Transferability of Covariates to Predict Soil Organic Carbon in Cropland Soils
Precise knowledge about the soil organic carbon (SOC) content in cropland soils is one requirement to design and execute effective climate and food policies. In digital soil mapping (DSM), machine learning algorithms are used to predict soil properties from covariates derived from traditional soil m...
| Main Authors: | , , , , , |
|---|---|
| פורמט: | Article |
| שפה: | English |
| יצא לאור: |
MDPI AG
2023-02-01
|
| סדרה: | Remote Sensing |
| נושאים: | |
| גישה מקוונת: | https://www.mdpi.com/2072-4292/15/4/876 |
| _version_ | 1827755681044758528 |
|---|---|
| author | Tom Broeg Michael Blaschek Steffen Seitz Ruhollah Taghizadeh-Mehrjardi Simone Zepp Thomas Scholten |
| author_facet | Tom Broeg Michael Blaschek Steffen Seitz Ruhollah Taghizadeh-Mehrjardi Simone Zepp Thomas Scholten |
| author_sort | Tom Broeg |
| collection | DOAJ |
| description | Precise knowledge about the soil organic carbon (SOC) content in cropland soils is one requirement to design and execute effective climate and food policies. In digital soil mapping (DSM), machine learning algorithms are used to predict soil properties from covariates derived from traditional soil mapping, digital elevation models, land use, and Earth observation (EO). However, such DSM models are trained for a specific dataset and region and have so far only allowed limited general statements to be made that would enable the models to be transferred to different regions. In this study, we test the transferability of SOC models for cropland soils using five different covariate groups: multispectral soil reflectance composites (satellite), soil legacy data (soil), digital elevation model derivatives (terrain), climate parameters (climate), and combined models (combined). The transferability was analyzed using data from two federal states in southern Germany: Bavaria and Baden-Wuerttemberg. First, baseline models were trained for each state with combined models performing best in both cases (<i>R</i><sup>2</sup> = 0.68/0.48). Next, the models were transferred and tested with soil samples from the other state whose data were not used during model calibration. Only satellite and combined models were transferable, but accuracy declined in both cases. In the final step, models were trained with samples from both states (mixed-data models) and applied to each state separately. This process significantly improved the accuracies of satellite, terrain, and combined models, while it showed no effect on climate models and decreased the models based on soil covariates. The experiment underlines the importance of EO for the transfer and extrapolation of DSM models. |
| first_indexed | 2024-03-11T08:12:26Z |
| format | Article |
| id | doaj.art-00d7aeb227724ceabce5fcc3ac6f87ce |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-11T08:12:26Z |
| publishDate | 2023-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-00d7aeb227724ceabce5fcc3ac6f87ce2023-11-16T23:00:42ZengMDPI AGRemote Sensing2072-42922023-02-0115487610.3390/rs15040876Transferability of Covariates to Predict Soil Organic Carbon in Cropland SoilsTom Broeg0Michael Blaschek1Steffen Seitz2Ruhollah Taghizadeh-Mehrjardi3Simone Zepp4Thomas Scholten5Thünen Institute of Farm Economics, Bundesallee 63, 38116 Braunschweig, GermanyState Authority for Geology, Resources and Mining, Albertstraße 5, 79104 Freiburg, GermanyDepartment of Geosciences, Soil Science and Geomorphology, University of Tübingen, 72070 Tübingen, GermanyDepartment of Geosciences, Soil Science and Geomorphology, University of Tübingen, 72070 Tübingen, GermanyGerman Remote Sensing Data Center (DFD), German Aerospace Center (DLR), Muenchener Str. 20, 82234 Wessling, GermanyDepartment of Geosciences, Soil Science and Geomorphology, University of Tübingen, 72070 Tübingen, GermanyPrecise knowledge about the soil organic carbon (SOC) content in cropland soils is one requirement to design and execute effective climate and food policies. In digital soil mapping (DSM), machine learning algorithms are used to predict soil properties from covariates derived from traditional soil mapping, digital elevation models, land use, and Earth observation (EO). However, such DSM models are trained for a specific dataset and region and have so far only allowed limited general statements to be made that would enable the models to be transferred to different regions. In this study, we test the transferability of SOC models for cropland soils using five different covariate groups: multispectral soil reflectance composites (satellite), soil legacy data (soil), digital elevation model derivatives (terrain), climate parameters (climate), and combined models (combined). The transferability was analyzed using data from two federal states in southern Germany: Bavaria and Baden-Wuerttemberg. First, baseline models were trained for each state with combined models performing best in both cases (<i>R</i><sup>2</sup> = 0.68/0.48). Next, the models were transferred and tested with soil samples from the other state whose data were not used during model calibration. Only satellite and combined models were transferable, but accuracy declined in both cases. In the final step, models were trained with samples from both states (mixed-data models) and applied to each state separately. This process significantly improved the accuracies of satellite, terrain, and combined models, while it showed no effect on climate models and decreased the models based on soil covariates. The experiment underlines the importance of EO for the transfer and extrapolation of DSM models.https://www.mdpi.com/2072-4292/15/4/876machine learningdigital soil mappingsoil organic carbonmodel transferextrapolationsoil reflectance composite |
| spellingShingle | Tom Broeg Michael Blaschek Steffen Seitz Ruhollah Taghizadeh-Mehrjardi Simone Zepp Thomas Scholten Transferability of Covariates to Predict Soil Organic Carbon in Cropland Soils Remote Sensing machine learning digital soil mapping soil organic carbon model transfer extrapolation soil reflectance composite |
| title | Transferability of Covariates to Predict Soil Organic Carbon in Cropland Soils |
| title_full | Transferability of Covariates to Predict Soil Organic Carbon in Cropland Soils |
| title_fullStr | Transferability of Covariates to Predict Soil Organic Carbon in Cropland Soils |
| title_full_unstemmed | Transferability of Covariates to Predict Soil Organic Carbon in Cropland Soils |
| title_short | Transferability of Covariates to Predict Soil Organic Carbon in Cropland Soils |
| title_sort | transferability of covariates to predict soil organic carbon in cropland soils |
| topic | machine learning digital soil mapping soil organic carbon model transfer extrapolation soil reflectance composite |
| url | https://www.mdpi.com/2072-4292/15/4/876 |
| work_keys_str_mv | AT tombroeg transferabilityofcovariatestopredictsoilorganiccarbonincroplandsoils AT michaelblaschek transferabilityofcovariatestopredictsoilorganiccarbonincroplandsoils AT steffenseitz transferabilityofcovariatestopredictsoilorganiccarbonincroplandsoils AT ruhollahtaghizadehmehrjardi transferabilityofcovariatestopredictsoilorganiccarbonincroplandsoils AT simonezepp transferabilityofcovariatestopredictsoilorganiccarbonincroplandsoils AT thomasscholten transferabilityofcovariatestopredictsoilorganiccarbonincroplandsoils |