Remote Sensing Based Soil Moisture Detection
Download as PPTX, PDF6 likes5,937 views
The document discusses the use of remote sensing techniques for soil moisture detection, highlighting the challenges of direct measurement and the potential of satellite data for wider area analysis. It introduces the PSMI (Permanent Soil Moisture Index) method utilizing GC/TIR space to model soil moisture, supported by Landsat imagery and field measurements. The findings indicate that PSMI correlates well with measured soil moisture and offers a promising tool for monitoring soil moisture distribution.
Remote Sensing Based Soil Moisture Detection
- 1. Remote Sensing Based Soil Moisture Detection Sanaz Shafian, Stephan J. Maas Department of Plant and Soil Science Texas Tech University Beyond Diagnostics: Insights and Recommendations from Remote Sensing
- 2. Texas Tech University Introduction Soil moisture influences Monitoring of plant water requirements Water resources and irrigation management Surface energy partitioning between the sensible and latent heat flux Beyond Diagnostics: Insights and Recommendations from Remote Sensing
- 3. Texas Tech University Introduction Challenges of directly soil moisture measurement Expensive Necessity of using surface meteorological observations Not readily available over large areas Produce point type measurements Restricted to specific locations Beyond Diagnostics: Insights and Recommendations from Remote Sensing
- 4. Texas Tech University Statement of problem Satellite remote sensing offers a means of measuring soil moisture Across a wide area Continuously Key variables in soil moisture estimation Vegetation cover Surface temperature Beyond Diagnostics: Insights and Recommendations from Remote Sensing
- 5. Texas Tech University Statement of problem Most current soil moisture estimation methods require Additional ancillary data Precise calibration of the surface temperature Expensive Time consuming Using NDVI in soil moisture estimation NDVI is a greenness index does not have physical interpretation Beyond Diagnostics: Insights and Recommendations from Remote Sensing
- 6. Texas Tech University Objectives To demonstrate how Landsat and other similar data may be used to estimate temporal and spatial patterns of soil moisture status To investigate the potentials of using a combination of multiple GCTIR spectral signatures to estimate soil moisture from space and to find the algorithm that will be best-suited for monitoring soil moisture To compare the results with soil moisture from direct measurements Beyond Diagnostics: Insights and Recommendations from Remote Sensing
- 7. Texas Tech University Literature review The Concept of using data from TIR band to monitor canopy water stress was originally proposed by Jackson(1977) Carlson (1989) studied the TsVI feature space properties and discovered that changes in soil moisture could be described within the TsVI ‘triangle’ Moran et al. (1994) introduced a concept termed the ‘vegetation index–temperature (VIT) trapezoid’ for the estimation of LE fluxes using the TsVI domain in areas of partial vegetation cover Beyond Diagnostics: Insights and Recommendations from Remote Sensing
- 8. Texas Tech University Literature review Gillies and Carlson (1995) introduced a method for the retrieval of spatially distributed maps of soil moisture availability (Mo), which they termed the ‘triangle’ method Sandholt et al. (2002) suggested a temperature vegetation dryness index (TVDI) for each pixel in trapezoid based on defining slope of dry edge Beyond Diagnostics: Insights and Recommendations from Remote Sensing
- 9. Texas Tech University GCTIR Space Observed properties of the GCTIR Space There is a relationship between ground cover (GC) and surface thermal emittance (TIR) of a given region Shape of the relationship is a truncated triangle or a trapezoid Beyond Diagnostics: Insights and Recommendations from Remote Sensing
- 10. Texas Tech University GCTIR Space Observed properties of the GCTIR Space GC increases along the y-axis Bare soil signal is gradually masked by vegetation contribution For a given GC, when TIR increases soil moisture will decrease Beyond Diagnostics: Insights and Recommendations from Remote Sensing
- 11. Texas Tech University GCTIR Space Observed properties of the GCTIR Space Minimum TIR value at the wet edge (maximum soil moisture) Maximum TIR value at the dry edge (Minimum soil moisture) The relative value of soil moisture at each pixel can be defined in terms of its position within the trapezoid /or triangle Beyond Diagnostics: Insights and Recommendations from Remote Sensing
- 12. Texas Tech University Description of the PSMI Method Modeling the trapezoid triangle Image processing Produce ground cover images by using PVI method • Red and NIR bands Beyond Diagnostics: Insights and Recommendations from Remote Sensing
- 13. Texas Tech University Description of the PSMI method Modeling the trapezoid triangle Image processing Produce GCTIR scatter plot for each image Normalizing TIR between 0 and 1 Produce Normalized GCTIR scatter plot Beyond Diagnostics: Insights and Recommendations from Remote Sensing
- 14. Texas Tech University Description of the PSMI method Decrease atmospheric effect Normalized TIR can be compared with normalized surface temperature Different scatter plots in different times can be compared GC and TIR are in the same range Beyond Diagnostics: Insights and Recommendations from Remote Sensing
- 15. Texas Tech University Description of the PSMI method Modeling the trapezoid triangle Consider the line that passes through the origin as the reference of soil moisture GC = 0 TIR = 0 Slope = - 45° Calculate perpendicular distance from each pixel from this line Beyond Diagnostics: Insights and Recommendations from Remote Sensing
- 16. Texas Tech University Description of the PSMI method Modeling the trapezoid triangle Normalizing the distance between 0 and 1 Considering the effect of GC Beyond Diagnostics: Insights and Recommendations from Remote Sensing
- 17. Texas Tech University Description of the PSMI method Calculate PSMI So, as PSMI goes from 0 to 1, you go from low to high soil moisture. Beyond Diagnostics: Insights and Recommendations from Remote Sensing
- 18. Texas Tech University Materials Study area Measuring soil moisture using TDR probe in 19 different fields Satellite Imagery 6 images from Landsat 7(ETM+)( 2012 and 2013 growing season) 4 images from Landsat 8(LCDM)( 2013 growing season) Beyond Diagnostics: Insights and Recommendations from Remote Sensing
- 19. Texas Tech University Results GC/TIR space is well defined in all cases Beyond Diagnostics: Insights and Recommendations from Remote Sensing
- 20. Texas Tech University Results Comparison between measured and estimated soil moisture Beyond Diagnostics: Insights and Recommendations from Remote Sensing
- 21. Texas Tech University Results Comparison between measured and estimated soil moisture Beyond Diagnostics: Insights and Recommendations from Remote Sensing
- 22. Texas Tech University Results Creating soil moisture map Spatial variation of soil moisture using PSMI Beyond Diagnostics: Insights and Recommendations from Remote Sensing
- 23. Texas Tech University Conclusions GCTIR space can be used instead VITs space to estimate soil moisture GCTIR space is well defined in all cases PSMI is always between 0 and 1 PSMI describes distribution of soil moisture in GCNormalized TIR space PSMI is closely related to measured soil moisture PSMI and measured soil moisture have similar spatial pattern Beyond Diagnostics: Insights and Recommendations from Remote Sensing
- 24. Texas Tech University Future work Using more data to test the robustness of the method over large areas Using different sets of satellite imagery (e.g. AVHRR) to derive PSMI Use of PSMI for driving, updating, and validating hydrological models Beyond Diagnostics: Insights and Recommendations from Remote Sensing
- 25. Texas Tech University Acknowledgment This project was funded by Texas Alliance Water Conservation (TAWC) We would like to thank John Deere Company for sharing soil moisture data Beyond Diagnostics: Insights and Recommendations from Remote Sensing