Examining the Spatial Variability of Summer Heat in New York City Using Landsat Land Surface Temperature

Author

• Ryan Kabir, CUNY City CollegeFollow

Date of Award

2026

Document Type

Thesis

Department

Earth and Atmospheric Sciences

First Advisor

James F. Booth

Keywords

temperature, environment, satellite

Abstract

Extreme heat is the leading weather-related cause of death in the United States, and in New York City (NYC) approximately 370 heat-related deaths occur annually on average. The spatial distribution of surface heat exposure across NYC is highly uneven, low-income neighborhoods and communities of color consistently face higher temperatures, yet the intraurban spatial structure of land surface temperature (LST) during heatwave events has not been systematically characterized at high resolution. The New York City Panel on Climate Change Fourth Assessment (NPCC4) identifies this as an explicit research gap, noting that its own projections, based on global climate models downscaled to point weather stations, describe citywide trends rather than the neighborhood-scale thermal structure that drives differential surface heat exposure.

This thesis attempts to address that gap by using Landsat Collection 2 Level 2 thermal imagery from Landsat 5, 7, 8, and 9 to examine the spatial variability of LST across New York City during summer heatwave and non-heatwave periods. Summer scenes from June through August spanning the Landsat archive are classified as heatwave or non-heatwave based on Central Park daily maximum temperature records, quality-filtered for cloud cover and valid pixel fractions, and analyzed using a three-map workflow: raw LST, citywide anomaly, and borough-normalized anomaly. Composite anomaly maps are produced by averaging across all heatwave and non-heatwave scenes separately. Two urban-only masking approaches, a land-use classification mask and a cool-threshold anomaly mask, are applied to isolate the thermal signal of the built environment from the confounding influence of parks and waterbodies.

The central finding is that heatwave conditions raise absolute LST citywide but do not substantially reorganize the spatial structure of thermal anomalies. The standard deviation of citywide anomalies increases by only 0.19°F between non-heatwave and heatwave composites, indicating that synoptic heatwave forcing amplifies the urban thermal field approximately uniformly rather than preferentially intensifying existing hotspots. Persistent hotspots are identified in southeastern Brooklyn, inner Queens, and the South Bronx across both regimes, supporting the interpretation that intraurban surface thermal inequality is a structural land cover outcome. Park and vegetation masking shifts the citywide reference mean upward by approximately 1.5°F and reveals that much of the apparent spatial variability in unmasked maps is attributable to park-driven cooling. Both masking approaches produce consistent results, supporting the robustness of the urban-only thermal signal. These findings directly address the NPCC4 research gap and have practical implications for spatially targeting heat adaptation interventions, including tree canopy programs, cool roof incentives, and cooling center placement.

Recommended Citation

Kabir, Ryan, "Examining the Spatial Variability of Summer Heat in New York City Using Landsat Land Surface Temperature" (2026). CUNY Academic Works.
https://academicworks.cuny.edu/cc_etds_theses/1330