A deep learning method to predict soil organic carbon content at a regional scale using satellite-based phenology variables
Obtaining the spatial distribution information of soil organic carbon (SOC) is significant to quantify the carbon budget and guide land management for migrating carbon emissions. Digital soil mapping of SOC at a regional scale is challenging due to the complex SOC-environment relationships. Vegetati...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Elsevier
2021-10-01
|
Series: | International Journal of Applied Earth Observations and Geoinformation |
Subjects: | |
Online Access: | http://www.sciencedirect.com/science/article/pii/S0303243421001355 |
_version_ | 1811328872278392832 |
---|---|
author | Lin Yang Yanyan Cai Lei Zhang Mao Guo Anqi Li Chenghu Zhou |
author_facet | Lin Yang Yanyan Cai Lei Zhang Mao Guo Anqi Li Chenghu Zhou |
author_sort | Lin Yang |
collection | DOAJ |
description | Obtaining the spatial distribution information of soil organic carbon (SOC) is significant to quantify the carbon budget and guide land management for migrating carbon emissions. Digital soil mapping of SOC at a regional scale is challenging due to the complex SOC-environment relationships. Vegetation phenology that directly indicates a long time vegetation growth characteristics can be potential environmental covariates for SOC prediction. Deep learning has been developed for soil mapping recently due to its ability of constructing high-level features from the raw data. However, only dozens of predictors were used in most of those studies. It is not clear that how deep learning with long term land surface phenology product performs for SOC prediction at a regional scale. This paper explored the effectiveness of ten-years MODIS MCD12Q2 phenology variables for SOC prediction with a convolutional neural network (CNN) model in Anhui province, China. Random forest (RF) was applied to compare with CNN using three groups of environmental variables. The results showed that adding the land surface phenology variables into the pool of the natural environmental variables improved the prediction accuracy of CNN by 5.57% of RMSE and 31.29% of R2. Adding phenology variables obtained a higher accuracy improvement than adding Normalized Differences Vegetation Indices. The CNN obtained a higher prediction accuracy than RF regardless of using which group of variables. This study proved that land surface phenology metrics were effective predictors and CNN was a promising method for soil mapping at a regional scale. |
first_indexed | 2024-04-13T15:33:49Z |
format | Article |
id | doaj.art-f69d839459a14968a31dc29f49df4052 |
institution | Directory Open Access Journal |
issn | 1569-8432 |
language | English |
last_indexed | 2024-04-13T15:33:49Z |
publishDate | 2021-10-01 |
publisher | Elsevier |
record_format | Article |
series | International Journal of Applied Earth Observations and Geoinformation |
spelling | doaj.art-f69d839459a14968a31dc29f49df40522022-12-22T02:41:20ZengElsevierInternational Journal of Applied Earth Observations and Geoinformation1569-84322021-10-01102102428A deep learning method to predict soil organic carbon content at a regional scale using satellite-based phenology variablesLin Yang0Yanyan Cai1Lei Zhang2Mao Guo3Anqi Li4Chenghu Zhou5School of Geography and Ocean Science, Nanjing University, Nanjing, 210023 ChinaSchool of Geography and Ocean Science, Nanjing University, Nanjing, 210023 ChinaSchool of Geography and Ocean Science, Nanjing University, Nanjing, 210023 ChinaSchool of Geography and Ocean Science, Nanjing University, Nanjing, 210023 ChinaSchool of Geography and Ocean Science, Nanjing University, Nanjing, 210023 ChinaSchool of Geography and Ocean Science, Nanjing University, Nanjing, 210023 China; State Key Laboratory of Resources and Environmental Information System, Institute of Geographical Sciences and Natural Resources Research, CAS, Beijing 100101, China; Corresponding author.Obtaining the spatial distribution information of soil organic carbon (SOC) is significant to quantify the carbon budget and guide land management for migrating carbon emissions. Digital soil mapping of SOC at a regional scale is challenging due to the complex SOC-environment relationships. Vegetation phenology that directly indicates a long time vegetation growth characteristics can be potential environmental covariates for SOC prediction. Deep learning has been developed for soil mapping recently due to its ability of constructing high-level features from the raw data. However, only dozens of predictors were used in most of those studies. It is not clear that how deep learning with long term land surface phenology product performs for SOC prediction at a regional scale. This paper explored the effectiveness of ten-years MODIS MCD12Q2 phenology variables for SOC prediction with a convolutional neural network (CNN) model in Anhui province, China. Random forest (RF) was applied to compare with CNN using three groups of environmental variables. The results showed that adding the land surface phenology variables into the pool of the natural environmental variables improved the prediction accuracy of CNN by 5.57% of RMSE and 31.29% of R2. Adding phenology variables obtained a higher accuracy improvement than adding Normalized Differences Vegetation Indices. The CNN obtained a higher prediction accuracy than RF regardless of using which group of variables. This study proved that land surface phenology metrics were effective predictors and CNN was a promising method for soil mapping at a regional scale.http://www.sciencedirect.com/science/article/pii/S0303243421001355Deep learningSoil organic carbonConvolutional neural network (CNN)Land surface phenology |
spellingShingle | Lin Yang Yanyan Cai Lei Zhang Mao Guo Anqi Li Chenghu Zhou A deep learning method to predict soil organic carbon content at a regional scale using satellite-based phenology variables International Journal of Applied Earth Observations and Geoinformation Deep learning Soil organic carbon Convolutional neural network (CNN) Land surface phenology |
title | A deep learning method to predict soil organic carbon content at a regional scale using satellite-based phenology variables |
title_full | A deep learning method to predict soil organic carbon content at a regional scale using satellite-based phenology variables |
title_fullStr | A deep learning method to predict soil organic carbon content at a regional scale using satellite-based phenology variables |
title_full_unstemmed | A deep learning method to predict soil organic carbon content at a regional scale using satellite-based phenology variables |
title_short | A deep learning method to predict soil organic carbon content at a regional scale using satellite-based phenology variables |
title_sort | deep learning method to predict soil organic carbon content at a regional scale using satellite based phenology variables |
topic | Deep learning Soil organic carbon Convolutional neural network (CNN) Land surface phenology |
url | http://www.sciencedirect.com/science/article/pii/S0303243421001355 |
work_keys_str_mv | AT linyang adeeplearningmethodtopredictsoilorganiccarboncontentataregionalscaleusingsatellitebasedphenologyvariables AT yanyancai adeeplearningmethodtopredictsoilorganiccarboncontentataregionalscaleusingsatellitebasedphenologyvariables AT leizhang adeeplearningmethodtopredictsoilorganiccarboncontentataregionalscaleusingsatellitebasedphenologyvariables AT maoguo adeeplearningmethodtopredictsoilorganiccarboncontentataregionalscaleusingsatellitebasedphenologyvariables AT anqili adeeplearningmethodtopredictsoilorganiccarboncontentataregionalscaleusingsatellitebasedphenologyvariables AT chenghuzhou adeeplearningmethodtopredictsoilorganiccarboncontentataregionalscaleusingsatellitebasedphenologyvariables AT linyang deeplearningmethodtopredictsoilorganiccarboncontentataregionalscaleusingsatellitebasedphenologyvariables AT yanyancai deeplearningmethodtopredictsoilorganiccarboncontentataregionalscaleusingsatellitebasedphenologyvariables AT leizhang deeplearningmethodtopredictsoilorganiccarboncontentataregionalscaleusingsatellitebasedphenologyvariables AT maoguo deeplearningmethodtopredictsoilorganiccarboncontentataregionalscaleusingsatellitebasedphenologyvariables AT anqili deeplearningmethodtopredictsoilorganiccarboncontentataregionalscaleusingsatellitebasedphenologyvariables AT chenghuzhou deeplearningmethodtopredictsoilorganiccarboncontentataregionalscaleusingsatellitebasedphenologyvariables |