Hi Luis,

I'm not sure I understand your question: what do you want to regress on LST? It may be easier to help if we understand better what you are after...

Michel.

Thank you Michael! My idea is very simple, I would like to use LST (in an urban area obtained from LANDSAT imagery) as an independent variable in a regression model where dependent variables are going to be a set of urban form indicators. The idea is to ascertain to what extent those urban form factors can affect LST, which are more important and what kind of recommendation can be drawn as urban design guidelines...

Dear Luis,

I think you are addressing a very complex question. At the same time, there has been a lot of work done in that area already. A quick search of the web with the keywords "city structure temperature landsat" yielded 561,000 results, so I'm sure you'll find a few references of interest. For example:

1. http://link.springer.com/article/10.1007%2Fs00024-013-0685-7

2. https://www.researchgate.net/publication/233011110_Surface-and_Air-Temperature_Studies_in_Relation_to_Land_UseLand_Cover_of_Nagpur_Urban_Area_Using_Landsat_5_TM_Data

3. http://www.isprs.org/proceedings/2011/ISRSE-34/211104015Final00211.pdf

4. https://www.jstage.jst.go.jp/article/jgeography1889/104/2/104_2_225/_article

Good luck. Michel.

Article Surface-and Air-Temperature Studies in Relation to Land Use/...

Thank you!

Dear Luis,

 In order to Model Heat Islands (this si the way to go for urban areas) you do have different alternatives –numerical, empirical and scale modeling approaches

One way can be to measure the relative warmth of air temperature near the ground (canopy layer) that is feasible with Landsat data, but you will need to use different seasons to be able to compare and come up with good explanatory factors.  Still you do have different models you can choose like:

BLUHI Boundary Layer Urban Heat Island

CLUHI Canopy Layer

Urban Heat Island

SUHI Surface Urban Heat Island

According to the model and resolution that you need you can use different sensors or a combination of LIDAR and VHR data, as well as thermal data but you will also need local temperatures taken in different points of the urban areas to calibrate your models (e.g. green vs. cement) and other data like wind that is a key variable among others. As explained by Michel, the amount of papers on the topic is quite important you can begin with RS Journals like PERS, IJRS and also look into climatological Journals to learn more about specific models and sensors

Have fun ! Cheers

Thank you very much Sandra! Looking forward to seeing you in Portland/Tartu!!!

abrazo

Luis

you can follow this article https://www.researchgate.net/publication/301797360_Analyzing_land_surface_temperature_distribution_in_response_to_land_useland_cover_change_using_split_window_algorithm_and_spectral_radiance_model_in_Sundarban_Biosphere_Reserve_India

Article Analyzing land surface temperature distribution in response ...