Evaluation of climate adaptation interventions using observational data

A warming climate is having demonstrable negative impacts on a range of societal outcomes, and understanding what actions limit these negative impacts is a key research and policy question. We seek studies that will help to rapidly build the evidence base on “what works” in climate adaptation. Specifically, we are interested in new evidence on interventions that reduce climate impacts, where we understand “intervention” very broadly to include everything from individual actions (e.g., using an air filter to reduce residential indoor wildfire smoke pollution) to climate-specific community/policy interventions (e.g. the use of community heat alerts to reduce temperature morbidity) to not-obviously-adaptation-focused interventions (e.g., the impact of Medicaid expansion on temperature-related mortality).  We seek quantitative evidence on the extent to which a given action moderates a climate impact.  Example “findings”: “Community heat alerts reduced the effect of a 30C day on ED visits by 20%”; “a program that subsidized air filters reduced indoor air pollution during smoke events by 50%”; etc. Examples in the existing literature are listed below. Of additional interest is information on the implementation costs of the action/intervention; although data on costs is not a requirement for funding, proposals with some ability to estimate costs will be given extra weight. We will accept proposals on a rolling basis.  Review of proposals will begin on Aug 25th, 2025, and continue thereafter on a rolling basis. 

Process and expectations:

• Research team. All research teams must have a Stanford PI who submits the proposal. Subawards can be made to collaborators external to Stanford.  Non-faculty researchers (e.g. graduate students) will need a faculty collaborator who can serve as PI – see below.
• Proposal. Your proposal should describe what climate impact you are interested in, what adaptation intervention you want to evaluate that has the possibility of lessening that impact, and what your specific empirical strategy is for identifying the impact of that intervention.  There is no need to describe why climate change is important or why adaptation is needed. You can get right into the specifics. See below for more detail. Included in your 2 pages should be a 1 paragraph budget justification and statement of work that tells us how much money you need (indirect costs are not allowed) and how you will spend the money. We have done a lot of these studies and have fairly precise knowledge of what they cost, so please reveal your true need and we will try to fund it.  We are not announcing funding limits so as to limit the tendency for budgets to magically come in a few dollars below the maximum.
• Proposals will be evaluated on a rolling basis and reviewed rapidly, and in some cases we might iterate with you on the proposal.  If we receive more high quality proposals than we can initially fund, we will randomize which receive funding.
• If your project is funded, we require a 1-page pre-analysis plan that will be posted publicly (much of which can be culled from your proposal, see below). This is due immediately prior to funds being wired.  This will succinctly specify the climate impact you are addressing, the intervention you are evaluating, and your proposed identification strategy and empirical specification. We will have an option to embargo the PAP until the draft results are due.
• A 1 paragraph project update due 6 months after funds land in your account.
• A draft, internet-ready set of initial reduced-form results that carry out your pre-analysis plan, written as a short pre-print, due 9 months after receiving funding in your account, and a github repo (or equivalent) for reproducing your main result that will be posted publicly;  the normal exceptions will be made for restricted access data.   What you do with the results after this is completely up to you, and initially posted pre-prints will be versioned and can be revised.  We will have a project dashboard that shows what projects have been funded and whether they have met these agreed-upon timelines. 

To both meet this ambitious timeline and to help keep costs down, we will employ multiple approaches to reducing costs and timeline:

• We have a highly optimized climate data pipeline to quickly generate most climate exposures for any geography of interest, and we can run it for you if you provide shapefiles/coordinates – or, if it’s a common set of spatial units (e.g. adm0,1,2,3), we can probably already provide it.
• We have access to some outcome data that might be useful (e.g. georeferenced mortality and morbidity datasets in the US, EU, Latin America), and might know others with other datasets.  
• We even possibly have highly-qualified research assistant labor to contribute, if such labor is needed and more convenient to hire at Stanford than at your institution. 

Indeed, a successful project proposal need not actually have either exposure or outcome data nor even the labor to combine it, but must at minimum have (1) intervention data that can be plausibly linked to available exposure and outcome data, and (2) a clear plan for how to isolate the impact of the intervention on a given climate impact. 

Things we will likely not fund:

• Prospective RCTs on adaptation interventions. These are wonderful and important but almost surely too expensive for us.
• Projects without a focus on an intervention. Studies that demonstrate heterogeneous effects of climate by income or demographics are useful but are not the focus of this call. Similarly, studies on what is sometimes termed “autonomous adaptation”, or what individuals or firms are already doing to adapt to a changing climate, are very useful but not our focus. We are looking for evaluations of actions that individuals, communities, or local, regional, or national policymakers could take to reduce climate impacts that most are not already taking, perhaps due to some constraint or market friction. 

Some great examples of observational adaptation studies that provide a model of the sorts of studies we’re after:

• Mullins, Jamie T., and Corey White. "Can access to health care mitigate the effects of temperature on mortality?" Journal of Public Economics 191 (2020): 104259.
• Cohen, François, and Antoine Dechezleprêtre. "Mortality, temperature, and public health provision: evidence from Mexico." American Economic Journal: Economic Policy 14.2 (2022): 161-192.
• Shrader, Jeffrey G., Laura Bakkensen, and Derek Lemoine. Fatal errors: The mortality value of accurate weather forecasts. NBER WP No. w31361, 2023.
• Sarmiento, Luis, et al. "Income Shocks, Adaptation, and Temperature-Related Mortality: Evidence from the Mexican Labor Market." (2024).
• Garg, Teevrat, Gordon C. McCord, and Aleister Montfort. "Can social protection reduce environmental damages?" J. Environmental Economics and Management 131, 2025.
• Annan, Francis, and Wolfram Schlenker. 2015. “Federal Crop Insurance and the Disincentive to Adapt to Extreme Heat.” American Economic Review 105 (5): 262–66.
• Pople, A., Hill, R., Dercon, S., & Brunckhorst, B. (2024). The importance of being early: anticipatory cash transfers for flood-affected households.
• Collier, Benjamin L., Sabrina T. Howell, and Lea Rendell. After the storm: How emergency liquidity helps small businesses following natural disasters. 2024.  NBER WP No. w32326.