Spatial downscaling of SMAP radiometer soil moisture using radar data: Application of machine learning to the SMAPEx and SMAPVEX campaigns
This study developed a random forest approach for downscaling the coarse-resolution (36 km) soil moisture measured by The National Aeronautics and Space Administration (NASA) Soil Moisture Active Passive (SMAP) mission to 1 km spatial resolution, utilizing airborne remotely sensed data (radar backsc...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-06-01
|
| Series: | Science of Remote Sensing |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666017224000063 |
| _version_ | 1797292917368291328 |
|---|---|
| author | Elaheh Ghafari Jeffrey P. Walker Liujun Zhu Andreas Colliander Alireza Faridhosseini |
| author_facet | Elaheh Ghafari Jeffrey P. Walker Liujun Zhu Andreas Colliander Alireza Faridhosseini |
| author_sort | Elaheh Ghafari |
| collection | DOAJ |
| description | This study developed a random forest approach for downscaling the coarse-resolution (36 km) soil moisture measured by The National Aeronautics and Space Administration (NASA) Soil Moisture Active Passive (SMAP) mission to 1 km spatial resolution, utilizing airborne remotely sensed data (radar backscatter and radiometer retrieved soil moisture), vegetation characteristics (normalized difference vegetation index), soil properties, topography, and ground soil moisture measurements from before the launch of SMAP for training a random forest model. The 36 km SMAP soil moisture product was then downscaled by the trained model to 1 km resolution using the information from SMAP. The downscaled soil moisture was evaluated using airborne retrieved soil moisture observations and ground soil moisture measurements. Considering the airborne retrieved soil moisture as a reference, the results demonstrated that the proposed random forest model could downscale the SMAP radiometer product to 1 km resolution with a correlation coefficient of 0.97, unbiased Root Mean Square Error of 0.048 m3 m−3 and bias of 0.016 m3 m−3. Accordingly, the downscaled soil moisture captured the spatial and temporal heterogeneity and demonstrated the potential of the proposed machine learning model for soil moisture downscaling. |
| first_indexed | 2024-03-07T20:04:25Z |
| format | Article |
| id | doaj.art-bc1c5f51b6734e02bfda8d0e9e2672a3 |
| institution | Directory Open Access Journal |
| issn | 2666-0172 |
| language | English |
| last_indexed | 2024-03-07T20:04:25Z |
| publishDate | 2024-06-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Science of Remote Sensing |
| spelling | doaj.art-bc1c5f51b6734e02bfda8d0e9e2672a32024-02-28T05:14:09ZengElsevierScience of Remote Sensing2666-01722024-06-019100122Spatial downscaling of SMAP radiometer soil moisture using radar data: Application of machine learning to the SMAPEx and SMAPVEX campaignsElaheh Ghafari0Jeffrey P. Walker1Liujun Zhu2Andreas Colliander3Alireza Faridhosseini4Department of Water Engineering, Ferdowsi University of Mashhad, Mashhad, Iran; Corresponding author.Department of Civil Engineering, Monash University, Melbourne, AustraliaDepartment of Civil Engineering, Monash University, Melbourne, Australia; Yangtze Institute for Conservation and Development, Hohai University, Nanjing, 210024, ChinaJet Propulsion Laboratory, NASA, California Institute of Technology, Pasadena, CA 91109, USADepartment of Water Engineering, Ferdowsi University of Mashhad, Mashhad, IranThis study developed a random forest approach for downscaling the coarse-resolution (36 km) soil moisture measured by The National Aeronautics and Space Administration (NASA) Soil Moisture Active Passive (SMAP) mission to 1 km spatial resolution, utilizing airborne remotely sensed data (radar backscatter and radiometer retrieved soil moisture), vegetation characteristics (normalized difference vegetation index), soil properties, topography, and ground soil moisture measurements from before the launch of SMAP for training a random forest model. The 36 km SMAP soil moisture product was then downscaled by the trained model to 1 km resolution using the information from SMAP. The downscaled soil moisture was evaluated using airborne retrieved soil moisture observations and ground soil moisture measurements. Considering the airborne retrieved soil moisture as a reference, the results demonstrated that the proposed random forest model could downscale the SMAP radiometer product to 1 km resolution with a correlation coefficient of 0.97, unbiased Root Mean Square Error of 0.048 m3 m−3 and bias of 0.016 m3 m−3. Accordingly, the downscaled soil moisture captured the spatial and temporal heterogeneity and demonstrated the potential of the proposed machine learning model for soil moisture downscaling.http://www.sciencedirect.com/science/article/pii/S2666017224000063Machine learningDownscalingSoil moistureSMAPRandom forest modelSMAPEx |
| spellingShingle | Elaheh Ghafari Jeffrey P. Walker Liujun Zhu Andreas Colliander Alireza Faridhosseini Spatial downscaling of SMAP radiometer soil moisture using radar data: Application of machine learning to the SMAPEx and SMAPVEX campaigns Science of Remote Sensing Machine learning Downscaling Soil moisture SMAP Random forest model SMAPEx |
| title | Spatial downscaling of SMAP radiometer soil moisture using radar data: Application of machine learning to the SMAPEx and SMAPVEX campaigns |
| title_full | Spatial downscaling of SMAP radiometer soil moisture using radar data: Application of machine learning to the SMAPEx and SMAPVEX campaigns |
| title_fullStr | Spatial downscaling of SMAP radiometer soil moisture using radar data: Application of machine learning to the SMAPEx and SMAPVEX campaigns |
| title_full_unstemmed | Spatial downscaling of SMAP radiometer soil moisture using radar data: Application of machine learning to the SMAPEx and SMAPVEX campaigns |
| title_short | Spatial downscaling of SMAP radiometer soil moisture using radar data: Application of machine learning to the SMAPEx and SMAPVEX campaigns |
| title_sort | spatial downscaling of smap radiometer soil moisture using radar data application of machine learning to the smapex and smapvex campaigns |
| topic | Machine learning Downscaling Soil moisture SMAP Random forest model SMAPEx |
| url | http://www.sciencedirect.com/science/article/pii/S2666017224000063 |
| work_keys_str_mv | AT elahehghafari spatialdownscalingofsmapradiometersoilmoistureusingradardataapplicationofmachinelearningtothesmapexandsmapvexcampaigns AT jeffreypwalker spatialdownscalingofsmapradiometersoilmoistureusingradardataapplicationofmachinelearningtothesmapexandsmapvexcampaigns AT liujunzhu spatialdownscalingofsmapradiometersoilmoistureusingradardataapplicationofmachinelearningtothesmapexandsmapvexcampaigns AT andreascolliander spatialdownscalingofsmapradiometersoilmoistureusingradardataapplicationofmachinelearningtothesmapexandsmapvexcampaigns AT alirezafaridhosseini spatialdownscalingofsmapradiometersoilmoistureusingradardataapplicationofmachinelearningtothesmapexandsmapvexcampaigns |