Why S.F. neighborhoods are up in arms over new environmental rankings that could cost them money

California recently updated the system it uses to decide which communities get top priority for state environmental cleanup funding, and dozens of groups in the Bay Area aren’t happy with it. 

In July, the Office of Environmental Health Hazard Assessment released its fifth version of CalEnviroScreen, a data tool that maps environmentally burdened communities in California. The tool added two new indicators – diabetes prevalence and small air toxic sites, such as gas stations – and incorporated updated metrics, such as levels of lead detected in children. 

But Bay Area community advocates say the new data in the final version dilutes the ranking of some neighborhoods, making them seem less polluted than they were before despite a lack of environmental improvements. The change means those communities could lose out on state funding. 

That’s because the California Environmental Protection Agency uses CalEnviroScreen to define what communities are “disadvantaged.” Under Assembly Bill 1550, the agency must direct 25% of so – called cap-and-invest funds — money paid by polluting companies and deposited into the state’s Greenhouse Gas Reduction Fund – to be set aside for disadvantaged communities.

In the draft version of CalEnviroScreen 5.0, six census tracts in San Francisco’s Chinatown and Tenderloin neighborhoods met the threshold for being environmentally burdened communities because they ranked in the top 25% on the index. 

The final version knocked those tracts out of the rankings, meaning they’d be less likely to receive money from the greenhouse gas fund. Other tracts facing pollution, water access  and housing challenges – including in the South of Market and the Mission District – didn’t make the top 25% at all.

Rankings changed statewide, but other densely populated urban areas — such as southeast Los Angeles and parts of Long Beach — still rank in the top 25% of the updated tool.

The final version “indicates the total failure of environmental science and how we approach these issues in the state of California,” said Eddie Ahn, executive director of Brightline Defense, a public policy organization that represents underserved communities. 

State environmental officials said census tracts statewide saw slight changes in scores between the draft and final versions because they added more recent data and made some corrections.

“CalEnviroScreen scores provide a statewide ranking of census tracts, rather than an absolute measure of environmental health burdens, to identify the most impacted communities,” said Amy Gilson, a spokesperson for the health hazard assessment office, in a joint statement with CalEPA.  

Gilson added that making sure disadvantaged communities are accurately reflected is a top priority for CalEPA. 

A subjective model? 

To update the CalEnviroScreen tool, the health hazard assessment office worked with eight community organizations and hosted seven in-person and virtual workshops to gather feedback from communities throughout the state. 

The agency works with communities, academics and other government agencies to determine ways to better characterize pollution burden and population vulnerability, said Álvaro Alvarado, CalEPA’s supervising toxicologist, in a January interview with CalMatters. 

“It’s a constant work in progress,” he said. 

The tool combines 23 indicators – from air pollution to garbage dumps – into a single index meant to help identify which California communities are most environmentally burdened.

But one researcher argues that combining indicators into a single composite score introduces subjectivity. That might be the reason more densely populated tracts in the Bay Area rank lower in CalEnviroScreen 5.0. 

“It’s a way of mashing a bunch of different numbers together,” said Benjamin Huynh, an assistant professor at Johns Hopkins Bloomberg School of Public Health. “Scientifically, it’s been known for decades that you’re not actually supposed to make decisions based on these composite indicators. They’re really just meant to be exploratory tools, just to get a sense of what the data show.” 

Because the model combines a mix of indicators to arrive at a single score, some areas that struggle with high air pollution, for instance, may be pushed up in the rankings compared to an area that doesn’t have as much air pollution. But that region may still struggle with other environmental factors that the tool does not consider.  

For instance, San Francisco neighborhoods may struggle with bad indoor air quality, but the tool doesn’t have an indicator for indoor air quality.

“We have data for that in San Francisco, but we don’t have data for that across the entire state,” Huynh said. 

Gilson said the agency uses the current model because it supports broad applicability, transparency and fairness. It follows recommendations for the National Academies of Science.  

Communities fear funding losses 

Disadvantaged communities have received at least $7.2  billion since 2015. That 25% floor is fixed by statute, but the size of the pool it applies to is not: state lawmakers have been negotiating over how much revenue flows into the Greenhouse Gas Reduction Fund and how it gets divided, meaning a fixed share of an uncertain and possibly shrinking pot.

CalEnviroScreen is also used as a guide by other local grant programs when deciding funding. 

The Tenderloin Housing Clinic, for example, has been able to help Brightline Defense distribute 300 air purifiers with funding from the James Cary Smith Community Grant program, administered by the Bay Area Air District. That program gave preference to projects serving environmental justice communities in the Bay Area, including those ranked in the top 30% of CalEnviroScreen.  

Pratibha Tekkey, director of community organizing for the Tenderloin Housing Clinic, said she doesn’t think CalEnviroScreen 5.0 reflects the reality of communities living in those neighborhoods.

“You can’t just be a scientist looking at data points and basically deciding this is what it would be,” she said. “You have to see who is living there in that community, what kind of health challenges they have, what kind of economic challenges they have”

“We feel like we’re going to be missing out on a lot of environmental‑justice‑oriented investments,” Tekkey added. 

Malcolm Yeung, chief executive office of the Chinatown Community Development Center, said the new tool puts densely populated urban areas in the Bay Area at a disadvantage. 

“Communities like Chinatown effectively are rendered non-competitive even though we’re a community with arguably the densest residential population in California,” Yeung said.

Gilson said the health hazard assessment office thoroughly reviewed comments from Bay Area advocates and evaluated the data to address potential errors. The agency made minor corrections for certain indicators– including diesel exhaust pollution, hazardous waste and fine-particle air pollution – such as accounting for more emissions sources and making “scoring method corrections.” 

“These data corrections resulted in very minor changes to the indicator scores for some Bay Area census tracts and only slightly impacted the overall share of tracts in the top 25% for the region,” Gilson said.  

Defining disadvantaged

The California Environmental Protection Agency periodically revisits its definition of disadvantaged communities. Under the current update, the agency is proposing to include census tracts in the top 25% of the new CalEnviroScreen tool; tracts in the top 5% of the tool’s pollution burden scores; and tracts that have some overlap with the previous CalEnviroScreen’s top 25%. The definition also includes tribal lands. 

CalEPA held two workshops in July to gather community feedback. Stakeholders had until Friday to submit public comments for consideration in the final definition. 

In a letter to state environmental officials, Brightline Defense proposed expanding the definition in several ways: including tracts located within a half a mile radius of tracts that score in the top 25% of CalEnviroScreen 5.0; expanding the definition to the top 35% overall; and including tracts that scored in the top 25% of CanEnviroScreen 3.0.  

“In the longer term, we urge the agency to champion broader reforms to the (disadvantaged communities) framework and funding allocation process that enable more efficient targeting of resources that meet hyper-localized needs,” Ahn wrote. 

The draft definition aims “to strike a balance of inclusivity while not diluting the designation and investments,” Gilson said.

She added communities outside of the top 25% of CalEnviroScreen are still considered for state funding  because certain state laws require low-income communities to be prioritized. 

Major changes to the model, she added, would require long-term evaluation and careful research. Suggestions on the model will be part of the agency’s evaluation process for the next version of the tool, which the health hazard assessment office will begin working on in 2027, she said.