Answers are grounded in this paper and cite their sources.
Extreme heat is the leading weather-related cause of death in New York City, yet vulnerability indices built from census and land-surface inputs capture where heat is present more readily than how residents experience and respond to it. We evaluated whether 311 service requests, a form of resident-generated social sensing, can serve as a complementary neighborhood-level indicator of heat burden. Daily 311 records for 2015–2025 were linked to heatwave days defined from the NOAA Central Park station (≥3 consecutive days ≥90°F), and nine candidate complaint categories (763,986 requests) were screened using Welch two-sample t-tests, with Poisson models as a sensitivity analysis for sparse categories. Hydrant-related requests showed the strongest heatwave signal (474.8 vs 59.1 mean daily requests), followed by power outage (39.9 vs 15.6) and ventilation requests (15.6 vs 8.4; all p<0.001). Aggregated to Modified ZIP Code Tabulation Areas, all three clustered spatially (Moran's I 0.507–0.674; all p<0.001), concentrating in the Bronx and Upper Manhattan, and all three were positively correlated with neighborhood poverty (r=0.35–0.66). In spatial lag models, hydrant requests were independently associated with Hispanic population share, power outage requests with all sociodemographic predictors examined, and ventilation requests with poverty. In exploratory secondary analyses, ventilation requests corresponded most closely with heat-illness EMS dispatch rates, while cooling center density showed weak or no association with any category. Heat-related 311 requests appear to index distinct dimensions of heat burden, including informal cooling, infrastructure strain, and housing conditions, and may complement conventional heat metrics in targeting urban heat interventions.
No takes yet. Share an insight, caveat, or question.
Sahr et al. (2026) studied this question.