Astronomical School’s Report, 2018, Volume 14, Issue 2, Pages 70–77

https://doi.org/10.18372/2411-6602.14.10
Download PDF
UDC 528.855+835

Estimation of the thermodynamic temperature field of the land surface using satellite data based on land cover classification

Zyelyk Ya.I., Pidgorodetska L.V., Chornyy S.V.

Space Research Institute NASU & SSAU, Hlushkova Avenue 40, building 4/1, 03680 Kyiv, Ukraine

Abstract

The method for estimation of the thermodynamic temperature field of the land surface using satellite data of the long-wave infrared band based on the land cover classification in the optical wavelength range using machine learning methods has been justified. The method is implemented in the Semi-Automatic Classification Plugin (SCP) for QGIS. The vector layer of the marsh mineral and peat soils for the lowland areas in the Kyiv region on the basis of Ukrainian soil maps and the SRTM Digital Elevation Data has been created. The vector layer of settlements within these soils has been constructed with the attributive information on places and dates of fires on peat lands. For the implementation and preliminary verification of the method, Landsat 8 OLI and TIRS satellite imagery were used, containing areas of fire-hazardous peat lands on the dates of the pronounced fire hazard. The rasters of the land cover classification based on the Maximum Likelihood Method and rasters of the thermal emissivity have been created. The thermal emissivity raster is characteristic for the obtained land cover classes. The rasters of the land surface thermodynamic temperature have been constructed by conversion of the effective brightness temperature at the satellite sensor using the thermal emissivity raster. In addition to long revisit period of the land surface area (16 days), the multispectral Landsat 8 (OLI and TIRS) data have the significant limitations that allow on their basis assess only the smoldering phase on the peat lands. The upper limit of the effective brightness temperature at the satellite sensor, based on the data of the thermal infrared band B10 (11 μm), due to “saturation” of its ADC, is T_brith_max = 94.88°C = 368 K. Data of shortwave infrared bands B6 (1.6 μm), B7 (2.2 μm) can be used to detect open flame hearth based only on nighttime Landsat 8 (OLI) imagery that are rare. It has been established that the contours of the land areas, obtained from the conditions for exceeding the experimentally set threshold values of the thermodynamic temperature from the constructed temperature raster, are consistent with ground-based information on smoldering and fires on peat lands. Further research will focus on the improvement and implementation of the developed method for the enhancement of the spatial resolution of thermal field images, smoke identification procedures based on classification methods. The regional (for the Kyiv region) index of peat lands fires will be assessed, which is proportional to the integrated during the period of fires value of the difference between the amount of atmospheric precipitations and the amount of evapotranspiration from the land, water structures and plants.

Keywords: thermodynamic temperature of the land surface; brightness (radiation) temperature; temperature raster; thermal emissivity; land cover classification, maximum likelihood method; fire hazardous peat lands; smoldering phase; flaming phase; Landsat 8 (OLI, TIRS)

References

  1. Semi-Automatic Classification Plugin Documentation [Elektronny resurs], Rezhim dostupu: https://semiautomaticclassificationmanual-v5.readthedocs.io/en/latest/# (01.11.2018)
  2. QGIS. A Free and Open Source Geographic Information System [Elektronny resurs], Rezhim dostupu: https://www.qgis.org/en/site/ (01.11.2018).
  3. Sobrino J., Jiménez-Muñoz J.C., Paolini L. (2004). Land surface temperature retrieval from LANDSAT TM 5. Remote Sensing of Environment, 90, 434–440. https://doi.org/10.1016/j.rse.2004.02.003
  4. Moran M., Jackson R., Slater P., Teillet P. (1992). Evaluation of simplified procedures for retrieval of land surface reflectance factors from satellite sensor output. Remote Sensing of Environment, 41, 169–184. https://doi.org/10.1016/0034-4257(92)90076-v
  5. Richards J.A., Jia X. Remote Sensing Digital Image Analysis: An Introduction. Springer, 2006. https://doi.org/10.1007/978-3-662-03978-6
  6. Weng Q., Lu D., Schubring J. (2004). Estimation of land surface temperature-vegetation abundance relationship for urban heat island studies. Remote Sensing of Environment, 89, 467–483. https://doi.org/10.1016/j.rse.2003.11.005
  7. Mallick J., Singh C.K., Shashtri S., Rahman A., Mukherjee S. (2012). Land surface emissivity retrieval based on moisture index from LANDSAT TM satellite data over heterogeneous surfaces of Delhi city. International Journal of Applied Earth Observation and Geoinformation, 19, 348–358. https://doi.org/10.1016/j.jag.2012.06.002
  8. Using the USGS Landsat Level-1 Data Product [Elektronny resurs], Rezhim dostupu: https://landsat.usgs.gov/using-usgs-landsat-8-product (01.11.2018).
  9. Elvidge C.D., Zhizhin M., Feng-Chi Hsu, Baugh K., Khomarudin M.R., Vetrita Y., Sofan P., Suwarsono, Dadang Hilman (2015). Long-wave infrared identification of smoldering peat fires in Indonesia with nighttime Landsat data. Environ. Res. Lett., 10, No. 6. https://doi.org/10.1088/1748-9326/10/6/065002

Download PDF