i.landsat8.swlst is a GRASS GIS add-on, implementing a practical Split-Window (SW) algorithm, estimating land surface temperature (LST), from the Thermal Infra-Red Sensor (TIRS) aboard Landsat 8 with an accuracy of better than 1.0 K.

Quick examples

After installation (see section Installation below), from within a GRASS-GIS session, retrieve usage details via i.landsat8.swlst --help

The shortest call for processing a complete Landsat8 scene normally is:

i.landsat8.swlst mtl=MTL prefix=B landcover=FROM_GLC

where:

• mtl= the name of the MTL metadata file (normally with a .txt extension)
• prefix= the prefix of the band names imported in GRASS GIS' data base
• landcover= the name of a FROM-GLC map product that covers the extent of the Landsat8 scene under processing. FORM-GLC products are available at http://data.ess.tsinghua.edu.cn/.

A faster call is to use existing maps for all in-between processing steps: at-satellite temperatures, cloud and emissivity maps.

• At-satellite temperature maps (options t10, t11) may be derived via the i.landsat.toar module. Note that i.landsat.toar does not process single bands selectively.

• The clouds option can be any user-defined map. Essentialy, it applies the given map as an inverted mask.

• The emissivity maps, derived by the module itself, can be saved once via the emissivity_out and delta_emissivity_out options and used afterwards via the emissivity and delta_emissivity options. Expert users, however, may use emissivity maps from other sources directly. An example command may be:

  i.landsat8.swlst t10=T10 t11=T11 clouds=Cloud_Map emissivity=Average_Emissivity_Map delta_emissivity=Delta_Emissivity_Map landcover=FROM_GLC -k -c 

For details and more examples, read the manual.

Description

The algorithm removes the atmospheric effect through differential atmospheric absorption in the two adjacent thermal infrared channels centered at about 11 and 12 μm.

The components of the algorithm estimating LST values are at-satellite brightness temperature (BT); land surface emissivities (LSEs); and the coefficients of the main Split-Window equation (SWCs).

LSEs are derived from an established look-up table linking the FROM-GLC classification scheme to average emissivities. The NDVI and the FVC are not computed each time an LST estimation is requested. Read [0] for details.

The SWCs depend on each pixel's column water vapor (CWV). CWV values are retrieved based on a modified Split-Window Covariance-Variance Matrix Ratio method (MSWCVMR) [1, 2]. Note, the spatial discontinuity found in the images of the retrieved CWV, is attributed to the data gap in the images caused by stray light outside of the FOV of the TIRS instrument [2]. In addition, the size of the spatial window querying for CWV values in adjacent pixels, is a key parameter of the MSWCVMR method. It influences accuracy and performance. In [2] it is stated:

A small window size n (N = n * n, see equation (1a)) cannot ensure a high correlation between two bands' temperatures due to the instrument noise. In contrast, the size cannot be too large because the variations in the surface and atmospheric conditions become larger as the size increases.

The combination of the brightness temperatures to estimate the LST bases upon the equation:

LST = b0 + + ( b1 + b2 * ( 1 - ae ) / ae + b3 * de / ae^2 ) * ( t10 + t11 ) / 2 + + ( b4 + b5 * ( 1 - ae ) / ae + b6 * de / ae^2 ) * ( t10 - t11 ) / 2 + + b7 * ( t10 - t11 )^2

Note, however, the last quadratic term of the Split-Window equation is applied only over barren land. [Reference Required!]

BTs are derived from Landsat 8's TIRS channels 10 and 11. Prior to any processing, the raw digital numbers are filtered for clouds.

To produce an LST map, the algorithm requires at minimum:

• TIRS bands 10 and 11
• the acquisition's metadata file (MTL)
• a Finer Resolution Observation & Monitoring of Global Land Cover (FROM-GLC) product

Installation

Requirements

---

see GRASS Addons SVN repository, README file, Installation - Code Compilation

Steps

Making the script i.lansat8.swlst available from within any GRASS-GIS ver. 7.x session, may be done via the following steps:

1. launch a GRASS-GIS’ ver. 7.x session

2. navigate into the script’s source directory

3. execute make MODULE_TOPDIR=$GISBASE

Implementation notes

• Created on Wed Mar 18 10:00:53 2015
• First all-through execution: Tue May 12 21:50:42 EEST 2015
• Updated for Python 3 and added new functionality: Jun 2020

To Do

[High Priority]

• Fix retrieval of adjacent subranges (exclude 6, get it only if there is no match for subranges 1, 2, 3, 4, and 5)

• Evaluate BIG mapcalc expressions -- are they correct? I guess so ;-)

  • Still missing automatic tests!
  • Expression for Column Water Vapor
  • CWV output values range -- is it rational? It was not. There is a typo in paper [0]. The correct order of the coefficients is in papers [1, 2].
  • Expression for Land Surface Temperature
  • LST output values range -- is it rational? At the moment, not! Fixed. The main Split-Window equation was wrong.

• Why is the LST out of range when using a fixed land cover class? Cloudy pixels are, mainly, the reason. Better cloud masking is the solution.

• Why does the multi-step approach on deriving the CWV map differ from the single big mapcalc expression? Fixed

• Implement direct conversion of B10, B11 to brightness temperature values. Done

• Get the FROM-GLC map, Found

• implement mechanism to read land cover classes from it and use it to retrieve emissivities Done

• How to use the FVC? Don't. Just use the Look-up table (see [*] for details).

• Save average emissivity and delta emissivity maps for caching (re-use in subsequent trials, huge time saver!) Implemented

[Mid]

• Redo the example screenshots for the manual after corrections for CWV, LST equations.
• Use existing i.emissivity? Not exactly compatible -- read paper for details. Anyhow, options to input average and delta emissivity maps implemented.
• Raster Row I/O -- Maybe not an option: see discussion with Pietro Zambelli
• How to perform pixel value validity checks for in-between and end products? r.mapcalc can't do this. Best to implement a test checking the values on-the-fly while they are created. A C-function?

[Low]

• Test for too small region? Works for a region of 267 rows x 267 cols (71289 cells)
• Deduplicate code in split_window_lst class, in functions _build_average_emissivity_mapcalc() and _build_delta_emissivity_mapcalc()
• Implement a median window filter, as another option in addition to mean.
• Profiling
• Implement a complete cloud masking function using the BQA image. Support for user requested confidence or types of clouds (?). Eg: options= clouds,cirrus,high,low ?
• Multi-Threading? Note, r.mapcalc is already.

[*] Details: the authors followed the CBEM method. Based on the FROM-GLC map, they derived the following look-up table (LUT):

Emissivity Class|TIRS10|TIRS11
Cropland|0.971|0.968
Forest|0.995|0.996
Grasslands|0.97|0.971
Shrublands|0.969|0.97
Wetlands|0.992|0.998
Waterbodies|0.992|0.998
Tundra|0.98|0.984
Impervious|0.973|0.981
Barren Land|0.969|0.978
Snow and ice|0.992|0.998

References

• [0] Du, Chen; Ren, Huazhong; Qin, Qiming; Meng, Jinjie; Zhao, Shaohua. 2015. "A Practical Split-Window Algorithm for Estimating Land Surface Temperature from Landsat 8 Data." Remote Sens. 7, no. 1: 647-665. http://www.mdpi.com/2072-4292/7/1/647/htm\#sthash.ba1pt9hj.dpuf

• [1] Huazhong Ren, Chen Du, Qiming Qin, Rongyuan Liu, Jinjie Meng, and Jing Li. "Atmospheric Water Vapor Retrieval from Landsat 8 and Its Validation." 3045--3048. IEEE, 2014.

• [2] Ren, H., Du, C., Liu, R., Qin, Q., Yan, G., Li, Z. L., & Meng, J. (2015). Atmospheric water vapor retrieval from Landsat 8 thermal infrared images. Journal of Geophysical Research: Atmospheres, 120(5), 1723-1738.

• [3] FROM-GLC products, http://data.ess.tsinghua.edu.cn/

Ευχαριστώ

• Yann Chemin
• Pietro Zambelli
• StackExchange contributors
• Sajid Pareeth
• Georgios Alexandris, Anthoula Alexandri
• Special thanks to author Huazhong Ren for commenting on questions (personal communication)
• Stefan Blumentrath