+
+
+
+
+
+
+In [1]:
+
+
+
+
+
+## Title: Evaluating Equity and Impact
+### Author: David P. Adams
+
+
+
+
+
+
+
+In [2]:
+
+
+
+
+
+%pip install pandas numpy matplotlib seaborn
+
+
+
+
+
+
+
+
+
+
+
+Requirement already satisfied: pandas in /home/dadams/Repos/california_equity_git/.venv/lib/python3.12/site-packages (2.2.3) +Requirement already satisfied: numpy in /home/dadams/Repos/california_equity_git/.venv/lib/python3.12/site-packages (2.1.2) +Requirement already satisfied: matplotlib in /home/dadams/Repos/california_equity_git/.venv/lib/python3.12/site-packages (3.9.2) +Requirement already satisfied: seaborn in /home/dadams/Repos/california_equity_git/.venv/lib/python3.12/site-packages (0.13.2) +Requirement already satisfied: python-dateutil>=2.8.2 in /home/dadams/Repos/california_equity_git/.venv/lib/python3.12/site-packages (from pandas) (2.9.0.post0) +Requirement already satisfied: pytz>=2020.1 in /home/dadams/Repos/california_equity_git/.venv/lib/python3.12/site-packages (from pandas) (2024.2) +Requirement already satisfied: tzdata>=2022.7 in /home/dadams/Repos/california_equity_git/.venv/lib/python3.12/site-packages (from pandas) (2024.2) +Requirement already satisfied: contourpy>=1.0.1 in /home/dadams/Repos/california_equity_git/.venv/lib/python3.12/site-packages (from matplotlib) (1.3.0) +Requirement already satisfied: cycler>=0.10 in /home/dadams/Repos/california_equity_git/.venv/lib/python3.12/site-packages (from matplotlib) (0.12.1) +Requirement already satisfied: fonttools>=4.22.0 in /home/dadams/Repos/california_equity_git/.venv/lib/python3.12/site-packages (from matplotlib) (4.54.1) +Requirement already satisfied: kiwisolver>=1.3.1 in /home/dadams/Repos/california_equity_git/.venv/lib/python3.12/site-packages (from matplotlib) (1.4.7) +Requirement already satisfied: packaging>=20.0 in /home/dadams/Repos/california_equity_git/.venv/lib/python3.12/site-packages (from matplotlib) (24.1) +Requirement already satisfied: pillow>=8 in /home/dadams/Repos/california_equity_git/.venv/lib/python3.12/site-packages (from matplotlib) (11.0.0) +Requirement already satisfied: pyparsing>=2.3.1 in /home/dadams/Repos/california_equity_git/.venv/lib/python3.12/site-packages (from matplotlib) (3.2.0) +Requirement already satisfied: six>=1.5 in /home/dadams/Repos/california_equity_git/.venv/lib/python3.12/site-packages (from python-dateutil>=2.8.2->pandas) (1.16.0) +Note: you may need to restart the kernel to use updated packages. ++
+
+
+
+
+
+
+In [3]:
+
+
+
+
+
+# Importing the necessary libraries
+import pandas as pd
+import numpy as np
+import matplotlib.pyplot as plt
+import seaborn as sns
+import os
+
+
+
+
+
+
+
+In [4]:
+
+
+
+
+
+## set directory
+import os
+os.chdir('/home/dadams/Repos/california_equity_git')
+
+
+
+
+
+
+
+In [5]:
+
+
+
+
+
+# read in the data
+data = pd.read_csv('data_raw/cci_programs_data.csv', low_memory=False)
+
+
+
+
+
+
+
+In [6]:
+
+
+
+
+
+columns_data = pd.DataFrame(data.columns)
+
+# export the 'DataFrame' to a csv file
+columns_data.to_csv('data_raw/columns.csv', index=False)
+
+
+
+
+
+
+
+In [7]:
+
+
+
+
+
+# check the data types
+data.dtypes
+
+# save the data types to a csv file
+data.dtypes.to_csv('data_raw/data_types.csv', header = False)
+
+
+
+
+
+
+
+In [8]:
+
+
+
+
+
+# describe the data
+data.describe()
+
+
+
+
+
+
+
+
+Out[8]:
+
+
+
+
+
+
+
+
+| + | Census Tract | +Total Project Cost | +Total Program GGRFFunding | +Total Project GHGReductions | +Annual Project GHGReductions | +Project Count | +Total GGRFDisadvantaged Community Funding | +Funding Benefiting Disadvantaged Communities | +Estimated Num Vehicles In Service | +Funding Within Disadvantage Communities | +... | +Indirect Jobs Fte | +Induced Jobs Fte | +Compost Produced Tons | +Compost Produced Tons Yr | +Net Density DUA | +Applicants Assisted | +Invasive Cover 12 Months | +Invasive Cover 36 Months | +Project Acreage | +Intermediary Admin Expenses Calc | +
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| count | +1.193700e+05 | +1.414290e+05 | +1.414290e+05 | +1.414290e+05 | +141429.000000 | +141429.000000 | +5.518700e+04 | +55187.000000 | +141429.000000 | +5.518700e+04 | +... | +141429.000000 | +141429.000000 | +141429.000000 | +141429.0 | +141429.000000 | +141429.000000 | +141429.000000 | +141429.000000 | +1.414290e+05 | +1.414290e+05 | +
| mean | +6.053889e+09 | +9.206412e+05 | +7.791664e+04 | +7.717972e+02 | +0.205389 | +4.090872 | +2.736820e+04 | +110.217551 | +0.047331 | +2.030028e+04 | +... | +0.064567 | +0.117056 | +0.440977 | +0.0 | +0.082260 | +0.019642 | +0.010656 | +0.010578 | +9.771087e+00 | +1.911114e+03 | +
| std | +2.641870e+07 | +3.736191e+07 | +1.011645e+06 | +2.371604e+04 | +3.361723 | +18.381861 | +6.327936e+05 | +1738.772195 | +1.972262 | +5.590536e+05 | +... | +1.444316 | +2.261146 | +45.712955 | +0.0 | +4.382593 | +1.406914 | +0.758250 | +0.757677 | +3.669526e+03 | +1.196317e+05 | +
| min | +6.001400e+09 | +0.000000e+00 | +0.000000e+00 | +-2.785930e+05 | +0.000000 | +0.000000 | +0.000000e+00 | +0.000000 | +0.000000 | +0.000000e+00 | +... | +0.000000 | +0.000000 | +0.000000 | +0.0 | +0.000000 | +0.000000 | +0.000000 | +0.000000 | +0.000000e+00 | +0.000000e+00 | +
| 25% | +6.037271e+09 | +4.000000e+03 | +3.500000e+03 | +8.000000e+00 | +0.000000 | +1.000000 | +0.000000e+00 | +0.000000 | +0.000000 | +0.000000e+00 | +... | +0.000000 | +0.000000 | +0.000000 | +0.0 | +0.000000 | +0.000000 | +0.000000 | +0.000000 | +0.000000e+00 | +0.000000e+00 | +
| 50% | +6.059022e+09 | +8.000000e+03 | +7.500000e+03 | +1.500000e+01 | +0.000000 | +1.000000 | +0.000000e+00 | +0.000000 | +0.000000 | +0.000000e+00 | +... | +0.000000 | +0.000000 | +0.000000 | +0.0 | +0.000000 | +0.000000 | +0.000000 | +0.000000 | +0.000000e+00 | +0.000000e+00 | +
| 75% | +6.073016e+09 | +2.150000e+04 | +1.900000e+04 | +4.000000e+01 | +0.000000 | +3.000000 | +5.500000e+03 | +0.000000 | +0.000000 | +1.500000e+03 | +... | +0.000000 | +0.000000 | +0.000000 | +0.0 | +0.000000 | +0.000000 | +0.000000 | +0.000000 | +0.000000e+00 | +0.000000e+00 | +
| max | +6.115041e+09 | +5.767700e+09 | +1.412670e+08 | +4.748924e+06 | +336.000000 | +2072.000000 | +6.443700e+07 | +102348.000000 | +503.000000 | +6.443700e+07 | +... | +110.170000 | +151.000000 | +10365.000000 | +0.0 | +706.000000 | +320.000000 | +85.000000 | +85.000000 | +1.380000e+06 | +2.000000e+07 | +
8 rows × 82 columns
+
+
+
+
+
+
+
+In [9]:
+
+
+
+
+
+import geopandas as gpd
+
+# Load the shapefile
+shapefile_path = '/home/dadams/Repos/california_equity_git/california_enviroscreen/calif_enviroscreen_shape/CES4 Final Shapefile.shp'
+gdf = gpd.read_file(shapefile_path)
+
+# Print the head of the GeoDataFrame
+print(gdf.head())
+
+
+
+
+
+
+
+
+
+
+
+Tract ZIP County ApproxLoc TotPop19 CIscore \ +0 6.083002e+09 93454 Santa Barbara Santa Maria 4495 36.019653 +1 6.083002e+09 93455 Santa Barbara Santa Maria 13173 37.030667 +2 6.083002e+09 93454 Santa Barbara Santa Maria 2398 31.213140 +3 6.083002e+09 93455 Santa Barbara Orcutt 4496 6.639331 +4 6.083002e+09 93455 Santa Barbara Orcutt 4008 14.022852 + + CIscoreP Ozone OzoneP PM2_5 ... Elderly65 Hispanic \ +0 69.162885 0.034190 10.566273 7.567724 ... 12.5028 68.9210 +1 70.637922 0.035217 11.561917 7.624775 ... 5.3519 78.6229 +2 61.069087 0.034190 10.566273 7.548835 ... 12.8857 65.7214 +3 5.988401 0.036244 13.615432 7.660570 ... 14.4128 22.9537 +4 23.121533 0.036244 13.615432 7.663210 ... 18.8872 33.4082 + + White AfricanAm NativeAm OtherMult Shape_Leng Shape_Area \ +0 20.8899 0.4004 0.2670 1.3126 6999.357689 2.847611e+06 +1 13.2240 2.5051 0.0000 0.9489 19100.578232 1.635292e+07 +2 30.6088 0.9591 0.0000 2.1685 4970.985897 1.352329e+06 +3 69.1948 0.9342 0.7117 2.5356 6558.956012 2.417717e+06 +4 59.7804 0.6986 1.4721 1.3723 6570.368730 2.608422e+06 + + AAPI geometry +0 8.2091 POLYGON ((-39795.07 -341919.191, -38126.384 -3... +1 4.6990 POLYGON ((-39795.07 -341919.191, -39803.632 -3... +2 0.5421 POLYGON ((-38115.747 -341130.248, -38126.384 -... +3 3.6699 POLYGON ((-37341.662 -348530.437, -37252.307 -... +4 3.2685 POLYGON ((-39465.107 -348499.262, -38244.305 -... + +[5 rows x 67 columns] ++
+
+
+
+
+
+
+In [11]:
+
+
+
+
+
+# Let's look at the actual column names in the merged dataset
+print("Columns containing 'County':")
+print([col for col in data.columns if 'County' in col])
+
+# Then modify our analysis using the correct county column name:
+geographic_analysis = data.groupby(['County', 'project_category']).agg({
+ 'Total Project Cost': 'sum',
+ 'Total Project GHGReductions': 'sum',
+ 'Total GGRFDisadvantaged Community Funding': 'sum'
+}).reset_index()
+
+# Rest of the analysis:
+# 1. First look at program distributions
+print("\nProgram Types:")
+print(data['Program Name'].value_counts().head(10))
+print("\nProject Types:")
+print(data['Project Type'].value_counts())
+print("\nFunding Recipient Types:")
+print(data['PRIMARY_FUNDING_RECIPIENT_TYPE'].value_counts())
+
+
+
+
+
+
+
+
+
+
+
+Columns containing 'County': +['County'] + +Program Types: +Program Name +Low Carbon Transportation 105965 +Funding Agricultural Replacement Measures for Emission Reductions Program 8554 +Low-Income Weatherization Program 6468 +Community Air Protection 5514 +Water-Energy Efficiency 5111 +Climate Smart Agriculture 1688 +Low Carbon Transit Operations Program 1003 +Transformative Climate Communities 973 +Urban and Community Forestry Program 967 +Woodsmoke Reduction Program 959 +Name: count, dtype: int64 + +Project Types: +Project Type +Voucher 2808 +Voucher Incentive 1730 +Air Filtration (Air Filtration) 1570 +Water Use and Energy Efficiency 1217 +Off-Road Equipment - Mobile Agricultural (Tractors) 891 + ... +Procurement Requirement 1 +Education and outreach and enforcement/inspections 1 +Personnel, education and outreach, and procurement 1 +Education & Outreach and Planning 1 +Equipment, Education and Outreach 1 +Name: count, Length: 1007, dtype: int64 + +Funding Recipient Types: +PRIMARY_FUNDING_RECIPIENT_TYPE +Business 423 +Government Agency 285 +Nonprofit 113 +Educational institution 21 +California Native American Tribe 4 +Government agency 2 +Name: count, dtype: int64 ++
+
+
+
+
+
+
+
+In [13]:
+
+
+
+
+
+# 1. Fix the classification function
+def classify_project_type_v2(row):
+ program = str(row['Program Name']).lower()
+ project_type = str(row['Project Type']).lower()
+
+ # Individual incentive programs
+ if program == 'low carbon transportation':
+ if 'voucher' in project_type.lower():
+ return 'Individual - Vehicle'
+ return 'Transportation - Other'
+ elif program in ['low-income weatherization program', 'woodsmoke reduction program']:
+ return 'Individual - Home Energy'
+ elif program == 'water-energy efficiency':
+ return 'Water-Energy Projects'
+ elif program == 'community air protection':
+ return 'Air Quality Projects'
+ elif program == 'climate smart agriculture':
+ return 'Agricultural Projects'
+ elif 'voucher' in project_type.lower():
+ return 'Individual - Other'
+ elif program == 'transformative climate communities':
+ return 'Community Projects'
+
+ # Institutional by recipient type
+ elif pd.notna(row['PRIMARY_FUNDING_RECIPIENT_TYPE']):
+ recipient = str(row['PRIMARY_FUNDING_RECIPIENT_TYPE'])
+ if recipient in ['Government Agency', 'Government agency', 'California Native American Tribe']:
+ return 'Institutional - Government'
+ elif recipient == 'Educational institution':
+ return 'Institutional - Education'
+ elif recipient == 'Nonprofit':
+ return 'Institutional - Nonprofit'
+ elif recipient == 'Business':
+ if row['Total Project Cost'] > 100000:
+ return 'Institutional - Large Business'
+ return 'Business - Small'
+
+ return 'Other'
+
+# 2. Add safe division function for metrics
+def safe_divide(a, b):
+ if b == 0:
+ return 0
+ return a / b
+
+# 3. Recalculate metrics
+data['project_category'] = data.apply(classify_project_type_v2, axis=1)
+data['dac_funding_ratio'] = data.apply(lambda x: safe_divide(x['Total GGRFDisadvantaged Community Funding'], x['Total Project Cost']), axis=1)
+data['ghg_per_dollar'] = data.apply(lambda x: safe_divide(x['Total Project GHGReductions'], x['Total Project Cost']), axis=1)
+
+# 4. Rerun analysis
+updated_analysis = data.groupby('project_category').agg({
+ 'Total Project Cost': ['count', 'sum', 'mean'],
+ 'Total Project GHGReductions': ['sum', 'mean'],
+ 'dac_funding_ratio': ['mean', 'median'],
+ 'ghg_per_dollar': ['mean', 'median']
+}).round(4)
+
+print("\nUpdated Project Category Distribution:")
+print(data['project_category'].value_counts())
+print("\nUpdated Analysis:")
+print(updated_analysis)
+
+
+
+
+
+
+
+
+
+
+
++Updated Project Category Distribution: +project_category +Transportation - Other 101427 +Other 14340 +Individual - Home Energy 7427 +Air Quality Projects 5514 +Water-Energy Projects 5111 +Individual - Vehicle 4538 +Agricultural Projects 1688 +Community Projects 973 +Institutional - Government 269 +Institutional - Nonprofit 100 +Institutional - Large Business 23 +Institutional - Education 17 +Business - Small 2 +Name: count, dtype: int64 + +Updated Analysis: + Total Project Cost \ + count sum mean +project_category +Agricultural Projects 1688 886514974 5.251866e+05 +Air Quality Projects 5514 1211608583 2.197331e+05 +Business - Small 2 65376 3.268800e+04 +Community Projects 973 508817727 5.229370e+05 +Individual - Home Energy 7427 320008647 4.308720e+04 +Individual - Vehicle 4538 1286418141 2.834769e+05 +Institutional - Education 17 5561672 3.271572e+05 +Institutional - Government 269 16939974420 6.297388e+07 +Institutional - Large Business 23 513921827 2.234443e+07 +Institutional - Nonprofit 100 1402247011 1.402247e+07 +Other 14340 104572013338 7.292330e+06 +Transportation - Other 101427 2473090031 2.438296e+04 +Water-Energy Projects 5111 85125916 1.665543e+04 + + Total Project GHGReductions \ + sum mean +project_category +Agricultural Projects 24249378 14365.7453 +Air Quality Projects 250801 45.4844 +Business - Small 0 0.0000 +Community Projects 154010 158.2837 +Individual - Home Energy 693914 93.4313 +Individual - Vehicle 993992 219.0375 +Institutional - Education 457 26.8824 +Institutional - Government 6580289 24462.0409 +Institutional - Large Business 388663 16898.3913 +Institutional - Nonprofit 4791834 47918.3400 +Other 67313019 4694.0738 +Transportation - Other 3312930 32.6632 +Water-Energy Projects 425225 83.1980 + + dac_funding_ratio ghg_per_dollar \ + mean median mean +project_category +Agricultural Projects 0.2388 0.0000 0.0106 +Air Quality Projects NaN NaN 0.0002 +Business - Small NaN NaN 0.0000 +Community Projects NaN NaN 0.0004 +Individual - Home Energy 0.6504 0.7405 0.0038 +Individual - Vehicle 0.1660 0.0245 0.0017 +Institutional - Education NaN NaN 0.0008 +Institutional - Government 0.0000 0.0000 0.0022 +Institutional - Large Business NaN NaN 0.0038 +Institutional - Nonprofit 0.0000 0.0000 0.0182 +Other 0.1483 0.0000 0.0023 +Transportation - Other 0.4190 0.0000 0.0027 +Water-Energy Projects 0.4359 0.0000 0.1027 + + + median +project_category +Agricultural Projects 0.0051 +Air Quality Projects 0.0000 +Business - Small 0.0000 +Community Projects 0.0000 +Individual - Home Energy 0.0026 +Individual - Vehicle 0.0010 +Institutional - Education 0.0000 +Institutional - Government 0.0001 +Institutional - Large Business 0.0005 +Institutional - Nonprofit 0.0004 +Other 0.0001 +Transportation - Other 0.0032 +Water-Energy Projects 0.0096 ++
+
+
+
+
+
+
+
+
+This refined analysis reveals some fascinating patterns about California's climate investments and their environmental justice implications. Let's break down the key findings:
+-
+
- Program Distribution and Scale: +
-
+
- Transportation dominates numerically:
-
+
- Transportation-Other: 101,427 projects (71.6% of all projects) +
- Individual Vehicle incentives: 4,538 projects +
+ - Individual programs have high reach but moderate costs:
-
+
- Home Energy: 7,427 projects (~$43K average) +
- Vehicle incentives: 4,538 projects (~$283K average) +
+ - Institutional projects are fewer but larger:
-
+
- Government: 269 projects (~$63M average) +
- Nonprofits: 100 projects (~$14M average) +
+
-
+
- Environmental Justice (DAC Funding Ratio): +
-
+
- Home Energy programs show strongest DAC focus (65% DAC funding ratio) +
- Water-Energy (44%) and Transportation-Other (42%) also show good DAC targeting +
- Individual Vehicle programs have lower DAC funding (17%) +
- Institutional programs show minimal DAC funding ratios +
-
+
- GHG Reduction Efficiency (GHG per dollar): +
-
+
- Water-Energy Projects are most efficient (0.1027 tons/$) +
- Nonprofit projects show high efficiency (0.0182 tons/$) +
- Agricultural Projects also perform well (0.0106 tons/$) +
- Individual incentives show lower efficiency:
-
+
- Home Energy: 0.0038 tons/$ +
- Vehicle: 0.0017 tons/$ +
+
Key Environmental Justice Implications:
+-
+
- Individual vs Institutional Trade-offs: +
-
+
- Individual programs (especially home energy) are better at reaching disadvantaged communities +
- But institutional programs tend to be more cost-effective for GHG reductions +
- This suggests a potential tension between equity and efficiency goals +
-
+
- Program-Specific Patterns: +
-
+
- Home Energy programs show the best balance of DAC reach and moderate GHG efficiency +
- Vehicle incentives have lower DAC reach and efficiency, supporting your hypothesis about wealth bias +
- Water-Energy projects show promising combination of good DAC reach and high efficiency +
-
+
- Policy Implications: +
-
+
- Consider expanding successful programs like Water-Energy that combine good DAC reach with high efficiency +
- Look for ways to improve DAC access to vehicle incentives +
- Consider bundling institutional and individual programs to balance efficiency and equity goals +
Would you like to explore any of these aspects in more detail? We could:
+-
+
- Drill deeper into geographic distribution +
- Analyze temporal trends in these patterns +
- Look at specific program characteristics that drive DAC success +
+
+
+
+
+
+
+In [16]:
+
+
+
+
+
+import matplotlib.pyplot as plt
+import seaborn as sns
+
+# Prepare data for plotting
+plot_data = data.groupby('project_category').agg({
+ 'Total Project Cost': ['count', 'mean'],
+ 'dac_funding_ratio': 'mean',
+ 'ghg_per_dollar': 'mean'
+}).reset_index()
+
+# Flatten column names
+plot_data.columns = ['project_category', 'count', 'avg_cost', 'dac_ratio', 'ghg_efficiency']
+
+# Create figure with multiple subplots
+fig, (ax1, ax2, ax3) = plt.subplots(3, 1, figsize=(15, 20))
+
+# 1. Project Counts
+sns.barplot(data=plot_data,
+ x='count',
+ y='project_category',
+ ax=ax1,
+ palette='viridis')
+ax1.set_title('Number of Projects by Category')
+ax1.set_xlabel('Number of Projects')
+
+# 2. DAC Funding Ratio
+mask = plot_data['dac_ratio'].notna() # Filter out NaN values
+sns.barplot(data=plot_data[mask],
+ x='dac_ratio',
+ y='project_category',
+ ax=ax2,
+ palette='viridis')
+ax2.set_title('Disadvantaged Community Funding Ratio by Category')
+ax2.set_xlabel('DAC Funding Ratio')
+
+# 3. GHG Efficiency
+mask = plot_data['ghg_efficiency'].notna() # Filter out NaN values
+sns.barplot(data=plot_data[mask],
+ x='ghg_efficiency',
+ y='project_category',
+ ax=ax3,
+ palette='viridis')
+ax3.set_title('GHG Reduction Efficiency by Category (tons CO2e per dollar)')
+ax3.set_xlabel('GHG Reduction per Dollar')
+
+plt.tight_layout()
+plt.show()
+
+# Create a scatter plot to show relationship between project size and efficiency
+plt.figure(figsize=(12, 8))
+plt.scatter(plot_data['avg_cost'],
+ plot_data['ghg_efficiency'],
+ alpha=0.6)
+
+# Add labels for each point
+for i, row in plot_data.iterrows():
+ plt.annotate(row['project_category'],
+ (row['avg_cost'], row['ghg_efficiency']),
+ xytext=(5, 5),
+ textcoords='offset points')
+
+plt.xscale('log') # Use log scale for cost due to wide range
+plt.xlabel('Average Project Cost (log scale)')
+plt.ylabel('GHG Reduction Efficiency (tons CO2e per dollar)')
+plt.title('Project Cost vs GHG Reduction Efficiency')
+plt.grid(True)
+plt.show()
+
+
+
+
+
+
+
+
+
+
+
+/tmp/ipykernel_738944/3534785368.py:18: FutureWarning: + +Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `y` variable to `hue` and set `legend=False` for the same effect. + + sns.barplot(data=plot_data, +/tmp/ipykernel_738944/3534785368.py:28: FutureWarning: + +Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `y` variable to `hue` and set `legend=False` for the same effect. + + sns.barplot(data=plot_data[mask], +/tmp/ipykernel_738944/3534785368.py:38: FutureWarning: + +Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `y` variable to `hue` and set `legend=False` for the same effect. + + sns.barplot(data=plot_data[mask], ++
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+In [18]:
+
+
+
+
+
+# First let's see how many counties we have data for and their distribution
+county_analysis = data.groupby(['County', 'project_category']).agg({
+ 'Total Project Cost': ['count', 'sum'],
+ 'Total Project GHGReductions': 'sum',
+ 'dac_funding_ratio': 'mean'
+}).round(2)
+
+# Reset index for easier manipulation
+county_analysis = county_analysis.reset_index()
+
+# Flatten column names
+county_analysis.columns = ['County', 'project_category', 'project_count', 'total_cost', 'total_ghg', 'dac_ratio']
+
+print("\nTop 10 Counties by Project Count:")
+print(data['County'].value_counts().head(10))
+
+print("\nProject Categories by County (top 5 counties):")
+top_counties = data['County'].value_counts().head(5).index
+for county in top_counties:
+ print(f"\n{county}:")
+ county_cats = data[data['County'] == county]['project_category'].value_counts()
+ print(county_cats)
+
+# Calculate percentage of projects in each county by category
+county_category_pcts = pd.crosstab(data['County'],
+ data['project_category'],
+ values=data['Total Project Cost'],
+ aggfunc='sum',
+ normalize='index') * 100
+
+print("\nPercentage of Funding by Category in Top 5 Counties:")
+print(county_category_pcts.loc[top_counties].round(2))
+
+
+
+
+
+
+
+
+
+
+
++Top 10 Counties by Project Count: +County +Los Angeles 38032 +Orange 10694 +San Diego 8468 +Fresno 7854 +Santa Clara 7154 +Alameda 6381 +Riverside 6190 +San Bernardino 5159 +Sacramento 4781 +San Joaquin 4202 +Name: count, dtype: int64 + +Project Categories by County (top 5 counties): + +Los Angeles: +project_category +Transportation - Other 30411 +Individual - Home Energy 2233 +Individual - Vehicle 2187 +Water-Energy Projects 1945 +Other 794 +Air Quality Projects 352 +Community Projects 79 +Institutional - Government 19 +Institutional - Nonprofit 6 +Agricultural Projects 3 +Institutional - Large Business 3 +Name: count, dtype: int64 + +Orange: +project_category +Transportation - Other 9866 +Individual - Vehicle 290 +Individual - Home Energy 184 +Water-Energy Projects 169 +Other 129 +Air Quality Projects 51 +Institutional - Government 4 +Community Projects 1 +Name: count, dtype: int64 + +San Diego: +project_category +Transportation - Other 7571 +Other 375 +Individual - Vehicle 294 +Air Quality Projects 105 +Individual - Home Energy 83 +Agricultural Projects 30 +Institutional - Government 6 +Institutional - Nonprofit 3 +Community Projects 1 +Name: count, dtype: int64 + +Fresno: +project_category +Transportation - Other 2541 +Other 2144 +Air Quality Projects 1747 +Individual - Home Energy 769 +Water-Energy Projects 338 +Agricultural Projects 142 +Individual - Vehicle 96 +Community Projects 64 +Institutional - Government 7 +Institutional - Education 3 +Institutional - Large Business 2 +Institutional - Nonprofit 1 +Name: count, dtype: int64 + +Santa Clara: +project_category +Transportation - Other 6214 +Water-Energy Projects 530 +Individual - Vehicle 177 +Other 89 +Individual - Home Energy 74 +Air Quality Projects 50 +Agricultural Projects 14 +Institutional - Government 3 +Institutional - Nonprofit 2 +Institutional - Large Business 1 +Name: count, dtype: int64 + +Percentage of Funding by Category in Top 5 Counties: +project_category Agricultural Projects Air Quality Projects \ +County +Los Angeles 0.00 0.36 +Orange 0.00 1.44 +San Diego 0.05 1.61 +Fresno 6.12 10.44 +Santa Clara 0.03 0.61 + +project_category Business - Small Community Projects \ +County +Los Angeles 0.0 0.14 +Orange 0.0 0.15 +San Diego 0.0 0.00 +Fresno 0.0 17.55 +Santa Clara 0.0 0.00 + +project_category Individual - Home Energy Individual - Vehicle \ +County +Los Angeles 0.06 0.91 +Orange 0.09 2.90 +San Diego 0.10 1.45 +Fresno 4.74 2.50 +Santa Clara 0.12 2.07 + +project_category Institutional - Education Institutional - Government \ +County +Los Angeles 0.00 5.35 +Orange 0.00 37.65 +San Diego 0.00 2.90 +Fresno 0.02 0.73 +Santa Clara 0.00 15.35 + +project_category Institutional - Large Business Institutional - Nonprofit \ +County +Los Angeles 0.26 0.71 +Orange 0.00 0.00 +San Diego 0.00 2.54 +Fresno 4.07 0.01 +Santa Clara 3.88 0.04 + +project_category Other Transportation - Other Water-Energy Projects +County +Los Angeles 91.48 0.73 0.00 +Orange 51.20 6.46 0.12 +San Diego 88.01 3.33 0.00 +Fresno 47.73 5.73 0.36 +Santa Clara 73.37 4.43 0.09 ++
+
+
+
+
+
+
+In [19]:
+
+
+
+
+
+# Create a stacked bar chart for project distribution across top counties
+plt.figure(figsize=(15, 8))
+county_category_pcts.loc[top_counties].plot(kind='bar', stacked=True)
+plt.title('Distribution of Project Funding by Category Across Top Counties')
+plt.xlabel('County')
+plt.ylabel('Percentage of Total Funding')
+plt.legend(bbox_to_anchor=(1.05, 1), loc='upper left')
+plt.tight_layout()
+plt.show()
+
+# Create a visualization for DAC funding ratio by county
+plt.figure(figsize=(15, 8))
+sns.boxplot(data=data[data['dac_funding_ratio'].notna()],
+ x='County',
+ y='dac_funding_ratio',
+ order=top_counties)
+plt.title('DAC Funding Ratio Distribution by County')
+plt.xticks(rotation=45)
+plt.tight_layout()
+plt.show()
+
+
+
+
+
+
+
+
+
+
+
+<Figure size 1500x800 with 0 Axes>+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+In [20]:
+
+
+
+
+
+# First, let's create a clearer categorization focused on our question
+def classify_for_ej_analysis(row):
+ program = str(row['Program Name']).lower()
+ project_type = str(row['Project Type']).lower()
+
+ # Individual incentives
+ if program == 'low carbon transportation' and 'voucher' in project_type.lower():
+ return 'Individual - Vehicle'
+ elif program == 'low-income weatherization program':
+ return 'Individual - Solar/Energy'
+ elif program == 'woodsmoke reduction program':
+ return 'Individual - Solar/Energy'
+ # Large projects
+ elif row['Total Project Cost'] > 100000: # threshold for "large" projects
+ return 'Large Projects'
+ else:
+ return 'Other'
+
+data['ej_category'] = data.apply(classify_for_ej_analysis, axis=1)
+
+# Analyze environmental justice metrics by category
+ej_analysis = data.groupby('ej_category').agg({
+ 'Total Project Cost': ['count', 'sum', 'mean'],
+ 'dac_funding_ratio': ['mean', 'median'],
+ 'Is Benefit Disadvantaged Communities': 'mean',
+ 'Is Low Income Communities': 'mean',
+ 'Total Project GHGReductions': 'sum'
+}).round(4)
+
+print("Environmental Justice Analysis by Category:")
+print(ej_analysis)
+
+# Calculate percentage of funds going to different income communities
+ej_distribution = data.groupby('ej_category').agg({
+ 'Total Project Cost': 'sum',
+ 'Total GGRFDisadvantaged Community Funding': 'sum',
+ 'Low Income Amount': 'sum'
+})
+
+ej_distribution['DAC_percentage'] = (ej_distribution['Total GGRFDisadvantaged Community Funding'] /
+ ej_distribution['Total Project Cost'] * 100)
+ej_distribution['LowIncome_percentage'] = (ej_distribution['Low Income Amount'] /
+ ej_distribution['Total Project Cost'] * 100)
+
+print("\nPercentage of Funding to Disadvantaged and Low-Income Communities:")
+print(ej_distribution[['DAC_percentage', 'LowIncome_percentage']].round(2))
+
+# Visualize the distribution
+plt.figure(figsize=(12, 6))
+bar_width = 0.35
+categories = ej_distribution.index
+
+x = np.arange(len(categories))
+plt.bar(x - bar_width/2, ej_distribution['DAC_percentage'],
+ bar_width, label='DAC Funding %', alpha=0.8)
+plt.bar(x + bar_width/2, ej_distribution['LowIncome_percentage'],
+ bar_width, label='Low Income Funding %', alpha=0.8)
+
+plt.xlabel('Project Category')
+plt.ylabel('Percentage of Total Funding')
+plt.title('Distribution of Funding to Disadvantaged and Low-Income Communities')
+plt.xticks(x, categories, rotation=45)
+plt.legend()
+plt.tight_layout()
+plt.show()
+
+# Let's also look at the per-project impact
+impact_analysis = data.groupby('ej_category').agg({
+ 'Total Project GHGReductions': 'mean',
+ 'Total Project Cost': 'mean'
+}).round(2)
+
+impact_analysis['GHG_reduction_per_dollar'] = (impact_analysis['Total Project GHGReductions'] /
+ impact_analysis['Total Project Cost']).round(4)
+
+print("\nImpact Analysis:")
+print(impact_analysis)
+
+
+
+
+
+
+
+
+
+
+
+Environmental Justice Analysis by Category: + Total Project Cost \ + count sum mean +ej_category +Individual - Solar/Energy 7427 320008647 4.308720e+04 +Individual - Vehicle 4538 1286418141 2.834769e+05 +Large Projects 10009 127060954384 1.269467e+07 +Other 119455 1537986491 1.287503e+04 + + dac_funding_ratio \ + mean median +ej_category +Individual - Solar/Energy 0.6504 0.7405 +Individual - Vehicle 0.1660 0.0245 +Large Projects 0.2761 0.0000 +Other 0.4129 0.0000 + + Is Benefit Disadvantaged Communities \ + mean +ej_category +Individual - Solar/Energy 0.7824 +Individual - Vehicle 0.3455 +Large Projects 0.0664 +Other 0.3275 + + Is Low Income Communities \ + mean +ej_category +Individual - Solar/Energy 0.9338 +Individual - Vehicle 0.4125 +Large Projects 0.5602 +Other 0.5708 + + Total Project GHGReductions + sum +ej_category +Individual - Solar/Energy 693914 +Individual - Vehicle 993992 +Large Projects 103527784 +Other 3938822 + +Percentage of Funding to Disadvantaged and Low-Income Communities: + DAC_percentage LowIncome_percentage +ej_category +Individual - Solar/Energy 41.46 4.97 +Individual - Vehicle 5.13 2.01 +Large Projects 0.91 1.47 +Other 10.11 16.38 ++
+
+
+
+
+
+
+
+
+
+
+
++Impact Analysis: + Total Project GHGReductions Total Project Cost \ +ej_category +Individual - Solar/Energy 93.43 43087.20 +Individual - Vehicle 219.04 283476.89 +Large Projects 10343.47 12694670.24 +Other 32.97 12875.03 + + GHG_reduction_per_dollar +ej_category +Individual - Solar/Energy 0.0022 +Individual - Vehicle 0.0008 +Large Projects 0.0008 +Other 0.0026 ++
+
+
+
+
+
+
+
+
+Patterns of Equity in California's Climate Investments¶
-
+
- Individual Solar/Energy Programs Show Strong Equity Focus: +
-
+
- Highest DAC funding ratio (0.65 mean, 0.74 median) +
- 78% benefit disadvantaged communities +
- 93% benefit low-income communities +
- 41.46% of funds go to DACs, highest among all categories +
- However, relatively small total investment ($320M) +
-
+
- Individual Vehicle Programs Show Wealth Disparity: +
-
+
- Much lower DAC funding ratio (0.17 mean, 0.02 median) +
- Only 34.5% benefit disadvantaged communities +
- Only 5.13% of funds go to DACs +
- Larger individual project costs (~$283K vs $43K for solar) +
- This supports your hypothesis about vehicle incentives benefiting wealthier communities +
-
+
- Large Projects Show Mixed Results: +
-
+
- Low DAC funding ratio (0.28 mean, 0.00 median) +
- Only 6.6% benefit disadvantaged communities +
- But 56% benefit low-income communities +
- Largest total investment ($127B) +
- Highest total GHG reductions (103.5M tons) +
Key Findings:
+-
+
- Your hypothesis about vehicle incentives primarily benefiting wealthier communities is strongly supported by the data +
- However, individual solar/energy programs are actually very successful at reaching disadvantaged communities +
- Large projects, while achieving the most total GHG reductions, have mixed equity outcomes - good at reaching low-income communities but not DACs +
Policy Implications:
+-
+
- The solar/energy program model could potentially be adapted for vehicle incentives to improve equity +
- There may be structural barriers preventing DACs from accessing vehicle incentives that need to be addressed +
- Large projects might benefit from stronger DAC targeting requirements, given their current low DAC participation despite good low-income community reach +
Would you like to explore any of these aspects in more detail? We could:
+-
+
- Analyze what makes the solar/energy programs more successful at reaching DACs +
- Look at specific barriers in the vehicle incentive programs +
- Examine whether certain types of large projects are better at reaching disadvantaged communities +
+
+
+
+
+
+
+In [22]:
+
+
+
+
+
+# 1. First, let's analyze the overall spatial distribution of investments
+county_analysis = data.groupby('County').agg({
+ 'Total Project Cost': ['count', 'sum'],
+ 'Total Project GHGReductions': 'sum',
+ 'dac_funding_ratio': 'mean',
+ 'Is Benefit Disadvantaged Communities': 'mean',
+ 'Is Low Income Communities': 'mean'
+}).round(4)
+
+# Add per capita metrics (we would need to merge with county population data)
+# Add GHG reduction per dollar by county
+
+# 2. Let's analyze vulnerable communities specifically
+vulnerability_analysis = pd.DataFrame({
+ 'Total Projects': [
+ len(data),
+ len(data[data['Is Benefit Disadvantaged Communities'] == 1]),
+ len(data[data['Is Low Income Communities'] == 1])
+ ],
+ 'Total Investment': [
+ data['Total Project Cost'].sum(),
+ data[data['Is Benefit Disadvantaged Communities'] == 1]['Total Project Cost'].sum(),
+ data[data['Is Low Income Communities'] == 1]['Total Project Cost'].sum()
+ ],
+ 'GHG Reduction': [
+ data['Total Project GHGReductions'].sum(),
+ data[data['Is Benefit Disadvantaged Communities'] == 1]['Total Project GHGReductions'].sum(),
+ data[data['Is Low Income Communities'] == 1]['Total Project GHGReductions'].sum()
+ ]
+}, index=['All Communities', 'Disadvantaged Communities', 'Low Income Communities'])
+
+# 3. Project type distribution in vulnerable communities
+project_vulnerability = data.groupby(['project_category', 'Is Benefit Disadvantaged Communities']).agg({
+ 'Total Project Cost': ['count', 'sum'],
+ 'Total Project GHGReductions': 'sum'
+}).round(4)
+
+print("Overall Project Distribution by County:")
+print(county_analysis.sort_values(('Total Project Cost', 'sum'), ascending=False).head(10))
+
+print("\nVulnerable Communities Analysis:")
+print(vulnerability_analysis)
+
+print("\nProject Types in Vulnerable Communities:")
+print(project_vulnerability)
+
+# Visualizations
+plt.figure(figsize=(15, 8))
+# Plot investment distribution...
+plt.subplot(1, 2, 1)
+county_analysis[('Total Project Cost', 'sum')].sort_values(ascending=False).head(10).plot(kind='bar')
+plt.title('Top 10 Counties by Total Investment')
+plt.ylabel('Total Investment')
+plt.xlabel('County')
+
+# ...and GHG reductions
+plt.subplot(1, 2, 2)
+county_analysis[('Total Project GHGReductions', 'sum')].sort_values(ascending=False).head(10).plot(kind='bar')
+plt.title('Top 10 Counties by GHG Reductions')
+plt.ylabel('Total GHG Reductions')
+plt.xlabel('County')
+
+plt.tight_layout()
+plt.show()
+
+
+
+
+
+
+
+
+
+
+
+Overall Project Distribution by County: + Total Project Cost \ + count +County +Los Angeles 38032 +Alameda 6381 +Alameda, Contra Costa, Fresno, Kern, Kings, Mad... 9 +Sacramento 4781 +San Diego 8468 +San Francisco 2748 +Santa Clara 7154 +San Francisco, Santa Clara 6 +Orange 10694 +San Diego, Santa Barbara, Ventura 23 + + \ + sum +County +Los Angeles 69936060480 +Alameda 10379456877 +Alameda, Contra Costa, Fresno, Kern, Kings, Mad... 7333200000 +Sacramento 5858784824 +San Diego 4870580185 +San Francisco 3813460158 +Santa Clara 3178166479 +San Francisco, Santa Clara 2998442533 +Orange 2383485407 +San Diego, Santa Barbara, Ventura 2235315130 + + Total Project GHGReductions \ + sum +County +Los Angeles 11307708 +Alameda 10518799 +Alameda, Contra Costa, Fresno, Kern, Kings, Mad... 546406 +Sacramento 1975903 +San Diego 2233426 +San Francisco 1850982 +Santa Clara 1793848 +San Francisco, Santa Clara 1538067 +Orange 1180528 +San Diego, Santa Barbara, Ventura 791432 + + dac_funding_ratio \ + mean +County +Los Angeles 0.5751 +Alameda 0.2983 +Alameda, Contra Costa, Fresno, Kern, Kings, Mad... NaN +Sacramento 0.4194 +San Diego 0.1004 +San Francisco 0.1057 +Santa Clara 0.2247 +San Francisco, Santa Clara 0.0101 +Orange 0.3994 +San Diego, Santa Barbara, Ventura NaN + + Is Benefit Disadvantaged Communities \ + mean +County +Los Angeles 0.5420 +Alameda 0.2612 +Alameda, Contra Costa, Fresno, Kern, Kings, Mad... 0.0000 +Sacramento 0.3403 +San Diego 0.0993 +San Francisco 0.1121 +Santa Clara 0.2200 +San Francisco, Santa Clara 0.1667 +Orange 0.3571 +San Diego, Santa Barbara, Ventura 0.0000 + + Is Low Income Communities + mean +County +Los Angeles 0.6534 +Alameda 0.5557 +Alameda, Contra Costa, Fresno, Kern, Kings, Mad... 1.0000 +Sacramento 0.5162 +San Diego 0.5265 +San Francisco 0.5761 +Santa Clara 0.4911 +San Francisco, Santa Clara 0.8333 +Orange 0.6461 +San Diego, Santa Barbara, Ventura 1.0000 + +Vulnerable Communities Analysis: + Total Projects Total Investment GHG Reduction +All Communities 141429 130205367663 109154512 +Disadvantaged Communities 47167 14111685888 12626682 +Low Income Communities 82595 103691488875 55739585 + +Project Types in Vulnerable Communities: + Total Project Cost \ + count +project_category Is Benefit Disadvantaged Communities +Agricultural Projects False 1490 + True 198 +Air Quality Projects False 5514 +Business - Small False 2 +Community Projects False 973 +Individual - Home Energy False 1616 + True 5811 +Individual - Vehicle False 2970 + True 1568 +Institutional - Education False 17 +Institutional - Government False 269 +Institutional - Large Business False 23 +Institutional - Nonprofit False 100 +Other False 13788 + True 552 +Transportation - Other False 64674 + True 36753 +Water-Energy Projects False 2826 + True 2285 + + \ + sum +project_category Is Benefit Disadvantaged Communities +Agricultural Projects False 824640269 + True 61874705 +Air Quality Projects False 1211608583 +Business - Small False 65376 +Community Projects False 508817727 +Individual - Home Energy False 117426101 + True 202582546 +Individual - Vehicle False 839347111 + True 447071030 +Institutional - Education False 5561672 +Institutional - Government False 16939974420 +Institutional - Large Business False 513921827 +Institutional - Nonprofit False 1402247011 +Other False 91784138193 + True 12787875145 +Transportation - Other False 1888149934 + True 584940097 +Water-Energy Projects False 57783551 + True 27342365 + + Total Project GHGReductions + sum +project_category Is Benefit Disadvantaged Communities +Agricultural Projects False 22903912 + True 1345466 +Air Quality Projects False 250801 +Business - Small False 0 +Community Projects False 154010 +Individual - Home Energy False 203507 + True 490407 +Individual - Vehicle False 699758 + True 294234 +Institutional - Education False 457 +Institutional - Government False 6580289 +Institutional - Large Business False 388663 +Institutional - Nonprofit False 4791834 +Other False 57935224 + True 9377795 +Transportation - Other False 2392628 + True 920302 +Water-Energy Projects False 226747 + True 198478 ++
+
+
+
+
+/tmp/ipykernel_738944/3838036285.py:63: UserWarning: Tight layout not applied. The bottom and top margins cannot be made large enough to accommodate all Axes decorations. + plt.tight_layout() ++
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+In [27]:
+
+
+
+
+
+import geopandas as gpd
+import matplotlib.pyplot as plt
+import numpy as np
+
+# First, summarize the data at tract level
+tract_summary = data.groupby('Census Tract').agg({
+ 'Total Project Cost': ['count', 'sum'],
+ 'Total Project GHGReductions': 'sum',
+ 'Is Benefit Disadvantaged Communities': 'mean',
+ 'Is Low Income Communities': 'mean'
+}).round(4)
+
+# Reset the index to make Census Tract a column
+tract_summary = tract_summary.reset_index()
+
+# Create a few different maps to show different aspects
+fig, ((ax1, ax2), (ax3, ax4)) = plt.subplots(2, 2, figsize=(20, 20))
+
+# Function to create a map with proper legend
+def create_map(data, column, title, ax, cmap):
+ im = ax.scatter(data['Census Tract'],
+ data[column],
+ c=data[column],
+ cmap=cmap)
+ plt.colorbar(im, ax=ax, label=title)
+ ax.set_title(title)
+ ax.axis('off')
+
+# 1. Project Count Map
+create_map(tract_summary,
+ ('Total Project Cost', 'count'),
+ 'Number of CCI Projects by Census Tract',
+ ax1,
+ 'YlOrRd')
+
+# 2. Total Investment Map
+create_map(tract_summary,
+ ('Total Project Cost', 'sum'),
+ 'Total CCI Investment by Census Tract',
+ ax2,
+ 'viridis')
+
+# 3. DAC Benefits Map
+create_map(tract_summary,
+ ('Is Benefit Disadvantaged Communities', 'mean'),
+ 'Proportion of Projects Benefiting DACs',
+ ax3,
+ 'RdYlBu')
+
+# 4. GHG Reductions Map
+create_map(tract_summary,
+ ('Total Project GHGReductions', 'sum'),
+ 'Total GHG Reductions by Census Tract',
+ ax4,
+ 'YlGn')
+
+plt.tight_layout()
+plt.show()
+
+# Let's also look at the basic statistics
+print("\nSpatial Distribution Statistics:")
+print("\nTop 10 Census Tracts by Total Investment:")
+print(tract_summary.nlargest(10, ('Total Project Cost', 'sum')))
+
+print("\nTop 10 Census Tracts by Project Count:")
+print(tract_summary.nlargest(10, ('Total Project Cost', 'count')))
+
+print("\nTop 10 Census Tracts by GHG Reductions:")
+print(tract_summary.nlargest(10, ('Total Project GHGReductions', 'sum')))
+
+# Additional visualization of investment intensity
+plt.figure(figsize=(15, 10))
+tract_summary['investment_per_project'] = tract_summary[('Total Project Cost', 'sum')] / tract_summary[('Total Project Cost', 'count')]
+plt.scatter(tract_summary[('Is Benefit Disadvantaged Communities', 'mean')],
+ tract_summary['investment_per_project'],
+ alpha=0.5)
+plt.xlabel('Proportion of Projects Benefiting DACs')
+plt.ylabel('Investment per Project ($)')
+plt.title('Investment Intensity vs DAC Benefits')
+plt.show()
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
++Spatial Distribution Statistics: + +Top 10 Census Tracts by Total Investment: + Census Tract Total Project Cost Total Project GHGReductions \ + count sum sum +1686 6.037206e+09 369 128838282 71147 +1701 6.037208e+09 246 79854037 21955 +1250 6.037113e+09 115 67551314 45625 +5715 6.073006e+09 38 65309556 1546 +3340 6.037901e+09 61 37506825 44030 +2391 6.037408e+09 207 34961176 31158 +3482 6.037980e+09 67 30672161 23886 +6642 6.081602e+09 61 26686653 14972 +7164 6.085512e+09 49 24118370 17130 +2268 6.037403e+09 81 23674333 11183 + + Is Benefit Disadvantaged Communities Is Low Income Communities + mean mean +1686 0.5176 0.0108 +1701 0.7886 0.0447 +1250 0.0087 0.0435 +5715 0.0000 0.5789 +3340 0.0000 0.9672 +2391 0.8164 0.7923 +3482 0.0000 0.0149 +6642 0.0000 0.9508 +7164 0.0000 0.0204 +2268 0.3333 0.8148 + +Top 10 Census Tracts by Project Count: + Census Tract Total Project Cost Total Project GHGReductions \ + count sum sum +1686 6.037206e+09 369 128838282 71147 +1701 6.037208e+09 246 79854037 21955 +2391 6.037408e+09 207 34961176 31158 +6443 6.077002e+09 132 616605 650 +6448 6.077002e+09 121 401633 284 +1250 6.037113e+09 115 67551314 45625 +6452 6.077003e+09 113 438571 580 +2670 6.037504e+09 103 13744994 12914 +6449 6.077002e+09 102 534104 625 +2276 6.037403e+09 99 5585697 12839 + + Is Benefit Disadvantaged Communities Is Low Income Communities + mean mean +1686 0.5176 0.0108 +1701 0.7886 0.0447 +2391 0.8164 0.7923 +6443 0.0985 0.9848 +6448 0.0661 0.9835 +1250 0.0087 0.0435 +6452 0.0885 0.9469 +2670 0.1942 0.0583 +6449 0.1275 0.9804 +2276 0.0000 0.1818 + +Top 10 Census Tracts by GHG Reductions: + Census Tract Total Project Cost Total Project GHGReductions \ + count sum sum +419 6.009000e+09 13 6718556 735067 +7330 6.091010e+09 8 1842014 730641 +7963 6.113011e+09 21 9324110 297675 +7411 6.095253e+09 17 2631219 163765 +4432 6.061022e+09 13 1199979 78245 +1686 6.037206e+09 369 128838282 71147 +3677 6.053011e+09 26 5634176 53453 +1250 6.037113e+09 115 67551314 45625 +3340 6.037901e+09 61 37506825 44030 +4166 6.059086e+09 84 18951369 32458 + + Is Benefit Disadvantaged Communities Is Low Income Communities + mean mean +419 0.0000 0.1538 +7330 0.0000 0.7500 +7963 0.0000 0.1429 +7411 0.0000 0.2353 +4432 0.0000 0.0000 +1686 0.5176 0.0108 +3677 0.7692 0.3462 +1250 0.0087 0.0435 +3340 0.0000 0.9672 +4166 0.3214 0.2262 ++
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+Analysis of Spatial Distribution and Environmental Justice Implications¶
Key Findings:¶
-
+
- Individual vs. Large Project Analysis +Key Findings: +
-
+
Individual Program Distribution:
+-
+
- Home Energy/Solar Programs:
-
+
- Strong equity performance (78% benefit DACs) +
- High DAC funding ratio (0.65) +
- 93% benefit low-income communities +
- Lower total investment ($320M) +
+ - Vehicle Programs:
-
+
- Lower equity performance (35% benefit DACs) +
- Low DAC funding ratio (0.17) +
- Only 41% benefit low-income communities +
- Higher individual costs (~$283K vs $43K for solar) +
+
+- Home Energy/Solar Programs:
Large Project Performance:
+-
+
- Higher total investment ($127B) +
- Highest total GHG reductions (103.5M tons) +
- Mixed equity outcomes:
-
+
- Low DAC funding ratio (0.28) +
- Only 6.6% benefit DACs +
- But 56% benefit low-income communities +
+
+
-
+
- Spatial Distribution Analysis +Key Findings: +
-
+
Geographic Concentration:
+-
+
- Urban dominance (LA: 27% of projects, 54% of funds) +
- Top 3 counties (LA, Alameda, San Diego) = 37% of projects +
- Clear urban-rural divide in investment distribution +
+Environmental Justice Implications:
+-
+
- DACs receive 33% of projects but only 11% of funding +
- Low-income communities show better performance (58% of projects, 80% of funding) +
- Significant variation in DAC benefit rates across regions +
+
Next Steps:
+-
+
- Individual vs. Large Project Analysis +
-
+
- Detailed program design analysis:
-
+
- What makes home energy programs more successful at reaching DACs? +
- What barriers exist in vehicle programs? +
- How can large project DAC benefits be improved? +
+ - Cost-effectiveness comparison:
-
+
- Calculate and compare GHG reduction per dollar across programs +
- Analyze trade-offs between equity and efficiency +
+ - Policy recommendations:
-
+
- Identify best practices from successful programs +
- Suggest modifications for underperforming programs +
+
-
+
- Spatial Distribution Analysis +
-
+
- Enhanced geographic analysis:
-
+
- Create detailed maps showing investment patterns +
- Analyze urban-rural disparities +
- Examine relationship between CalEnviroScreen scores and investment patterns +
+ - Environmental justice metrics:
-
+
- Develop composite indicators of investment equity +
- Analyze temporal changes in distribution +
- Compare to demographic and socioeconomic patterns +
+
-
+
- Integration and Synthesis +
-
+
- Combine findings from both approaches to:
-
+
- Identify overlapping patterns +
- Develop comprehensive policy recommendations +
- Create framework for evaluating climate investment equity +
+
-
+
- Specific Data Analysis Tasks: +
# 1. Create program efficiency metrics
+program_efficiency = data.groupby('project_category').agg({
+ 'Total Project GHGReductions': 'sum',
+ 'Total Project Cost': 'sum'
+}).assign(
+ efficiency = lambda x: x['Total Project GHGReductions'] / x['Total Project Cost']
+)
+
+# 2. Analyze spatial equity
+spatial_equity = merged_data.groupby('CES_quartile').agg({
+ 'Total Project Cost': ['sum', 'mean', 'count'],
+ 'Total Project GHGReductions': ['sum', 'mean']
+})
+
+# 3. Create temporal analysis
+temporal_patterns = data.groupby(['project_category', 'Date Operational']).agg({
+ 'Total Project Cost': 'sum',
+ 'Is Benefit Disadvantaged Communities': 'mean'
+})
+Would you like me to focus on developing any of these next steps in more detail? We could:
+-
+
- Create specific visualizations for the paper +
- Develop detailed policy recommendations +
- Conduct additional statistical analyses +
- Create a framework for program evaluation +