A Spatial Downscaling Method for Remote Sensing Soil Moisture Based on Random Forest Considering Soil Moisture Memory and Mass Conservation
Remote sensing soil moisture (SM) has been widely used in various earth science studies and applications, but their low resolution limits their usage and downscaling of them is needed. In this study, we proposed a spatial downscaling method for SM based on random forest considering soil moisture mem...
Main Authors: | , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-08-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-4292/14/16/3858 |
_version_ | 1797442363798323200 |
---|---|
author | Taoning Mao Wei Shangguan Qingliang Li Lu Li Ye Zhang Feini Huang Jianduo Li Wei Liu Ruqing Zhang |
author_facet | Taoning Mao Wei Shangguan Qingliang Li Lu Li Ye Zhang Feini Huang Jianduo Li Wei Liu Ruqing Zhang |
author_sort | Taoning Mao |
collection | DOAJ |
description | Remote sensing soil moisture (SM) has been widely used in various earth science studies and applications, but their low resolution limits their usage and downscaling of them is needed. In this study, we proposed a spatial downscaling method for SM based on random forest considering soil moisture memory and mass conservation to improve downscaling performance. The lagged SM was added as a predictor to represent soil moisture memory, in addition to the regular predictors in previous downscaling studies. The Soil Moisture Active Passive (SMAP) SM data of the Pearl River Basin were used to test our downscaling method. The results show that the downscaling model obtained good performance on the test set (<i>R</i><sup>2</sup> = 0.848, ubRMSE = 0.034 m<sup>3</sup>/m<sup>3</sup> and Bias = 0.008 m<sup>3</sup>/m<sup>3</sup>). The spatial and temporal performance of the RF downscaling model can be improved by adding lagged SM variables. Downscaled data obtained can retain the information of the original SMAP SM data well and show more spatial details, and mass conservation correction is considered to be useful to eliminate systematic bias of the downscaling model. Downscaled SM achieved acceptable performance in in situ validation, though it was inevitably limited by the performance of the original SMAP data. The proposed downscaling method can serve as a powerful tool for the development of high-resolution SM information. |
first_indexed | 2024-03-09T12:40:49Z |
format | Article |
id | doaj.art-5c3ad2e99ba44842a8226438f1a1cfd7 |
institution | Directory Open Access Journal |
issn | 2072-4292 |
language | English |
last_indexed | 2024-03-09T12:40:49Z |
publishDate | 2022-08-01 |
publisher | MDPI AG |
record_format | Article |
series | Remote Sensing |
spelling | doaj.art-5c3ad2e99ba44842a8226438f1a1cfd72023-11-30T22:18:37ZengMDPI AGRemote Sensing2072-42922022-08-011416385810.3390/rs14163858A Spatial Downscaling Method for Remote Sensing Soil Moisture Based on Random Forest Considering Soil Moisture Memory and Mass ConservationTaoning Mao0Wei Shangguan1Qingliang Li2Lu Li3Ye Zhang4Feini Huang5Jianduo Li6Wei Liu7Ruqing Zhang8Southern Marine Science and Engineering Guangdong Laboratory (Zhuhai), Guangdong Province Key Laboratory for Climate Change and Natural Disaster Studies, School of Atmospheric Sciences, Sun Yat-sen University, Guangzhou 510275, ChinaSouthern Marine Science and Engineering Guangdong Laboratory (Zhuhai), Guangdong Province Key Laboratory for Climate Change and Natural Disaster Studies, School of Atmospheric Sciences, Sun Yat-sen University, Guangzhou 510275, ChinaCollege of Computer Science and Technology, Changchun Normal University, Changchun 130032, ChinaSouthern Marine Science and Engineering Guangdong Laboratory (Zhuhai), Guangdong Province Key Laboratory for Climate Change and Natural Disaster Studies, School of Atmospheric Sciences, Sun Yat-sen University, Guangzhou 510275, ChinaSouthern Marine Science and Engineering Guangdong Laboratory (Zhuhai), Guangdong Province Key Laboratory for Climate Change and Natural Disaster Studies, School of Atmospheric Sciences, Sun Yat-sen University, Guangzhou 510275, ChinaSouthern Marine Science and Engineering Guangdong Laboratory (Zhuhai), Guangdong Province Key Laboratory for Climate Change and Natural Disaster Studies, School of Atmospheric Sciences, Sun Yat-sen University, Guangzhou 510275, ChinaState Key Laboratory of Severe Weather, Chinese Academy of Meteorological Sciences, Beijing 100081, ChinaGuangdong Climate Center, Guangzhou 510275, ChinaSouthern Marine Science and Engineering Guangdong Laboratory (Zhuhai), Guangdong Province Key Laboratory for Climate Change and Natural Disaster Studies, School of Atmospheric Sciences, Sun Yat-sen University, Guangzhou 510275, ChinaRemote sensing soil moisture (SM) has been widely used in various earth science studies and applications, but their low resolution limits their usage and downscaling of them is needed. In this study, we proposed a spatial downscaling method for SM based on random forest considering soil moisture memory and mass conservation to improve downscaling performance. The lagged SM was added as a predictor to represent soil moisture memory, in addition to the regular predictors in previous downscaling studies. The Soil Moisture Active Passive (SMAP) SM data of the Pearl River Basin were used to test our downscaling method. The results show that the downscaling model obtained good performance on the test set (<i>R</i><sup>2</sup> = 0.848, ubRMSE = 0.034 m<sup>3</sup>/m<sup>3</sup> and Bias = 0.008 m<sup>3</sup>/m<sup>3</sup>). The spatial and temporal performance of the RF downscaling model can be improved by adding lagged SM variables. Downscaled data obtained can retain the information of the original SMAP SM data well and show more spatial details, and mass conservation correction is considered to be useful to eliminate systematic bias of the downscaling model. Downscaled SM achieved acceptable performance in in situ validation, though it was inevitably limited by the performance of the original SMAP data. The proposed downscaling method can serve as a powerful tool for the development of high-resolution SM information.https://www.mdpi.com/2072-4292/14/16/3858soil moisturemachine learningdownscalingSMAPremote sensing |
spellingShingle | Taoning Mao Wei Shangguan Qingliang Li Lu Li Ye Zhang Feini Huang Jianduo Li Wei Liu Ruqing Zhang A Spatial Downscaling Method for Remote Sensing Soil Moisture Based on Random Forest Considering Soil Moisture Memory and Mass Conservation Remote Sensing soil moisture machine learning downscaling SMAP remote sensing |
title | A Spatial Downscaling Method for Remote Sensing Soil Moisture Based on Random Forest Considering Soil Moisture Memory and Mass Conservation |
title_full | A Spatial Downscaling Method for Remote Sensing Soil Moisture Based on Random Forest Considering Soil Moisture Memory and Mass Conservation |
title_fullStr | A Spatial Downscaling Method for Remote Sensing Soil Moisture Based on Random Forest Considering Soil Moisture Memory and Mass Conservation |
title_full_unstemmed | A Spatial Downscaling Method for Remote Sensing Soil Moisture Based on Random Forest Considering Soil Moisture Memory and Mass Conservation |
title_short | A Spatial Downscaling Method for Remote Sensing Soil Moisture Based on Random Forest Considering Soil Moisture Memory and Mass Conservation |
title_sort | spatial downscaling method for remote sensing soil moisture based on random forest considering soil moisture memory and mass conservation |
topic | soil moisture machine learning downscaling SMAP remote sensing |
url | https://www.mdpi.com/2072-4292/14/16/3858 |
work_keys_str_mv | AT taoningmao aspatialdownscalingmethodforremotesensingsoilmoisturebasedonrandomforestconsideringsoilmoisturememoryandmassconservation AT weishangguan aspatialdownscalingmethodforremotesensingsoilmoisturebasedonrandomforestconsideringsoilmoisturememoryandmassconservation AT qingliangli aspatialdownscalingmethodforremotesensingsoilmoisturebasedonrandomforestconsideringsoilmoisturememoryandmassconservation AT luli aspatialdownscalingmethodforremotesensingsoilmoisturebasedonrandomforestconsideringsoilmoisturememoryandmassconservation AT yezhang aspatialdownscalingmethodforremotesensingsoilmoisturebasedonrandomforestconsideringsoilmoisturememoryandmassconservation AT feinihuang aspatialdownscalingmethodforremotesensingsoilmoisturebasedonrandomforestconsideringsoilmoisturememoryandmassconservation AT jianduoli aspatialdownscalingmethodforremotesensingsoilmoisturebasedonrandomforestconsideringsoilmoisturememoryandmassconservation AT weiliu aspatialdownscalingmethodforremotesensingsoilmoisturebasedonrandomforestconsideringsoilmoisturememoryandmassconservation AT ruqingzhang aspatialdownscalingmethodforremotesensingsoilmoisturebasedonrandomforestconsideringsoilmoisturememoryandmassconservation AT taoningmao spatialdownscalingmethodforremotesensingsoilmoisturebasedonrandomforestconsideringsoilmoisturememoryandmassconservation AT weishangguan spatialdownscalingmethodforremotesensingsoilmoisturebasedonrandomforestconsideringsoilmoisturememoryandmassconservation AT qingliangli spatialdownscalingmethodforremotesensingsoilmoisturebasedonrandomforestconsideringsoilmoisturememoryandmassconservation AT luli spatialdownscalingmethodforremotesensingsoilmoisturebasedonrandomforestconsideringsoilmoisturememoryandmassconservation AT yezhang spatialdownscalingmethodforremotesensingsoilmoisturebasedonrandomforestconsideringsoilmoisturememoryandmassconservation AT feinihuang spatialdownscalingmethodforremotesensingsoilmoisturebasedonrandomforestconsideringsoilmoisturememoryandmassconservation AT jianduoli spatialdownscalingmethodforremotesensingsoilmoisturebasedonrandomforestconsideringsoilmoisturememoryandmassconservation AT weiliu spatialdownscalingmethodforremotesensingsoilmoisturebasedonrandomforestconsideringsoilmoisturememoryandmassconservation AT ruqingzhang spatialdownscalingmethodforremotesensingsoilmoisturebasedonrandomforestconsideringsoilmoisturememoryandmassconservation |