Watershed scale soil moisture estimation model using machine learning and remote sensing in a data-scarce context
Soil moisture content can be used to predict drought impact on agricultural yield better than precipitation. Remote sensing is viable source of soil moisture data in instrument-scarce areas. However, space-based soil moisture estimates lack suitability for daily and high-resolution agricultural, hyd...
| Main Authors: | , , , , |
|---|---|
| Format: | Journal Article |
| Language: | Inglés |
| Published: |
Scientia Agropecuaria
2024
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10568/131095 |
| _version_ | 1855534177397506048 |
|---|---|
| author | Bueno, M. Baca García, C. Montoya, N. Rau, P. Loayza, H. |
| author_browse | Baca García, C. Bueno, M. Loayza, H. Montoya, N. Rau, P. |
| author_facet | Bueno, M. Baca García, C. Montoya, N. Rau, P. Loayza, H. |
| author_sort | Bueno, M. |
| collection | Repository of Agricultural Research Outputs (CGSpace) |
| description | Soil moisture content can be used to predict drought impact on agricultural yield better than precipitation. Remote sensing is viable source of soil moisture data in instrument-scarce areas. However, space-based soil moisture estimates lack suitability for daily and high-resolution agricultural, hydrological, and environmental applications. This study aimed to assess the potential of the random forest machine learning technique to enhance the spatial resolution of remote soil moisture products from the SMAP satellite. Models were built using random forest for spatial downscaling of SMAP-L3-E, then visually and statistically evaluated for disaggregation quality. The impact of topography, soil properties, and precipitation on the downscaled soil moisture was examined. The relationship between downscaled soil moisture and in-situ soil moisture was analyzed. The results indicate that the proposed method demonstrated spatial and hydrological coherence, along with a satisfactory downscaling quality. Statistical validation indicated suitable generalization error for scientific and practical use (RMSE < 0.05 cm3 cm-3). Random forest effectively achieved spatial downscaling of SMAP-L3-E in the study area. Principal component and spatial analysis revealed dependence of downscaled soil moisture on elevation, soil organic carbon content, clay content, and saturated hydraulic conductivity, mainly under near-saturation conditions. Regarding validation against in-situ data, downscaled soil moisture explained in-situ soil moisture well under low soil water content ( = 0.624). Downscaling performance deteriorates for water contents between 0.40 to 0.50 cm3 cm-3, suggesting inadequacy under near saturation conditions at a daily temporal frequency. However, coarser temporal aggregations (7 to 10 days) yielded an average 0.98 correlation coefficient, regardless of saturation conditions. These results could potentially be applied in irrigation planning, soil physics studies and hydrology monitoring, to forecasting the occurrence of droughts, leaching of contaminants, surface runoff modeling, carbon cycle studies, soil's capacity to store and provide nutrients. |
| format | Journal Article |
| id | CGSpace131095 |
| institution | CGIAR Consortium |
| language | Inglés |
| publishDate | 2024 |
| publishDateRange | 2024 |
| publishDateSort | 2024 |
| publisher | Scientia Agropecuaria |
| publisherStr | Scientia Agropecuaria |
| record_format | dspace |
| spelling | CGSpace1310952025-03-13T16:10:08Z Watershed scale soil moisture estimation model using machine learning and remote sensing in a data-scarce context Bueno, M. Baca García, C. Montoya, N. Rau, P. Loayza, H. soil water content remote sensing machine learning drought Soil moisture content can be used to predict drought impact on agricultural yield better than precipitation. Remote sensing is viable source of soil moisture data in instrument-scarce areas. However, space-based soil moisture estimates lack suitability for daily and high-resolution agricultural, hydrological, and environmental applications. This study aimed to assess the potential of the random forest machine learning technique to enhance the spatial resolution of remote soil moisture products from the SMAP satellite. Models were built using random forest for spatial downscaling of SMAP-L3-E, then visually and statistically evaluated for disaggregation quality. The impact of topography, soil properties, and precipitation on the downscaled soil moisture was examined. The relationship between downscaled soil moisture and in-situ soil moisture was analyzed. The results indicate that the proposed method demonstrated spatial and hydrological coherence, along with a satisfactory downscaling quality. Statistical validation indicated suitable generalization error for scientific and practical use (RMSE < 0.05 cm3 cm-3). Random forest effectively achieved spatial downscaling of SMAP-L3-E in the study area. Principal component and spatial analysis revealed dependence of downscaled soil moisture on elevation, soil organic carbon content, clay content, and saturated hydraulic conductivity, mainly under near-saturation conditions. Regarding validation against in-situ data, downscaled soil moisture explained in-situ soil moisture well under low soil water content ( = 0.624). Downscaling performance deteriorates for water contents between 0.40 to 0.50 cm3 cm-3, suggesting inadequacy under near saturation conditions at a daily temporal frequency. However, coarser temporal aggregations (7 to 10 days) yielded an average 0.98 correlation coefficient, regardless of saturation conditions. These results could potentially be applied in irrigation planning, soil physics studies and hydrology monitoring, to forecasting the occurrence of droughts, leaching of contaminants, surface runoff modeling, carbon cycle studies, soil's capacity to store and provide nutrients. 2024-03-11 2023-07-12T02:36:47Z 2023-07-12T02:36:47Z Journal Article https://hdl.handle.net/10568/131095 en Open Access Scientia Agropecuaria Bueno, M.; Baca García, C.; Montoya, N.; Rau, P.; Loayza, H. 2024. Watershed scale soil moisture estimation model using machine learning and remote sensing in a data-scarce context. Scientia Agropecuaria. ISSN 2306-6741. 15(1), 103-120. |
| spellingShingle | soil water content remote sensing machine learning drought Bueno, M. Baca García, C. Montoya, N. Rau, P. Loayza, H. Watershed scale soil moisture estimation model using machine learning and remote sensing in a data-scarce context |
| title | Watershed scale soil moisture estimation model using machine learning and remote sensing in a data-scarce context |
| title_full | Watershed scale soil moisture estimation model using machine learning and remote sensing in a data-scarce context |
| title_fullStr | Watershed scale soil moisture estimation model using machine learning and remote sensing in a data-scarce context |
| title_full_unstemmed | Watershed scale soil moisture estimation model using machine learning and remote sensing in a data-scarce context |
| title_short | Watershed scale soil moisture estimation model using machine learning and remote sensing in a data-scarce context |
| title_sort | watershed scale soil moisture estimation model using machine learning and remote sensing in a data scarce context |
| topic | soil water content remote sensing machine learning drought |
| url | https://hdl.handle.net/10568/131095 |
| work_keys_str_mv | AT buenom watershedscalesoilmoistureestimationmodelusingmachinelearningandremotesensinginadatascarcecontext AT bacagarciac watershedscalesoilmoistureestimationmodelusingmachinelearningandremotesensinginadatascarcecontext AT montoyan watershedscalesoilmoistureestimationmodelusingmachinelearningandremotesensinginadatascarcecontext AT raup watershedscalesoilmoistureestimationmodelusingmachinelearningandremotesensinginadatascarcecontext AT loayzah watershedscalesoilmoistureestimationmodelusingmachinelearningandremotesensinginadatascarcecontext |