3

I am curious about making a thermal map to display where the hot spots and where the cool spots in a community are. Obviously the roads will be hot and the shaded areas with trees will be cool but this is more to brush up on my skills.

How would I go about collecting this data for an area where none currently exists?

It will be some form of raster data but I am lost from there on out.

  • Would it have something to do with turning the color bands off and on for an arial image? – Rob Jul 3 '14 at 18:06
  • 2
    Yes and no. Thermal is a specific band or combination of bands, and you basically need to find a source that has it available at a suitable resolution. That won't be cheap, as most satellite imagery has low resolution for thermal data (say, most likely 30-90m range) so you're probably looking at a for-hire custom data collection flyover. The other option is collecting it yourself with a UAV and a thermal sensor... which again might not be cheap depending on your patience and equipment investment. – Chris W Jul 3 '14 at 20:43
  • 3
    In addition to comment of Chris W (low spatial resolution of satellite images) I would like to add: In the company where I worked we always used a cutom airbore thermal scanner data collection to determine the time of the data acquisition. In our urban climate studies we were interested in two times to analyze heat flow: one flyover shortly before the sunrise and another flyover shortly before the sunset. This is not possible with satellite data. – Jens Jul 4 '14 at 7:23
  • 1
    Very good point regarding the temporal component @jens - thermal readings will change throughout the day and depending on time of year (never mind weather). Introduction to Remote Sensing by Campbell and Wynne has an excellent chapter (9) on Thermal Imagery. They list a youtube video about home-built thermal UAV as a learning resource which might be of interest as a jumping off point to further information and resources. – Chris W Jul 4 '14 at 20:45
2

Landsat-8 (and previous missions) imagery is available for free and contains a thermal bands that could be used to perform such an analysis, but at a slightly coarser scale than you suggest. It might be possible to pan-sharpen the thermal band using the panchromatic band (15 m resolution) to improve the resolution (or maybe overlaying / interpolating) and perform a rough qualitative analysis.

0

There are many ways to get thermal data, firstly, you can use a manual radiometer like FLIR which is composed of a thermal camera, you can scan materials with that tool targeting every angle of an object, the main advantage is the low cost and the fact that you can scan materials any time of the day, secondly, using a multispectral camera mounted on an plane or UAV in order to get thermal imagery from above, the main advantage of this method is you can fly any time of the day, you can point your camera in angles, an so you can see oblique images, thirdly, processing Landsat or ASTER satellite imagery, some images are free to download (EarthExplorer), you can get images for a whole city, main disadvantage is the resolution is lower and the time of the day when the satellite collects the data do not correspond to the time when the materials are emitting the higher temperature, also the dates when you can get cloud free imagery is limited.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.