California has tens of thousands of active oil and gas wells, concentrated in Kern, Los Angeles, and Ventura counties. This tool maps active and idle wells and shows what residents living near them say about plugging and who should pay, based on the 2035 Initiative's November 2025 survey of residents across five counties.
Use the Explore the map panel to overlay county survey data and district boundaries. See the Public Opinion Data tab for survey charts, and Methods for survey and data details.
Survey results from the 2035 Initiative's November 2025 survey of residents living near oil and gas wells in California's five oil-producing counties. Click any county row to fly the map to that county.
Source: 2035 Initiative November 2025 Survey. Weighted by age and sex to 2023 ACS estimates. San Luis Obispo has a small sample (n=83); interpret with caution.
Rows are randomly assigned experimental conditions. Each respondent saw only one framing.
The survey included a framing experiment testing whether adding an environmental justice context increases support for three oil and gas regulatory policies. Each respondent received one of three randomly assigned framings: no EJ context, a children's health frame, or a racial justice frame. Results are shown by policy measure below.
EJ frames added text: children = "these facilities may particularly affect school-age children" · race = "this may particularly affect Black, Latino, and Indigenous populations"
Rows are randomly assigned experimental conditions. Each respondent saw only one framing.
Note: stop-permits policy has lowest support of the three measures.
Nearly 9 in 10 residents living near oil and gas wells support a program to plug idle and abandoned wells, including strong majorities of conservatives, moderates, and liberals.
86.9% support, 59.8% stronglyMore than two-thirds of respondents said oil and gas companies should fund a well plugging program, far ahead of property owners (11%), state government (9%), and the federal government (8%).
68.9% say oil & gas companies should paySupport is highest when plugging is framed around jobs for oil and gas workers (91%), compared with abandoned well hazards (87%), cost burden on companies (85%), or land use (84%).
91.0% support with the jobs frameAdding a children's health or racial justice frame slightly raises support for stronger oil and gas regulations compared with a no-frame baseline, with the racial justice frame producing the largest boost.
+2.8 points for enforcement with the racial justice frame2035 Initiative at UC Santa Barbara. (2025). November 2025 Survey of California Oil-Producing County Residents. Santa Barbara, CA: University of California, Santa Barbara. Retrieved from https://2035initiative.com
Well location data: California Department of Conservation, CalGEM WellSTAR database (queried live). County and district boundaries: U.S. Census Bureau TIGER/Line web services.
The figures and choropleth data shown in this tool are drawn from the 2035 Initiative's November 2025 Survey of California Oil-Producing County Residents, conducted by the 2035 Initiative at UC Santa Barbara. The survey targeted residents living within a 320-meter (~1,000-foot) buffer around active oil and gas well locations, meaning all respondents live in close proximity to oil or gas infrastructure.
The survey was fielded across six California counties with significant oil and gas production: Kern, Los Angeles, Ventura, Orange, Santa Barbara, and San Luis Obispo. Sample sizes vary by county. The Orange County sample was too small to report and is excluded from the figures and map. Santa Barbara (n=636) and San Luis Obispo (n=83) data are from the November 2025 tricounty wave.
The survey was conducted via online panel, with respondents recruited through postal mailings to residential addresses within the 320-meter well buffer. Fieldwork took place in November 2025.
Responses were weighted by age and sex to match 2023 American Community Survey (ACS) 5-year block-group population estimates for the sampled geographic areas.
Support for well plugging
Response options: Strongly support / Somewhat support / Somewhat oppose / Strongly oppose.
The survey included an experimental condition (plugcond) that randomly varied
the sentence completion "Plugging these wells…" across four frames: cost burden on companies,
abandoned well hazards, jobs for workers, and land use opportunity. The Public Opinion Data
tab shows results for each frame separately.
Who should pay
Respondents selected from: oil and gas companies; property owners who lease their land for oil and gas production; state government; federal government; county government.
Percentages displayed in the charts are county-level crosstab frequencies read directly from the survey output files. "All CA respondents" reflects the pooled sample across all six surveyed counties (labeled Topline in the source data).
The county choropleth map layers use discrete color bins to represent survey response distributions across counties. Bin thresholds were chosen to reflect meaningful breaks in the data rather than equal intervals.
No additional statistical modeling was applied. Specifically, no multilevel regression and poststratification (MRP) or regression adjustment was used. The tool displays crosstab averages as provided in the survey data files.
Well locations are retrieved live at page load from the California Department
of Conservation's CalGEM WellSTAR database via its ArcGIS REST API
(MapServer layer 0). Active wells are filtered by WellStatus='Active';
abandoned/idle wells by WellStatus='Idle'. Active wells load automatically;
the abandoned/idle layer loads on demand when selected. Well counts reflect the current state
of the database at the time the page is loaded and may differ from published static reports.
On the map, individual well markers are grouped into proportional clusters: bubble size scales logarithmically with the number of wells in each cluster, so denser areas appear as visibly larger circles. Active well clusters are shown in orange; abandoned/idle well clusters in grey. Individual wells become visible as you zoom in. The wells legend updates to reflect whichever well type(s) are currently displayed.
County boundaries are fetched live from the U.S. Census Bureau's
TIGER/Line web services (State_County/MapServer/1, filtered to California).
The default map view — Wells by county — shades each county by well count using a
five-class sequential color scale. A "Shade by" sub-option in the panel lets you switch
between active and abandoned/idle well counts for this view.
California Senate and Assembly district boundaries are fetched live from
the Census TIGER Legislative MapServer (MapServer/1 for Senate,
MapServer/2 for Assembly, both filtered to STATE='06').
Each district view includes a "Shade by" sub-option to shade districts by either active
or abandoned/idle well count. Selecting the abandoned/idle option automatically loads
the abandoned/idle well layer if it has not yet been fetched.
The optional CalEnviroScreen 4.0 overlay displays cumulative pollution burden and population vulnerability scores for California census tracts, as published by the California Office of Environmental Health Hazard Assessment (OEHHA). Scores are expressed as statewide percentiles (0–100); higher values indicate greater combined environmental and socioeconomic burden. All ~8,000 California census tracts are fetched in full when the layer is first selected and cached for the remainder of the session, so no additional requests are made when panning or zooming.
This interactive web tool was built with the assistance of Anthropic's Claude (Claude Sonnet 4.6). AI was used to help write and structure the application code (the map and subsequent tabs) and to organize the survey and data-layer documentation into plain-language explanations. The underlying survey, statistical analysis, and modeled estimates are the work of the research team; all figures shown here derive from their data and methods, and all AI-generated language was reviewed and revised by the research team. The survey data, the MRP estimates, and statistical results all describe real survey respondents.