How do i calculate Land Surface Temperature Landsat 8 level 2 image?
Do I have to convert DN (TOA) to radiance, then convert radiance to brightness temperature, then emissivity, and finally the LST calculation?
OR is it enough to rescale the thermal bands and convert kelvin to celsius degrees (since level 2 is already corrected)?
Lea Lea, you should be able to get the desired solution of calculating Land Surface Temperature (LST) derived from Landsat 8 Collection 2 Level 1 Thermal Infrared Sensor (TIRS) Band 10 data if you follow the first step, i.e., converting DN (Digital Number) to radiance, then to brightness temperature, and finally applying emissivity correction to get the accurate LST.
However, if you browse through the L8-9 C2 L2 Science guide (link attached), Surface Temperature (ST) / LST product is already provided bundled with the SR and other bands. The only thing that you need to do is use the scaling and offset values from page 12 on ProductID_ST_B10 to get the actual LST values. Emissivity is also available as a separate product that was estimated from ASTER GED. Having said that, I don't think night time L8-9 scenes have ST products!
I'm curious to know which LST product is more accurate, i.e., manually calculated LST or USGS ST product? I suppose it depends on how you calculate Emissivity, whether through ASTER GED or NDVI method or some other way...? Do you have ground surface temperature measurements to validate the results?
https://www.usgs.gov/media/files/landsat-8-9-collection-2-level-2-science-product-guide
James Varghese Thank you for your comment!
Just to be sure - if i did rescale and offset for landsat 8 L2 band 10 and then transformed the values to degree celcius - does it concidered as LST?
Or should i use the whole formula of LST=BT/1+w*(BT/p)*In(e) including the emissivity? Do i need to calculate the emissivity? If not, how do i extract the specific emissivity area from the usgs ? And the BT from the formula is the degree band that i rescaled ?
Lea Lea let me give you an example, which will clarify your doubts. I selected a C2 L2 Landsat 8 image from USGS EarthExplorer over Israel and downloaded the Surface Temperature (ST) and the Emissivity (Emis) bands. Here are the details of the Landsat 8 Scene:
ID: LC08_L2SP_174038_20230716_20230725_02_T1
Date Acquired: 2023/07/16
Path: 174
Row: 038
After downloading the ST and Emis bands, I did the following:
I used the Multiplicative Scale Factor and the Additive Offset from the science guide to get the actual LST values in Kelvin. To get the values in °C, subtract it with 273.15
- ("LC08_L2SP_174038_20230716_20230725_02_T1_ST_B10.TIF" * 0.00341802 + 149 ) - 273.15
To get the actual Emissivity values, I used the Multiplicative Scale Factor from the science guide
- "LC08_L2SP_174038_20230716_20230725_02_T1_ST_EMIS.TIF" * 0.0001
Check attached images for reference.
In summary, when you download C2 L2 product, Land Surface Temperature (LST) or Surface Temperature (ST) is already provided as a ready to use product and by utilizing the scaling and offset values you get actual LST range. L2 products don't provide you with Brightness Temperature (BT) band or B10 Thermal band. As I said earlier, if you wish to separately calculate LST/ST, you can download the Landsat 8/9 thermal bands (B10 or B11) from the Collection 2 Level 1 data catalog and go through the manual calculation of B10/11 --> Radiance --> BT --> Emissivity --> LST.
Let me know if you need further clarification?
James Varghese Thank you for the explantations!
Just one thing that i'm not quite sure - If you're saying that i get the LST from rescaling and converting to °C - when do i use the emissivity data and for what purpose?
I'm asking because I want to check the connection between LST and NDVI before and after megafire in Australia. So i took 2 landsat 8 L2 images, did rescale and so on as you told (withought the emissivity), and extracted the pixel data to a graph (2 files attached).
Can i surely say that the proccess that you wrote is LST values? withought the emissivity? And how can i validate the outputs?
Lea Lea
This was my question too. But I understood this;
To obtain "LST" from Collection 2 Level 2 Landsat images, the following formula is sufficient: (DN * 0.00341802 + 149) - 273.15.
However, if you want to calculate LST using NDVI, Pv, and emissivity, you will need to download the Level 1 collection of Landsat.
Hey Lea Lea
For C2 L2 products Emissivity (Emis) band is provided for reference, although you can use this band for other research applications. If your purpose is to only get the LST output image, then you only need the ST band which ends with the following naming convention '...T1_ST_B10.TIF'
I would suggest you go through the Science Guide as it contains more details. I'm just quoting two relevant paragraphs from the guide:
'The Landsat 8-9 Surface Temperature (ST) product is generated from the single channel algorithm version 1.3.0 (derived from June 2017 version of RIT ST code). The Landsat 8-9 Collection 2 ST is derived from the Collection 2 Level 1 Thermal Infrared Sensor (TIRS) band 10 using Top of Atmosphere (TOA) Reflectance, TOA Brightness Temperature (BT), Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) Global Emissivity Dataset (GED) data, ASTER Normalized Difference Vegetation Index (NDVI) data, and atmospheric profiles of geopotential height, specific humidity, and air temperature extracted from reanalysis data."
'The ST intermediate bands are delivered with the ST product. These layer bands are used in the creation of the ST band. The ST_TRAD band contains the Level 1 thermal band converted to thermal surface radiance. The ST_URAD and ST_DRAD bands respectively contain the upwelled and downwelled thermal radiance. The ST_ATRAN band contains the atmospheric transmittance. The ST_EMIS band and ST_EMSD band contain the calculated surface emissivity and its expected standard deviation.'
If you are still in doubt you can write to USGS Support and let us know if they have some other answer!
Regarding your question about validating the output, that was my question to you in my first response. I asked previously, 'Do you have ground surface temperature measurements to validate the results?'. It is easy to generate a satellite derived LST output image but validating the results is always crucial. Refer to other research paper and see what others have done.
- Badges
- Science topic
- Similar topics
- Geoscience
- Remote Sensing