Land surface temperature vs. air temperature, explained
Two different things get called "temperature"
The National Weather Service number is air temperature, measured a couple of meters off the ground, in the shade, usually over grass, usually at an airport. It describes the air a person is standing in.
A thermal satellite measures something else: the radiant heat coming off whatever surface it's looking at. A parking lot, a flat rooftop membrane, a school's blacktop, a strip of grass. That's land surface temperature, and on a 91-degree afternoon it is routine for a satellite to read a sun-baked roof or an asphalt lot at 115 to 130 degrees. The forecast didn't lie. It's answering a different question than the one the thermal band is asking.
New planners often double-check the thermal layer against the forecast the first time the two numbers don't match, assuming something's broken. Nothing is broken. The forecast reports air temperature. The thermal band reports surface temperature, and the gap between the two is exactly what shows which surfaces are holding the day's heat.
Why the satellite reads hotter
Surfaces absorb and re-radiate solar energy at wildly different rates depending on what they're made of. Dark asphalt and dark roofing can sit 40 to 60 degrees above the day's air temperature at peak sun. A mature tree canopy or an irrigated park can land close to air temperature, sometimes below it. The weather station's grass plot is deliberately chosen to represent "typical" conditions, not the worst material in the neighborhood.
Land surface temperature maps where heat concentrates once the air temperature everyone already agrees on hits a built environment that is not uniform. A single stretch of rowhouses can run a 20-degree surface-temperature spread between a parking lot and the shaded lawn forty feet away, on the exact same afternoon, under the exact same sky.
Why weather stations can't show this
A weather station is a point. The city might have two or three of them, plus an airport reading. None of them sit on the roof of a brick rowhouse with no canopy, which is usually the place a health department needs eyes on. Thermal satellite imagery covers the ground between those stations at a resolution fine enough to separate one stretch of street from the next, which is the resolution a siting decision runs on. Weather stations tell you the air was hot that day. They can't tell you which part of the neighborhood had nowhere to escape it.
Reading the gap for cooling investment
For siting work, the comparison that matters is one street segment against the next: surface temperature on a rowhouse row with no street trees against surface temperature on a shaded block two streets over, read from the same satellite pass. The first will consistently run hotter than the second on every seasonal pass, regardless of what the thermometer says that particular afternoon. That pattern holds up across a season, and it's what's worth building a cooling center around or planting shade trees against.
Our Heat Exposure Mapping overlays that surface-temperature layer against building type and demographic density, so the hot spots that show up on the thermal band get checked against who actually lives there before any cooling-centre or shade-tree dollar goes out the door.
Land surface temperature and air temperature measure two different things, and once that's clear, a satellite reading hotter than the forecast stops looking like noise and starts looking like the map you were missing. If you're ready to see which blocks in your city carry that gap, we'd like to hear what you're working on.