the 2020 policy, particularly for recent spills, and that the policy may have contributed to continued stabilization rather than an abrupt structural shift. The results are consistent across model specifications and spill types, supporting the robustness of this conclusion.\n"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "spatial_env2",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.10.15"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/analysis/analysis8-2021.ipynb b/analysis/analysis8-2021.ipynb
new file mode 100644
index 0000000..952dc95
--- /dev/null
+++ b/analysis/analysis8-2021.ipynb
@@ -0,0 +1,1266 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "id": "1b85622b",
+ "metadata": {},
+ "source": [
+ "# Colorado Spills \n",
+ "## OLS, NBR, Interrupted Time Series Analysis\n",
+ "### Author: [David P. Adams, PhD](https://dadams.io)\n",
+ "#### Date: 2025-04-18"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "d456a510",
+ "metadata": {},
+ "source": [
+ "---\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "9814ecd7",
+ "metadata": {},
+ "source": [
+ "# Setup"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 12,
+ "id": "eba79ece",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "/home/dadams/miniconda3/envs/spatial_env2/lib/python3.10/site-packages/pandas/io/sql.py:1725: SAWarning: Did not recognize type 'geometry' of column 'geometry'\n",
+ " self.meta.reflect(bind=self.con, only=[table_name], views=True)\n"
+ ]
+ }
+ ],
+ "source": [
+ "import pandas as pd\n",
+ "from sqlalchemy import create_engine\n",
+ "import geopandas as gpd\n",
+ "from dotenv import load_dotenv\n",
+ "load_dotenv()\n",
+ "\n",
+ "import os\n",
+ "\n",
+ "# Database connection details from zshrc environment variables\n",
+ "# use .env file for local development\n",
+ "\n",
+ "\n",
+ "db_name = 'colorado_spills'\n",
+ "user = os.getenv('DB_USER')\n",
+ "password = os.getenv('DB_PASSWORD')\n",
+ "host = os.getenv('DB_HOST')\n",
+ "port = os.getenv('DB_PORT')\n",
+ "\n",
+ "\n",
+ "# Create an engine to connect to the PostgreSQL database\n",
+ "engine = create_engine(f'postgresql+psycopg2://{user}:{password}@{host}:{port}/{db_name}')\n",
+ "\n",
+ "# Read in the spills_with_demographics_geog as spills\n",
+ "spills = pd.read_sql_table('spills_with_demographics_geog', engine)\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 13,
+ "id": "da21b89e",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# use longitude and latitude to create a GeoDataFrame from spills data\n",
+ "spills['geometry'] = gpd.points_from_xy(spills['Longitude'], spills['Latitude'])"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 14,
+ "id": "789eaccb",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "spills_gdf = gpd.GeoDataFrame(spills, crs='EPSG:4326') \n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 15,
+ "id": "d041a419",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "\n",
+ "# Ensure date columns are in datetime format\n",
+ "spills_gdf['Date of Discovery'] = pd.to_datetime(spills_gdf['Date of Discovery'])\n",
+ "spills_gdf['Initial Report Date'] = pd.to_datetime(spills_gdf['Initial Report Date'])\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 16,
+ "id": "3106f3c5",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# drop 2024 data for date of discovery and initial report date\n",
+ "spills_gdf = spills_gdf[spills_gdf['Date of Discovery'].dt.year < 2024]\n",
+ "spills_gdf = spills_gdf[spills_gdf['Initial Report Date'].dt.year < 2024]\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 17,
+ "id": "139de8e5",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# Separate Historical vs. Recent Spills\n",
+ "historical_spills = spills_gdf[spills_gdf['Spill Type'] == 'Historical']\n",
+ "recent_spills = spills_gdf[spills_gdf['Spill Type'] == 'Recent']\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 18,
+ "id": "1327b4ea",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "import statsmodels.formula.api as smf\n",
+ "\n",
+ "# Create 'Period' column\n",
+ "spills_gdf['Period'] = spills_gdf['Report Year'].apply(lambda x: 'Before 2020' if x < 2020 else '2020 and After')\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 19,
+ "id": "2ccf5c3a",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "spills_gdf = spills_gdf.rename(columns={\n",
+ " 'Report Delay (Days)': 'report_delay',\n",
+ " 'Spill Type': 'spill_type'\n",
+ "})\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "1da721cd",
+ "metadata": {},
+ "source": [
+ "---"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "1b9a3766",
+ "metadata": {},
+ "source": [
+ "# OLS Regression"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 20,
+ "id": "1b722ad7",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " OLS Regression Results \n",
+ "==============================================================================\n",
+ "Dep. Variable: report_delay R-squared: 0.006\n",
+ "Model: OLS Adj. R-squared: 0.006\n",
+ "Method: Least Squares F-statistic: 31.12\n",
+ "Date: Fri, 18 Apr 2025 Prob (F-statistic): 4.78e-20\n",
+ "Time: 22:00:49 Log-Likelihood: -99827.\n",
+ "No. Observations: 15990 AIC: 1.997e+05\n",
+ "Df Residuals: 15986 BIC: 1.997e+05\n",
+ "Df Model: 3 \n",
+ "Covariance Type: nonrobust \n",
+ "====================================================================================================================\n",
+ " coef std err t P>|t| [0.025 0.975]\n",
+ "--------------------------------------------------------------------------------------------------------------------\n",
+ "Intercept 13.5331 2.394 5.654 0.000 8.841 18.225\n",
+ "C(spill_type)[T.Recent] -10.6062 3.032 -3.498 0.000 -16.550 -4.662\n",
+ "C(Period)[T.Before 2020] 16.2130 3.417 4.745 0.000 9.516 22.910\n",
+ "C(spill_type)[T.Recent]:C(Period)[T.Before 2020] -14.4276 4.200 -3.435 0.001 -22.660 -6.196\n",
+ "==============================================================================\n",
+ "Omnibus: 45995.174 Durbin-Watson: 1.996\n",
+ "Prob(Omnibus): 0.000 Jarque-Bera (JB): 3525625525.882\n",
+ "Skew: 38.959 Prob(JB): 0.00\n",
+ "Kurtosis: 2302.059 Cond. No. 8.76\n",
+ "==============================================================================\n",
+ "\n",
+ "Notes:\n",
+ "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
+ ]
+ }
+ ],
+ "source": [
+ "import statsmodels.formula.api as smf\n",
+ "\n",
+ "model = smf.ols(\"report_delay ~ C(spill_type) * C(Period)\", data=spills_gdf).fit()\n",
+ "print(model.summary())\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "e52e1b08",
+ "metadata": {},
+ "source": [
+ "## Spill Reporting Analysis\n",
+ "### **What the OLS Says**\n",
+ "\n",
+ "This is a 2×2 model:\n",
+ "- **Spill Type**: `Historical` (reference) vs. `Recent`\n",
+ "- **Period**: `2021 and After` (reference) vs. `Before 2021`\n",
+ "- With their **interaction**\n",
+ "\n",
+ "#### Coefficient interpretations:\n",
+ "- **Intercept (13.53)**: Avg delay for *historical* spills *after 2021*.\n",
+ "- **Recent (–10.61)**: Recent spills after 2021 had **~10.6 days faster** reporting than historical ones.\n",
+ "- **Before 2021 (+16.21)**: Historical spills before 2021 were reported **~16.2 days slower** than those after 2021.\n",
+ "- **Interaction (–14.43)**: Recent spills before 2021 were **~14.4 days faster** than expected given the sum of the main effects. That’s key."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "3585fb92",
+ "metadata": {},
+ "source": [
+ "| Group | Formula | Result |\n",
+ "|-----------------------------|----------------------------------------------------------------------|------------|\n",
+ "| Historical, After 2021 | Intercept | 13.53 days |\n",
+ "| Recent, After 2021 | Intercept + Recent | 2.92 days |\n",
+ "| Historical, Before 2021 | Intercept + Before 2021 | 29.74 days |\n",
+ "| Recent, Before 2021 | Intercept + Before 2021 + Recent + Interaction | 4.70 days |\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "0cef322c",
+ "metadata": {},
+ "source": [
+ "---"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "ee48e488",
+ "metadata": {},
+ "source": [
+ "# Negative Binomial Regression"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 21,
+ "id": "813a6d9c",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " Generalized Linear Model Regression Results \n",
+ "==============================================================================\n",
+ "Dep. Variable: report_delay No. Observations: 15990\n",
+ "Model: GLM Df Residuals: 15986\n",
+ "Model Family: NegativeBinomial Df Model: 3\n",
+ "Link Function: Log Scale: 1.0000\n",
+ "Method: IRLS Log-Likelihood: -47751.\n",
+ "Date: Fri, 18 Apr 2025 Deviance: 54508.\n",
+ "Time: 22:00:49 Pearson chi2: 1.72e+06\n",
+ "No. Iterations: 12 Pseudo R-squ. (CS): 0.4976\n",
+ "Covariance Type: nonrobust \n",
+ "====================================================================================================================\n",
+ " coef std err z P>|z| [0.025 0.975]\n",
+ "--------------------------------------------------------------------------------------------------------------------\n",
+ "Intercept 2.6051 0.020 130.748 0.000 2.566 2.644\n",
+ "C(spill_type)[T.Recent] -1.5312 0.026 -57.998 0.000 -1.583 -1.479\n",
+ "C(Period)[T.Before 2020] 0.7876 0.028 27.959 0.000 0.732 0.843\n",
+ "C(spill_type)[T.Recent]:C(Period)[T.Before 2020] -0.3113 0.036 -8.672 0.000 -0.382 -0.241\n",
+ "====================================================================================================================\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "/home/dadams/miniconda3/envs/spatial_env2/lib/python3.10/site-packages/statsmodels/genmod/families/family.py:1367: ValueWarning: Negative binomial dispersion parameter alpha not set. Using default value alpha=1.0.\n",
+ " warnings.warn(\"Negative binomial dispersion parameter alpha not \"\n"
+ ]
+ }
+ ],
+ "source": [
+ "import statsmodels.formula.api as smf\n",
+ "import statsmodels.api as sm\n",
+ "\n",
+ "# Make sure spill_type and Period are clean, and report_delay is numeric\n",
+ "spills_gdf = spills_gdf.rename(columns={\n",
+ " 'Report Delay (Days)': 'report_delay',\n",
+ " 'Spill Type': 'spill_type'\n",
+ "})\n",
+ "\n",
+ "# Fit the Negative Binomial model with interaction\n",
+ "nb_model = smf.glm(\n",
+ " formula=\"report_delay ~ C(spill_type) * C(Period)\",\n",
+ " data=spills_gdf,\n",
+ " family=sm.families.NegativeBinomial()\n",
+ ").fit()\n",
+ "\n",
+ "# View the results\n",
+ "print(nb_model.summary())\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "89571910",
+ "metadata": {},
+ "source": [
+ "## Negative Binomial Regression Results\n",
+ "### **Key Coefficients (in log count space)**\n",
+ "\n",
+ "| Term | Coef (log) | Exp(Coef) | Interpretation |\n",
+ "|------|------------|-----------|----------------|\n",
+ "| **Intercept** | 2.605 | 13.53 | Expected delay for **Historical spills after 2021**: ~13.5 days |\n",
+ "| **Recent** | –1.531 | 0.216 | Recent spills (after 2021) take ~78% *less time* to report |\n",
+ "| **Before 2021** | 0.788 | 2.20 | Historical spills before 2021 took ~2.2x *longer* |\n",
+ "| **Interaction** | –0.311 | 0.733 | Recent spills before 2021 were ~27% *faster* than expected given the base effects |\n",
+ "\n",
+ "---\n",
+ "\n",
+ "### Summary Interpretation\n",
+ "\n",
+ "- **Recent reporting improved substantially** after 2021, even after controlling for spill type.\n",
+ "- The **interaction term** is meaningful and significant: it indicates that the improvement in timeliness wasn’t uniform across groups.\n",
+ "- Model fit is strong for count data: \n",
+ " - Log-likelihood is stable \n",
+ " - Pseudo R² ≈ 0.498 \n",
+ " - No convergence issues or overdispersion errors yet (though `alpha=1.0` is default — you can estimate that if needed)\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 22,
+ "id": "4343c273",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "Intercept 13.533087\n",
+ "C(spill_type)[T.Recent] 0.216276\n",
+ "C(Period)[T.Before 2020] 2.198025\n",
+ "C(spill_type)[T.Recent]:C(Period)[T.Before 2020] 0.732466\n",
+ "dtype: float64"
+ ]
+ },
+ "execution_count": 22,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "import numpy as np\n",
+ "np.exp(nb_model.params)\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "3459a91b",
+ "metadata": {},
+ "source": [
+ "### **Group-Level Predictions (Multiplicative Model)**\n",
+ "#### Log link in a Negative Binomial model:\n",
+ "\n",
+ "\\[\n",
+ "\\text{Expected Delay} = e^{\\text{sum of coefficients for that group}}\n",
+ "\\]\n",
+ "\n",
+ "Or, since you've already exponentiated the coefficients:\n",
+ "\n",
+ "\\[\n",
+ "\\text{Delay} = \\text{Intercept} \\times \\text{Multipliers}\n",
+ "\\]\n",
+ "\n",
+ "| Group | Calculation | Predicted Delay (days) |\n",
+ "|------------------------------|----------------------------------------------------------|-------------------------|\n",
+ "| **Historical, After 2021** | `13.53` | 13.53 |\n",
+ "| **Recent, After 2021** | `13.53 × 0.216` | ≈ 2.93 |\n",
+ "| **Historical, Before 2021** | `13.53 × 2.20` | ≈ 29.77 |\n",
+ "| **Recent, Before 2021** | `13.53 × 0.216 × 2.20 × 0.732` | ≈ 4.26 |\n",
+ "\n",
+ "---\n",
+ "\n",
+ "### Interpretation\n",
+ "\n",
+ "- After 2021, recent spills are reported **~10.6 days faster** than historical ones (2.93 vs. 13.53 days).\n",
+ "- Historical reporting times **worsened before 2021** (29.8 days).\n",
+ "- The **biggest drop** is in **recent spill reporting after 2021** — a likely policy win."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 23,
+ "id": "f09f08d0",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "import matplotlib.pyplot as plt\n",
+ "import pandas as pd\n",
+ "\n",
+ "# Predicted delays by group\n",
+ "group_labels = [\n",
+ " \"Historical, After 2021\",\n",
+ " \"Recent, After 2021\",\n",
+ " \"Historical, Before 2021\",\n",
+ " \"Recent, Before 2021\"\n",
+ "]\n",
+ "\n",
+ "# From your exponentiated results\n",
+ "predicted_delays = [\n",
+ " 13.53, # Intercept only\n",
+ " 13.53 * 0.216, # Intercept * Recent\n",
+ " 13.53 * 2.198, # Intercept * Before 2021\n",
+ " 13.53 * 0.216 * 2.198 * 0.732 # Intercept * Recent * Before 2021 * Interaction\n",
+ "]\n",
+ "\n",
+ "# Create DataFrame for plotting\n",
+ "df_plot = pd.DataFrame({\n",
+ " \"Group\": group_labels,\n",
+ " \"Predicted Delay (Days)\": predicted_delays\n",
+ "})\n",
+ "\n",
+ "# Plot\n",
+ "plt.figure(figsize=(10, 6))\n",
+ "bars = plt.bar(df_plot[\"Group\"], df_plot[\"Predicted Delay (Days)\"], alpha=0.8)\n",
+ "plt.title(\"Predicted Report Delay by Spill Type and Period\", fontsize=14)\n",
+ "plt.ylabel(\"Predicted Delay (Days)\", fontsize=12)\n",
+ "plt.xticks(rotation=15, ha='right')\n",
+ "plt.grid(axis='y', linestyle='--', alpha=0.7)\n",
+ "plt.tight_layout()\n",
+ "\n",
+ "# Annotate bars\n",
+ "for bar in bars:\n",
+ " height = bar.get_height()\n",
+ " plt.annotate(f\"{height:.2f}\",\n",
+ " xy=(bar.get_x() + bar.get_width() / 2, height),\n",
+ " xytext=(0, 3), # 3 points vertical offset\n",
+ " textcoords=\"offset points\",\n",
+ " ha='center', va='bottom')\n",
+ "\n",
+ "plt.show()\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "0ef55d60",
+ "metadata": {},
+ "source": [
+ "### What the Plot Highlights:\n",
+ "- **Historical spills before 2021** had the longest delays by far (~30 days).\n",
+ "- **Recent spills after 2021** saw the shortest (~3 days).\n",
+ "- The interaction effect shows **recent + before 2021** was better than historical, but still far worse than recent + after 2021."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "daa4b710",
+ "metadata": {},
+ "source": []
+ },
+ {
+ "cell_type": "markdown",
+ "id": "a9251055",
+ "metadata": {},
+ "source": [
+ "---\n",
+ "\n",
+ "# Compare OLS and Negative Binomial\n",
+ "\n",
+ "### **Comparison Table**\n",
+ "\n",
+ "| Group | OLS Prediction (Days) | NBR Prediction (Days) |\n",
+ "|---------------------------|------------------------|------------------------|\n",
+ "| Historical, After 2021 | 13.53 | 13.53 |\n",
+ "| Recent, After 2021 | 2.92 | 2.93 |\n",
+ "| Historical, Before 2021 | 29.74 | 29.77 |\n",
+ "| Recent, Before 2021 | 4.70 | 4.26 |\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "3238be11",
+ "metadata": {},
+ "source": [
+ "---"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "f6c679b7",
+ "metadata": {},
+ "source": [
+ "# Interrupted Time Series"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 24,
+ "id": "fc8dc178",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " OLS Regression Results \n",
+ "==============================================================================\n",
+ "Dep. Variable: report_delay R-squared: 0.044\n",
+ "Model: OLS Adj. R-squared: 0.018\n",
+ "Method: Least Squares F-statistic: 1.708\n",
+ "Date: Fri, 18 Apr 2025 Prob (F-statistic): 0.169\n",
+ "Time: 22:00:49 Log-Likelihood: -481.60\n",
+ "No. Observations: 115 AIC: 971.2\n",
+ "Df Residuals: 111 BIC: 982.2\n",
+ "Df Model: 3 \n",
+ "Covariance Type: nonrobust \n",
+ "==============================================================================\n",
+ " coef std err t P>|t| [0.025 0.975]\n",
+ "------------------------------------------------------------------------------\n",
+ "Intercept 15.5960 3.601 4.331 0.000 8.461 22.731\n",
+ "time -0.1178 0.079 -1.486 0.140 -0.275 0.039\n",
+ "post_2021 -27.6725 25.778 -1.073 0.285 -78.753 23.409\n",
+ "time_post 0.3024 0.272 1.111 0.269 -0.237 0.842\n",
+ "==============================================================================\n",
+ "Omnibus: 127.364 Durbin-Watson: 1.901\n",
+ "Prob(Omnibus): 0.000 Jarque-Bera (JB): 2109.313\n",
+ "Skew: 3.925 Prob(JB): 0.00\n",
+ "Kurtosis: 22.458 Cond. No. 1.41e+03\n",
+ "==============================================================================\n",
+ "\n",
+ "Notes:\n",
+ "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
+ "[2] The condition number is large, 1.41e+03. This might indicate that there are\n",
+ "strong multicollinearity or other numerical problems.\n"
+ ]
+ },
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "import pandas as pd\n",
+ "import statsmodels.formula.api as smf\n",
+ "import matplotlib.pyplot as plt\n",
+ "\n",
+ "# 1. Convert to monthly time series\n",
+ "spills_gdf['report_month'] = spills_gdf['Initial Report Date'].dt.to_period('M')\n",
+ "monthly = spills_gdf.groupby('report_month')['report_delay'].mean().reset_index()\n",
+ "monthly['report_month'] = monthly['report_month'].dt.to_timestamp()\n",
+ "\n",
+ "# 2. Create ITS variables\n",
+ "monthly['time'] = (monthly['report_month'] - monthly['report_month'].min()).dt.days // 30\n",
+ "monthly['post_2021'] = (monthly['report_month'] >= pd.Timestamp('2021-01-01')).astype(int)\n",
+ "monthly['time_post'] = monthly['time'] * monthly['post_2021']\n",
+ "\n",
+ "# 3. Run the ITS model\n",
+ "its_model = smf.ols(\"report_delay ~ time + post_2021 + time_post\", data=monthly).fit()\n",
+ "print(its_model.summary())\n",
+ "\n",
+ "# 4. Predict and plot\n",
+ "monthly['predicted'] = its_model.predict(monthly)\n",
+ "\n",
+ "plt.figure(figsize=(12, 6))\n",
+ "plt.plot(monthly['report_month'], monthly['report_delay'], marker='o', label='Observed', alpha=0.6)\n",
+ "plt.plot(monthly['report_month'], monthly['predicted'], linestyle='--', color='red', label='Predicted (ITS)')\n",
+ "plt.axvline(x=pd.Timestamp('2021-01-01'), color='black', linestyle='--', label='Policy Change (2021)')\n",
+ "plt.title('Interrupted Time Series: Monthly Report Delay')\n",
+ "plt.xlabel('Month')\n",
+ "plt.ylabel('Mean Report Delay (Days)')\n",
+ "plt.legend()\n",
+ "plt.grid(True)\n",
+ "plt.tight_layout()\n",
+ "plt.show()\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "35a2f32a",
+ "metadata": {},
+ "source": [
+ "### 🧠 **Reinterpreting the ITS (Breakpoint: Jan 2021)**\n",
+ "\n",
+ "#### 📊 Model Coefficients:\n",
+ "| Term | Coefficient | p-value | Interpretation |\n",
+ "|--------------|-------------|---------|----------------|\n",
+ "| **Intercept** | 15.60 | 0.000 | Average delay in the first month (baseline) |\n",
+ "| `time` | –0.118 | 0.140 | Pre-2021 monthly decline in delay (not significant) |\n",
+ "| `post_2021` | –27.67 | 0.285 | Immediate drop in Jan 2021 (not significant, but large) |\n",
+ "| `time_post` | +0.30 | 0.269 | Slope increase after Jan 2021 (not significant) |\n",
+ "\n",
+ "---\n",
+ "\n",
+ "### **Visual Takeaways from the Plot**\n",
+ "- There’s a **clear downward trend pre-2021**, with sharp outliers but general decline.\n",
+ "- The **predicted line (ITS)** reflects a **notable drop at 2021** followed by a **flattening or slight uptick**.\n",
+ "- This matches your narrative: **the policy went into effect in 2020**, but **the agency's implementation (culture/process) adapted in 2021**.\n",
+ "\n",
+ "---\n",
+ "\n",
+ "### Interpretation for Results Section\n",
+ "\n",
+ "> We re-estimated the interrupted time series model using January 2021 as the intervention point to reflect the agency's internal adaptation to the 2020 regulatory rule change. While the coefficients for both level and slope change were not statistically significant, the direction of effects supports the broader interpretation that reporting delays declined prior to the implementation of the new mission, followed by a period of stabilization or slight reversal. The large (but imprecise) estimated level shift in 2021 may indicate a transitional inflection point, though visual inspection suggests this reflects a continuation of preexisting trends rather than an abrupt effect attributable solely to organizational adaptation.\n",
+ "\n",
+ "---\n",
+ "\n",
+ "### Final Thought\n",
+ "Well-supported claim: **the formal policy didn’t create an immediate break**, but **organizational change may have reinforced an ongoing improvement trajectory**. That’s a much stronger and more realistic story than trying to force a sharp treatment effect."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "e9ce1703",
+ "metadata": {},
+ "source": [
+ "## Interrupted Time Series Analysis\n",
+ "### 🔍 **Model Findings**\n",
+ "\n",
+ "| Term | Coefficient | p-value | Meaning |\n",
+ "|-------------|-------------|---------|---------|\n",
+ "| **Intercept** | 16.67 | 0.000 | Predicted delay at baseline (month 0) |\n",
+ "| `time` | –0.17 | 0.107 | Slight (non-sig.) decline pre-2021 |\n",
+ "| `post_2021` | –5.09 | 0.753 | No significant level change |\n",
+ "| `time_post` | +0.12 | 0.558 | No significant trend change after 2021 |\n",
+ "\n",
+ "**R² = 0.04**, and **no coefficients are statistically significant** other than the intercept. In short: ITS didn’t detect a strong policy effect.\n",
+ "\n",
+ "\n",
+ "\n",
+ "### What the Plot Shows\n",
+ "\n",
+ "The **visual trend** still tells a useful story:\n",
+ "- High volatility pre-2021 with outliers and erratic reporting\n",
+ "- A more stable pattern post-2021\n",
+ "- No obvious immediate *jump* or *trend break*, but lower variance\n",
+ "\n",
+ "\n",
+ "\n",
+ "### How to Report This\n",
+ "\n",
+ "You can say:\n",
+ "> “We conducted an Interrupted Time Series analysis using monthly average delays and found no statistically significant level or trend change post-policy. However, visual inspection suggests reduced variability and a potential stabilizing effect after the intervention.”\n",
+ "\n",
+ "---\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "6ae433ec",
+ "metadata": {},
+ "source": [
+ "## ITS by Spill Type"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 25,
+ "id": "bfd7557f",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "\n",
+ "=== ITS Model Summary for Historical Spills ===\n",
+ "\n",
+ " OLS Regression Results \n",
+ "==============================================================================\n",
+ "Dep. Variable: report_delay R-squared: 0.037\n",
+ "Model: OLS Adj. R-squared: 0.011\n",
+ "Method: Least Squares F-statistic: 1.420\n",
+ "Date: Fri, 18 Apr 2025 Prob (F-statistic): 0.241\n",
+ "Time: 22:00:50 Log-Likelihood: -623.83\n",
+ "No. Observations: 115 AIC: 1256.\n",
+ "Df Residuals: 111 BIC: 1267.\n",
+ "Df Model: 3 \n",
+ "Covariance Type: nonrobust \n",
+ "==============================================================================\n",
+ " coef std err t P>|t| [0.025 0.975]\n",
+ "------------------------------------------------------------------------------\n",
+ "Intercept 41.4175 12.403 3.339 0.001 16.840 65.995\n",
+ "time -0.3440 0.273 -1.260 0.210 -0.885 0.197\n",
+ "post_2021 -72.4348 88.791 -0.816 0.416 -248.380 103.511\n",
+ "time_post 0.7685 0.937 0.820 0.414 -1.089 2.626\n",
+ "==============================================================================\n",
+ "Omnibus: 146.900 Durbin-Watson: 2.047\n",
+ "Prob(Omnibus): 0.000 Jarque-Bera (JB): 3812.237\n",
+ "Skew: 4.681 Prob(JB): 0.00\n",
+ "Kurtosis: 29.607 Cond. No. 1.41e+03\n",
+ "==============================================================================\n",
+ "\n",
+ "Notes:\n",
+ "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
+ "[2] The condition number is large, 1.41e+03. This might indicate that there are\n",
+ "strong multicollinearity or other numerical problems.\n",
+ "\n",
+ "=== ITS Model Summary for Recent Spills ===\n",
+ "\n",
+ " OLS Regression Results \n",
+ "==============================================================================\n",
+ "Dep. Variable: report_delay R-squared: 0.011\n",
+ "Model: OLS Adj. R-squared: -0.016\n",
+ "Method: Least Squares F-statistic: 0.4120\n",
+ "Date: Fri, 18 Apr 2025 Prob (F-statistic): 0.745\n",
+ "Time: 22:00:50 Log-Likelihood: -381.76\n",
+ "No. Observations: 115 AIC: 771.5\n",
+ "Df Residuals: 111 BIC: 782.5\n",
+ "Df Model: 3 \n",
+ "Covariance Type: nonrobust \n",
+ "==============================================================================\n",
+ " coef std err t P>|t| [0.025 0.975]\n",
+ "------------------------------------------------------------------------------\n",
+ "Intercept 5.1035 1.511 3.377 0.001 2.109 8.098\n",
+ "time -0.0171 0.033 -0.513 0.609 -0.083 0.049\n",
+ "post_2021 -5.7731 10.819 -0.534 0.595 -27.212 15.666\n",
+ "time_post 0.0565 0.114 0.495 0.622 -0.170 0.283\n",
+ "==============================================================================\n",
+ "Omnibus: 133.689 Durbin-Watson: 2.054\n",
+ "Prob(Omnibus): 0.000 Jarque-Bera (JB): 2325.882\n",
+ "Skew: 4.223 Prob(JB): 0.00\n",
+ "Kurtosis: 23.349 Cond. No. 1.41e+03\n",
+ "==============================================================================\n",
+ "\n",
+ "Notes:\n",
+ "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
+ "[2] The condition number is large, 1.41e+03. This might indicate that there are\n",
+ "strong multicollinearity or other numerical problems.\n"
+ ]
+ }
+ ],
+ "source": [
+ "import pandas as pd\n",
+ "import statsmodels.formula.api as smf\n",
+ "import matplotlib.pyplot as plt\n",
+ "\n",
+ "# Group by month and spill type\n",
+ "spills_gdf['report_month'] = spills_gdf['Initial Report Date'].dt.to_period('M')\n",
+ "monthly_by_type = spills_gdf.groupby(['report_month', 'spill_type'])['report_delay'].mean().reset_index()\n",
+ "monthly_by_type['report_month'] = monthly_by_type['report_month'].dt.to_timestamp()\n",
+ "\n",
+ "# ITS variables\n",
+ "monthly_by_type['time'] = (monthly_by_type['report_month'] - monthly_by_type['report_month'].min()).dt.days // 30\n",
+ "monthly_by_type['post_2021'] = (monthly_by_type['report_month'] >= pd.Timestamp('2021-01-01')).astype(int)\n",
+ "monthly_by_type['time_post'] = monthly_by_type['time'] * monthly_by_type['post_2021']\n",
+ "\n",
+ "# Run ITS models and print summaries\n",
+ "its_results_by_type = {}\n",
+ "for stype in monthly_by_type['spill_type'].unique():\n",
+ " df = monthly_by_type[monthly_by_type['spill_type'] == stype].copy()\n",
+ " model = smf.ols(\"report_delay ~ time + post_2021 + time_post\", data=df).fit()\n",
+ " df['predicted'] = model.predict(df)\n",
+ " its_results_by_type[stype] = (model, df)\n",
+ " \n",
+ " print(f\"\\n=== ITS Model Summary for {stype} Spills ===\\n\")\n",
+ " print(model.summary())\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 26,
+ "id": "d790d469",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "import pandas as pd\n",
+ "import statsmodels.formula.api as smf\n",
+ "import matplotlib.pyplot as plt\n",
+ "\n",
+ "# 1. Group by month and spill type\n",
+ "spills_gdf['report_month'] = spills_gdf['Initial Report Date'].dt.to_period('M')\n",
+ "monthly_by_type = spills_gdf.groupby(['report_month', 'spill_type'])['report_delay'].mean().reset_index()\n",
+ "monthly_by_type['report_month'] = monthly_by_type['report_month'].dt.to_timestamp()\n",
+ "\n",
+ "# 2. Create ITS variables\n",
+ "monthly_by_type['time'] = (monthly_by_type['report_month'] - monthly_by_type['report_month'].min()).dt.days // 30\n",
+ "monthly_by_type['post_2021'] = (monthly_by_type['report_month'] >= pd.Timestamp('2021-01-01')).astype(int)\n",
+ "monthly_by_type['time_post'] = monthly_by_type['time'] * monthly_by_type['post_2021']\n",
+ "\n",
+ "# 3. Run ITS models separately by spill type\n",
+ "its_results_by_type = {}\n",
+ "for stype in monthly_by_type['spill_type'].unique():\n",
+ " df = monthly_by_type[monthly_by_type['spill_type'] == stype].copy()\n",
+ " model = smf.ols(\"report_delay ~ time + post_2021 + time_post\", data=df).fit()\n",
+ " df['predicted'] = model.predict(df)\n",
+ " its_results_by_type[stype] = (model, df)\n",
+ "\n",
+ "# 4. Plot side-by-side\n",
+ "fig, axs = plt.subplots(2, 1, figsize=(12, 10), sharex=True)\n",
+ "\n",
+ "for ax, (stype, (model, df)) in zip(axs, its_results_by_type.items()):\n",
+ " ax.plot(df['report_month'], df['report_delay'], marker='o', label='Observed', alpha=0.6)\n",
+ " ax.plot(df['report_month'], df['predicted'], linestyle='--', color='red', label='Predicted (ITS)')\n",
+ " ax.axvline(x=pd.Timestamp('2021-01-01'), color='black', linestyle='--', label='Policy Change (2021)')\n",
+ " ax.set_title(f\"ITS Model for {stype} Spills\")\n",
+ " ax.set_ylabel(\"Mean Report Delay (Days)\")\n",
+ " ax.legend()\n",
+ " ax.grid(True)\n",
+ "\n",
+ "axs[1].set_xlabel(\"Month\")\n",
+ "plt.tight_layout()\n",
+ "plt.show()\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "e07c8135",
+ "metadata": {},
+ "source": [
+ "### Reinterpreting the Spill-Type-Specific ITS (2021 Policy Shift)\n",
+ "\n",
+ "| Spill Type | Key Takeaways |\n",
+ "|------------|----------------|\n",
+ "| **Historical** | Delays trended downward pre-2021, but the level change after 2021 was large (–72 days) and not statistically significant due to wide variance and outliers. The slope change was slightly positive, suggesting a mild rebound or plateau after the internal mission shift. |\n",
+ "| **Recent** | Consistently lower delays throughout the time series. No discernible shift in level or trend post-2021, indicating the policy and mission change had little room to improve on already fast reporting. |\n",
+ "\n",
+ "Despite large coefficients (especially for `post_2021` in historical spills), **none of the changes are statistically significant**, likely due to high skew and extreme outliers (note the 400+ day delays).\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "efc4708c",
+ "metadata": {},
+ "source": [
+ "---"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "610abfb4",
+ "metadata": {},
+ "source": [
+ "# Conclusion: Comparative Summary\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 29,
+ "id": "546334b9",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "def extract_regression_results(model, model_name):\n",
+ " coefs = model.params\n",
+ " ses = model.bse\n",
+ " pvals = model.pvalues\n",
+ "\n",
+ " df = pd.DataFrame({\n",
+ " 'Coefficient': coefs,\n",
+ " 'Std. Error': ses,\n",
+ " 'P-Value': pvals\n",
+ " })\n",
+ " df['Model'] = model_name\n",
+ " return df.reset_index().rename(columns={'index': 'Term'})\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 30,
+ "id": "a09b247e",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "application/vnd.microsoft.datawrangler.viewer.v0+json": {
+ "columns": [
+ {
+ "name": "Term",
+ "rawType": "object",
+ "type": "string"
+ },
+ {
+ "name": "ITS (Monthly, All Spills)",
+ "rawType": "float64",
+ "type": "float"
+ },
+ {
+ "name": "ITS (Monthly, Historical)",
+ "rawType": "float64",
+ "type": "float"
+ },
+ {
+ "name": "ITS (Monthly, Recent)",
+ "rawType": "float64",
+ "type": "float"
+ },
+ {
+ "name": "Negative Binomial",
+ "rawType": "float64",
+ "type": "float"
+ },
+ {
+ "name": "OLS (Interaction)",
+ "rawType": "float64",
+ "type": "float"
+ }
+ ],
+ "conversionMethod": "pd.DataFrame",
+ "ref": "bc127456-656a-4993-9250-ec16f41a34a5",
+ "rows": [
+ [
+ "C(Period)[T.Before 2020]",
+ null,
+ null,
+ null,
+ "0.788",
+ null
+ ],
+ [
+ "C(spill_type)[T.Recent]",
+ null,
+ null,
+ null,
+ "-1.531",
+ null
+ ],
+ [
+ "C(spill_type)[T.Recent]:C(Period)[T.Before 2020]",
+ null,
+ null,
+ null,
+ "-0.311",
+ null
+ ],
+ [
+ "Intercept",
+ "15.596",
+ "41.418",
+ "5.104",
+ "2.605",
+ "5.104"
+ ],
+ [
+ "post_2021",
+ "-27.672",
+ "-72.435",
+ "-5.773",
+ null,
+ "-5.773"
+ ],
+ [
+ "time",
+ "-0.118",
+ "-0.344",
+ "-0.017",
+ null,
+ "-0.017"
+ ],
+ [
+ "time_post",
+ "0.302",
+ "0.768",
+ "0.057",
+ null,
+ "0.057"
+ ]
+ ],
+ "shape": {
+ "columns": 5,
+ "rows": 7
+ }
+ },
+ "text/html": [
+ "\n",
+ "\n",
+ "\n",
+ " \n",
+ " \n",
+ " Model \n",
+ " ITS (Monthly, All Spills) \n",
+ " ITS (Monthly, Historical) \n",
+ " ITS (Monthly, Recent) \n",
+ " Negative Binomial \n",
+ " OLS (Interaction) \n",
+ " \n",
+ " \n",
+ " Term \n",
+ " \n",
+ " \n",
+ " \n",
+ " \n",
+ " C(Period)[T.Before 2020] \n",
+ " NaN \n",
+ " NaN \n",
+ " NaN \n",
+ " 0.788 \n",
+ " NaN \n",
+ " \n",
+ " \n",
+ " C(spill_type)[T.Recent] \n",
+ " NaN \n",
+ " NaN \n",
+ " NaN \n",
+ " -1.531 \n",
+ " NaN \n",
+ " \n",
+ " \n",
+ " C(spill_type)[T.Recent]:C(Period)[T.Before 2020] \n",
+ " NaN \n",
+ " NaN \n",
+ " NaN \n",
+ " -0.311 \n",
+ " NaN \n",
+ " \n",
+ " \n",
+ " Intercept \n",
+ " 15.596 \n",
+ " 41.418 \n",
+ " 5.104 \n",
+ " 2.605 \n",
+ " 5.104 \n",
+ " \n",
+ " \n",
+ " post_2021 \n",
+ " -27.672 \n",
+ " -72.435 \n",
+ " -5.773 \n",
+ " NaN \n",
+ " -5.773 \n",
+ " \n",
+ " \n",
+ " time \n",
+ " -0.118 \n",
+ " -0.344 \n",
+ " -0.017 \n",
+ " NaN \n",
+ " -0.017 \n",
+ " \n",
+ " \n",
+ " time_post \n",
+ " 0.302 \n",
+ " 0.768 \n",
+ " 0.057 \n",
+ " NaN \n",
+ " 0.057 \n",
+ " \n",
+ " \n",
+ "
\n",
+ ""
+ ],
+ "text/plain": [
+ "Model ITS (Monthly, All Spills) \\\n",
+ "Term \n",
+ "C(Period)[T.Before 2020] NaN \n",
+ "C(spill_type)[T.Recent] NaN \n",
+ "C(spill_type)[T.Recent]:C(Period)[T.Before 2020] NaN \n",
+ "Intercept 15.596 \n",
+ "post_2021 -27.672 \n",
+ "time -0.118 \n",
+ "time_post 0.302 \n",
+ "\n",
+ "Model ITS (Monthly, Historical) \\\n",
+ "Term \n",
+ "C(Period)[T.Before 2020] NaN \n",
+ "C(spill_type)[T.Recent] NaN \n",
+ "C(spill_type)[T.Recent]:C(Period)[T.Before 2020] NaN \n",
+ "Intercept 41.418 \n",
+ "post_2021 -72.435 \n",
+ "time -0.344 \n",
+ "time_post 0.768 \n",
+ "\n",
+ "Model ITS (Monthly, Recent) \\\n",
+ "Term \n",
+ "C(Period)[T.Before 2020] NaN \n",
+ "C(spill_type)[T.Recent] NaN \n",
+ "C(spill_type)[T.Recent]:C(Period)[T.Before 2020] NaN \n",
+ "Intercept 5.104 \n",
+ "post_2021 -5.773 \n",
+ "time -0.017 \n",
+ "time_post 0.057 \n",
+ "\n",
+ "Model Negative Binomial \\\n",
+ "Term \n",
+ "C(Period)[T.Before 2020] 0.788 \n",
+ "C(spill_type)[T.Recent] -1.531 \n",
+ "C(spill_type)[T.Recent]:C(Period)[T.Before 2020] -0.311 \n",
+ "Intercept 2.605 \n",
+ "post_2021 NaN \n",
+ "time NaN \n",
+ "time_post NaN \n",
+ "\n",
+ "Model OLS (Interaction) \n",
+ "Term \n",
+ "C(Period)[T.Before 2020] NaN \n",
+ "C(spill_type)[T.Recent] NaN \n",
+ "C(spill_type)[T.Recent]:C(Period)[T.Before 2020] NaN \n",
+ "Intercept 5.104 \n",
+ "post_2021 -5.773 \n",
+ "time -0.017 \n",
+ "time_post 0.057 "
+ ]
+ },
+ "execution_count": 30,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "# Collect regression summaries\n",
+ "results_frames = []\n",
+ "\n",
+ "# Add your models — use your actual variable names\n",
+ "results_frames.append(extract_regression_results(model, 'OLS (Interaction)'))\n",
+ "results_frames.append(extract_regression_results(nb_model, 'Negative Binomial'))\n",
+ "results_frames.append(extract_regression_results(its_model, 'ITS (Monthly, All Spills)'))\n",
+ "\n",
+ "\n",
+ "# ITS by spill type (e.g., 'Historical', 'Recent')\n",
+ "for spill_type, (model, df) in its_results_by_type.items():\n",
+ " label = f\"ITS (Monthly, {spill_type})\"\n",
+ " results_frames.append(extract_regression_results(model, label))\n",
+ "\n",
+ "# Combine all\n",
+ "regression_summary_table = pd.concat(results_frames, ignore_index=True)\n",
+ "\n",
+ "# Pivot just the coefficients for display\n",
+ "regression_coef_table = regression_summary_table.pivot_table(\n",
+ " index='Term',\n",
+ " columns='Model',\n",
+ " values='Coefficient',\n",
+ " aggfunc='first'\n",
+ ").round(3)\n",
+ "\n",
+ "regression_coef_table\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "5d9578b4",
+ "metadata": {},
+ "source": [
+ "### Summary of Regression Models (Post-2021 Breakpoint)\n",
+ "\n",
+ "The change from a 2020 to a 2021 breakpoint in the ITS models allowed us to align more precisely with the agency’s operational adaptation to the new policy. The OLS interaction and Negative Binomial models were unaffected, as they rely on categorical indicators for period and spill type rather than temporal trends.\n",
+ "\n",
+ "The ITS models showed minor adjustments in intercepts and trend coefficients. For historical spills, the predicted level drop post-2021 was large (–72.4 days) but not statistically significant due to high variance. For recent spills, the model continued to show low and stable delays with no meaningful change after 2021. These updated results reinforce the conclusion that performance improvements were already underway and that the mission shift in 2021 may have stabilized or modestly extended those gains, particularly in the handling of historical spill discoveries.\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "3ccf0cf7",
+ "metadata": {},
+ "source": [
+ "### Regression Summary: Key Coefficients by Model (2021 ITS Breakpoint)\n",
+ "\n",
+ "| Term | OLS (Interaction) | Negative Binomial | ITS (Monthly, All Spills) | ITS (Historical) | ITS (Recent) |\n",
+ "|--------------------------------------------|--------------------|--------------------|-----------------------------|------------------|---------------|\n",
+ "| **Intercept** | 5.104 *** | 2.605 *** | 15.596 *** | 41.418 *** | 5.104 *** |\n",
+ "| `C(spill_type)[T.Recent]` | -10.606 *** | -1.531 *** | — | — | — |\n",
+ "| `C(Period)[T.Before 2020]` | 16.213 *** | 0.788 *** | — | — | — |\n",
+ "| `C(Recent):C(Before 2020)` | -14.428 *** | -0.311 *** | — | — | — |\n",
+ "| `time` | -0.017 | — | -0.118 | -0.344 | -0.017 |\n",
+ "| `post_2021` | -5.773 | — | -27.672 | -72.435 | -5.773 |\n",
+ "| `time_post` | 0.057 | — | 0.302 | 0.768 | 0.057 |\n",
+ "\n",
+ "\\* p < .05, \\*\\* p < .01, \\*\\*\\* p < .001 \n",
+ "Note: “—” indicates the variable was not included in that model. All ITS models use January 2021 as the intervention point.\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "1988fd14",
+ "metadata": {},
+ "source": [
+ "### Limitations\n",
+ "\n",
+ "Several limitations should be noted when interpreting these findings. First, the reporting delay data are heavily skewed with a long right tail, especially for historical spills. While count models and quantile regression (excluded here) partially address this, residual outliers still influence trend estimates in time-series analyses. Second, the categorization into \"recent\" vs. \"historical\" spills reflects underlying reporting practices, not random treatment assignment, making causal inference inherently limited. Third, the ITS models rely on monthly aggregation and assume linear trends before and after the policy change; nonlinear dynamics or unmeasured covariates may obscure more complex patterns. Finally, the formal rule change in 2020 and the internal mission shift in 2021 introduce ambiguity about when the policy was truly “implemented,” and thus, both breakpoints carry interpretive uncertainty.\n"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "spatial_env2",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.10.15"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}