The short answer
A high-risk neighborhood cannot be found from temperature alone. Stack four data layers: satellite land surface temperature (e.g., Landsat, roughly 30 m), near-surface air temperature and humidity where people actually are (vehicle-mounted or stationary sensors), land cover and vegetation (canopy, impervious surfaces), and social vulnerability (age, income, air-conditioning access, housing). Overlaying hazard, exposure and sensitivity turns a heat map into a risk map that prioritizes the right blocks.
Key takeaways
- Surface and air temperature differ: human risk is driven by air temperature and humidity, not only by how hot asphalt gets.
- Build at least four layers: satellite temperature, field air readings, land cover and vegetation, and social vulnerability indicators.
- Satellites provide continuous coverage, but sparse revisits and clouds force warm-season composites from many clear scenes.
- Field drives in the morning, afternoon and evening on one of the hottest days capture air temperature where people are.
- Equal heat is unequal risk: an older, unshaded neighborhood without air conditioning is more vulnerable than a young, shaded one.
- Index weights are a policy choice; publish them so the map stays transparent and contestable.
Define the outcome before you map temperature
An urban heat map is a decision tool, so define which decision it will support before choosing data. The common error is equating the hottest surface with the highest risk. A useful heat-risk map merges three dimensions: hazard (how hot an area gets), exposure (who and what is there during a heat event) and sensitivity (how hard it is for residents to cope, from age and health to air conditioning and shade). Many cities combine these into one index rather than a single temperature layer.
Clarify the outcome. Is the goal to site cooling centers, target tree planting, justify a cool-roof ordinance, or inform an emergency heat plan? Each goal changes which blocks matter and which layers dominate. A school district protecting children will weight exposure and shade differently than a public-health office planning outreach to older adults living alone.
- Cooling centers and emergency response: weight exposure, access and mobility.
- Tree planting and shading: weight vegetation deficit and pedestrian routes.
- Cool-roof policy: weight roof materials and the area of exposed buildings.
Layer 1: satellite land surface temperature
Surface temperature from satellites gives a consistent, citywide baseline. Landsat Collection 2 land surface temperature products use the satellite's thermal band at roughly 30 m and are distributed analysis-ready by the USGS; annual mean and mean maximum temperature composites for selected cities go back to 1985. Landsat revisits a site about every 16 days, and cloud cover forces many scenes to be discarded, so warm-season composites are usually assembled from many clear overpasses rather than one snapshot.
Newer products help where Landsat is limited. NSF NCAR's Urban Heat MiniCubes combine Landsat 8/9 surface temperature and reflectance at 30 m with GOES geostationary brightness temperatures at about 2 km updated every 10 minutes, letting analysts track the daily thermal cycle instead of a single morning pass. Because thermal resolution and revisit frequency trade off, few analysts rely on a single sensor.
Layer 2: air temperature where people actually are
People experience air temperature and humidity, not the surface temperature a satellite sees, and surfaces can run several degrees hotter than what is felt. Measuring near-surface air temperature is therefore the second essential layer. The NOAA urban heat island mapping program, run with community partners since 2017, sends volunteers along pre-mapped routes on one of the hottest days, in the morning, afternoon and evening, with sensors recording temperature, humidity, time and location every second. In 2021, 799 volunteers recorded 1.2 million measurements across 24 communities; a 2022 round covered 14 U.S. cities plus Freetown and Rio de Janeiro.
Field readings are sparse in space, so programs model them. Machine learning combines the mobile air readings with satellite imagery to interpolate continuous maps of air temperature and heat index for morning, afternoon and evening. The NOAA program documents that neighborhoods can be roughly 20 °F hotter than nearby areas. Stationary sensors installed through the summer add a temporal check that a one-day campaign cannot provide.
Layer 3: land cover, canopy and adaptive capacity
The third layer explains why some areas overheat and what could cool them. Key indicators are vegetation and tree canopy (often from the normalized difference vegetation index, NDVI), impervious surface fraction such as asphalt and rooftops, building density, and material albedo. Sealed, dark surfaces store heat; canopy and water release it. EPA's Heat Island Community Actions Database catalogs dozens of local tree-canopy, green-roof and cool-roof programs and codes that this layer can justify.
The Local Climate Zone framework groups neighborhoods by structure and surface — compact mid-rise, open low-rise, dense trees, bare rock and so on — which predicts how each responds to sun and wind. Using such a classification converts raw temperature differences into classes you can address with specific interventions: shade structures for exposed transit stops, trees for parking-dominated zones, reflective roofs for industrial blocks.
Layer 4: who lives in the blocks
Two blocks at identical temperatures pose unequal risk if one holds many older residents without air conditioning and the other is young, air-conditioned and shaded. Census or statistical-office data adds age structure, income, housing type and tenure, home air-conditioning penetration, pre-existing health conditions and isolation. Overlaying temperature with this social layer is what turns a heat map into a heat-risk map, and it typically reveals that the hottest and least-served blocks coincide.
Methods range from a simple two-layer overlay to composite vulnerability indices built from normalized and weighted indicators. The exact weights are policy choices — publish them so the map remains transparent. In the United States this layering connects to environmental justice programs such as the federal Justice40 effort, where applications for heat-mapping projects are weighed partly by how much they address disadvantaged communities.
Turn the layers into priority blocks
Practical workflow: stack the raster layers into one grid, normalize each to a common scale such as 0–1 or 0–100, assign weights that reflect your decision, and sum them into a risk score per pixel or block. Then classify blocks into tiers — extreme, high, moderate and low — and ground-truth the top tier by walking it, checking shade, cooling centers and who is outdoors at midday. Because heat varies within a city block by block, inspect results at street scale.
Scenario tools test interventions before budgets are committed. WRI's Cool Cities Lab, launched in early 2026 for more than 25 cities, maps heat to the block and models how trees, shade structures and reflective roofs lower felt temperatures; in Hermosillo officials used the data to site a park in one of the hottest, least shaded areas, and Atlanta used analysis of reflective roofs to support a city cool-roof ordinance. Such modeling estimates the cooling benefit before a single tree is planted.
- Extreme tier: co-locate a cooling center and fund a shading program first.
- High tier: plant trees and add reflective roofs on public buildings.
- Moderate tier: adopt parking-lot shading rules and summer monitoring.
- Low tier: refresh data every few years as vegetation and development change.
Put it into practice
Heat-risk block prioritization checklist and scoring matrix
A reusable control list of the four data layers with source checks, followed by a scoring routine. For each layer, note the source, resolution and vintage, then normalize, weight and sum into a per-block risk score before field-verifying the top tier.
- Land surface temperature: list the product (e.g., Landsat Collection 2 LST), resolution and year; confirm the warm season uses clear, cloud-free scenes.
- Air temperature and humidity: record the field-drive date and hours (morning/afternoon/evening) or the period of stationary sensors.
- Vegetation: NDVI or a tree-canopy map; note the imagery source and year.
- Impervious surfaces: fraction of asphalt, rooftops and built area; assign each block a local climate zone class.
- Social indicators: age, income, air-conditioning access, housing type and health — note the census or statistics vintage.
- Normalize every layer to 0–1 or 0–100 and write down the formula.
- Assign weights per your decision goal and document them publicly.
- Sum the weighted layers into a risk score per pixel or block and split into tiers.
- Field-check the top tier: shade, access, and who is outdoors at midday.
- Model at least one intervention scenario (trees, shade, cool roofs) before committing funds.
- Set a refresh date — vegetation and built form change over a few years.
Questions people ask
What is the difference between land surface temperature and air temperature in heat mapping?
A satellite measures land surface temperature — how hot asphalt, rooftops and ground become — which can run several degrees above what a person feels. People experience near-surface air temperature combined with humidity. A useful risk map for people therefore pairs satellite surface temperature as an indicator of hot territory with field air-temperature readings, typically collected by sensors on vehicles or at stationary stations.
Which free satellite data give land surface temperature for a city?
The main open source is Landsat Collection 2 land surface temperature (thermal band, about 30 m), published analysis-ready by the USGS with annual composites for selected cities back to 1985. ECOSTRESS and geostationary GOES data add detail or temporal frequency. NSF NCAR's Urban Heat MiniCubes combine Landsat 8/9 at 30 m with GOES at about 2 km updated every 10 minutes, helping analysts track the daily cycle of heat rather than a single overpass.
How do NOAA urban heat mapping campaigns work and can my city join?
Since 2017 NOAA and partners run volunteer field campaigns in which residents drive pre-mapped routes on one of the hottest days, in the morning, afternoon and evening, while sensors log temperature, humidity and location each second. Machine learning merges these readings with satellite imagery into continuous maps of air temperature and heat index. Participation is arranged through the NIHHIS/NOAA program and typically requires a local team and partnerships; schedules and eligibility change yearly, so confirm current rounds on official NOAA pages.
Why do different maps show different hot neighborhoods?
The result depends on which metric was measured (surface or air), the time of day and date of collection, and which layers and weights entered the risk index. A morning drive and an afternoon drive can yield different heat leaders, and adding social indicators shifts priorities. When comparing maps, check the date, hours, sensor type and weighting method rather than trusting the final image alone.
How do I combine a temperature map with census data to prioritize neighborhoods?
Bring temperature and social indicators (age, income, air-conditioning access, health) to a common scale such as 0–1, assign weights aligned with your goal, and sum them into a risk score per block. Split blocks into tiers and field-check the highest tier. Publishing the formula and weights keeps the analysis transparent and useful for residents and decision-makers.
Which land-cover data should I include beyond temperature?
Add vegetation and tree canopy (typically from NDVI), the impervious-surface fraction such as pavement and roofs, building density, and surface albedo. A local climate zone classification groups blocks by structure and surface so you can match interventions — shade structures for open transit stops, trees for parking-heavy zones, reflective roofs for industrial districts. EPA's Heat Island Community Actions Database shows how cities translate such layers into ordinances and programs.
Sources and further reading
Sources were checked when this page was generated. Confirm changing dates, rules and prices with the original publisher.
- From Data to Action: How Urban Heat Mapping Campaigns Can Expose Vulnerabilities and Inform Local Heat PolicyNOAA Institutional Repository / Bulletin of the American Meteorological Society
- NOAA and communities to map heat inequities in 14 U.S. cities and countiesNOAA
- NOAA Urban Heat Islands Program (2017-2024)Center for Collaborative Heat Monitoring
- RELEASE: New Global Platform Maps Urban Heat Risks Block by Block—and Shows Cities How to Cool ThemWorld Resources Institute
- Developing an annual land surface temperature from 1985 to present for select international sitesU.S. Geological Survey
- Urban Heat MiniCubesNSF NCAR National Center for Atmospheric Research
- Heat Island Community Actions DatabaseUS Environmental Protection Agency