To calculate the land surface temperature, you have to calculate the emissivity of the area. There are several methods of calculating the emissivity of an area, but the most commonly used one is the NDVI approach. Try to go through this document http://esatjournals.org/Volumes/IJRET/2014V03/I05/IJRET20140305025.pdf. This explains the extraction from landsat 8. I haven't worked much on Landsat 7 and 5, but they  use the same method as they both have a single Thermal Infrared Band. You should also note that, temperature data extracted from satellites also needs to be verified on the ground so as to prove that the measurements are correct. I hope this will be helpful to you.

Hello Venkata,

There are a number of Algorithms to do the job.

If you know LSEs already then single-channel, multi-channel and multi-angle methods can be used. If LSEs are not known, then you can use either stepwise retrieval, simultaneous retrieval of LSEs and LST with known atmospheric info. and simultaneous retrieval with unknown atmospheric info. can be used.

In any case it is good to go through this nice review article on Satellite Based LST retrievals by Li et al..  You might also want to check out this handy youtube tutorial on LST estimation using Landsat 8 TIR Bands but bare in mind that the video does not incorporate atmospheric correction which is essential for LST.

Milton has already mentioned the importance of ground validation. I would like to add to the fact that it is rather crucial to realize in the end, Satellite Images are time stamps, captured in a moment of time. Therefore they can to a certain extent give you the "temperature of the skin of the earth" to some µm level at a certain location and at a certain instant. If then for an instant you can extend it with a time series analysis of a certain location by analyzing multiple LSTs. I think the time component would add much more value to the LST rather than looking at just one LST computed from a single Image. It gives a better understanding of the variability of LST at that particular location.

http://www.sciencedirect.com/science/article/pii/S0034425712004749

https://www.youtube.com/watch?v=uDQo2a5e7dM

Dear milton and james 

Thank you for your suggestions and i will go forward to calculate LST.

Dear milton,

While LSE was estimated using NDVI threshold method, does we use TOA refelctance values or DN for calculating NDVI? 

Dear Milton,

I have seen the attachment you gave. In that report he mentioned NDVIs, NDVIv can you tell me how to generate it from NDVI. Also he didn't mention about how he calculate the atmospheric water vapour content. Is this atmospheric water vapour varies from place to place and time of acquisition. Also comment on how to calculate  atmospheric water vapour content of Landsat 8 data.  

Dear Vanama,

There are several algorithms which are used to estimate the LST of a place from the TIRS. He used the split window algorithm. One of the problems of the algorithm is the difficulty of estimating the atmospheric water vapour. He obtained the NDVIs and NDVIv after classification. Go through this document I have attached, it is a review on LST estimation methods, I am sure that it is going to give you an insight on the matter. To calculate the NDVI, he used the Red (band 4) and NIR (band 5) bands.

you can follow this article for landsat 5 and landsat 8 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 ...

For Landsat8 imagery, one option is the https://grass.osgeo.org/grass72/manuals/addons/i.landsat8.swlst.html, a GRASS GIS addon.

ps- Late regarding the date when the question was posted. Yet potentiall usefull today.

Nikos, thanks for sharing this link for implementing the Split-Window (SW) algorithm to estimate LST in GRASS. Even though it is recommended not to use the SW to estimate LST using Landsat 8 TIRS, it is good to know that such an add-on exists.

https://landsat.usgs.gov/how-are-thermal-infrared-sensor-tirs-thermal-bands-aboard-landsat-8-used

Thanks for heads up James. The issue is known and has been partially tackled and acquisitions were/are reprocessed. See https://landsat.usgs.gov/landsat-8-data-users-handbook-appendix-a for uncertainty levels for both bands 10 and 11.

Once the "stray-light" issue is understood, the whole archive will be reprocessed. Still, Landsat8 TIRS is the "best" source for LST, publicly accessible at high resolution. Depending on the study and the post-processing, uncertainty of +/- 1-2 degrees can be regarded as acceptable.

Nikos, thanks for the clarification. I need to take a closer look at the information provided in appendix a of the handbook. I read the following statement in the FAQ section and didn’t go beyond that to use the SW: ‘it is recommended that users refrain from using band 11 data in science studies.’

For completeness: https://landsat.usgs.gov/april-25-2017-tirs-stray-light-correction-implemented-collection-1-processing

you can follow this article

https://www.researchgate.net/publication/311518778_Automated_Land_Surface_Temperature_Retrieval_from_Landsat_8_Satellite_Imagery_A_Case_Study_of_Kahramanmaras_-_Turkey

Please find attached my published article with can help You.

Regards

In one of my project i have estimated LST by applying a structured mathematical algorithm viz., Split-Window (SW) algorithm. It based on brightness temperature of two bands of TIR, mean and difference in land surface emissivity for estimating LST of an area. The algorithm is :

LST = TB10 + C1 (TB10-TB11) + C2 (TB10-TB11)2 + C0 + (C3+C4W) (1- ε) + (C5+C6W) ∆ ε

LST - Land Surface Temperature (K)

C0 to C6 - Split-Window Coefficient values (Skokovic et al, 2014; Sobrino et al, 1996; 2003; Shaouhua Zhao et al, 2009)

TB10 and TB11 – brightness temperature of band 10 and band 11 (K)

ε – mean LSE of TIR bands

W – Atmospheric water vapour content

∆ ε – Difference in LSE

Though there are number of algorithms for this hence, the following link may useful in this regards

(1). https://geogeek.xyz/how-to-calculate-land-surface-temperature-with-landsat-8-images.html

what about if I use ENVI 5.3 rather Arc Map?

The standard procedure for LST computation has been explained in

Article Land Surface Temperature Responses to Land Use Land Cover Dynamics

hello

you can choose the link : https://landsat.usgs.gov/april-25-2017-tirs-stray-light-correction-implemented-collection-1-processing

Hi Venkata Sai Krishna Vanama

Though there are number of algorithms for this hence, the following link may useful in this regards

https://geogeek.xyz/how-to-calculate-land-surface-temperature-with-landsat-8-images.html

Please see the paper and the related toolbox (Supplementary Material) at:

Article Land Surface Temperature Retrieval from Landsat 5, 7, and 8 ...

LST Model in Erdas Imagine software

Hi Venkata Sai Krishna Vanama

I agree with S. Bonafoni

Good luck.

Please see the paper

Article Land Surface Temperature Retrieval from LANDSAT-8 Thermal In...

You may want to look at:

https://www.researchgate.net/publication/332676563_Land_Surface_Temperature_Retrieval_from_LANDSAT-8_Thermal_Infrared_Sensor_Data_and_Validation_with_Infrared_Thermometer_Camera

https://www.researchgate.net/publication/339748426_Combination_of_visible_and_thermal_remotely_sensed_data_for_enhancement_of_Land_Cover_Classification_by_using_satellite_imagery

You May take help from this article Article Spatio-Temporal Analysis of Land Surface Temperature - Sikkim, India

Hi,

You can use the GEE code. You just need to run a JSON code.

Hello,

you can use the GEE cod. and use the User Guide of this platform.

https://developers.google.com/earth-engine/