View in EDS HTML Full Text PDF Full Text

Optimization of Sampling Density in Regional-Scale Soil Organic Matter Mapping and Its Impact on Prediction Accuracy.

Saved in:
Bibliographic Details
Title: Optimization of Sampling Density in Regional-Scale Soil Organic Matter Mapping and Its Impact on Prediction Accuracy.
Authors: Guo, Jin1 (AUTHOR), Zhang, Yuhong1,2 (AUTHOR) zhangyh@hrbnu.edu.cn, Kong, Depiao2,3 (AUTHOR), Zhao, Xueyao1,4 (AUTHOR), Li, Xue1,3 (AUTHOR), Luo, Chong2,4 (AUTHOR)
Source: Remote Sensing. May2026, Vol. 18 Issue 10, p1485. 26p.
Subjects: Sampling (Process), Digital soil mapping, Remote-sensing images, Organic compound content of soils, Forecasting, Environmental indicators, Statistical sampling, Boosting algorithms
Abstract: Highlights: What are the main findings? Integrating Sentinel-2 imagery from the April–May bare-soil period with selected environmental variables significantly improves the accuracy of the XGBoost prediction model for Soil Organic Matter (SOM). Using a soil-type-based Stratified Random (SR) sampling strategy allows the training dataset to be reduced to just 434 surface samples while maintaining stable predictive performance (R2 = 0.499, RMSE = 5.319 g/kg) and spatial pattern consistency. What are the implications of the main findings? Integrating environmental constraints with stratified thinning offers a cost-effective approach to maintain reliable mapping accuracy at significantly reduced sampling densities. This framework provides actionable guidance for optimizing surface soil monitoring networks and achieving low-cost, high-precision agricultural management, particularly in highly heterogeneous black-soil regions. High sampling density in digital soil mapping improves accuracy but increases costs. This study optimized sampling density for soil organic matter (SOM) mapping in Longjiang County using an XGBoost model, integrating Sentinel-2 imagery (April–May) and environmental variables. We evaluated the impact of reducing a dataset of 2059 surface soil samples using a soil-type-based stratified random (SR) sampling strategy. The optimal scale was determined using a cost–performance index restricting RMSE increases to under 10%. Results showed that integrating April–May environmental variables significantly improved model accuracy. Crucially, the SR strategy allowed training samples to be reduced to just 434 while maintaining stable predictive performance (R2 = 0.499, RMSE = 5.319 g/kg). Furthermore, SOM spatial prediction patterns at this optimized, lower density aligned closely with high-density results, showing no systematic bias. In conclusion, combining environmental constraints with stratified thinning allows regional SOM mapping to maintain reliable accuracy at significantly reduced sampling densities. This offers a cost-effective approach for optimizing soil monitoring networks in black-soil regions. [ABSTRACT FROM AUTHOR]
Copyright of Remote Sensing is the property of MDPI and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
Text:
  Availability: 1
Header DbId: egs
DbLabel: Engineering Source
An: 194141010
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Optimization of Sampling Density in Regional-Scale Soil Organic Matter Mapping and Its Impact on Prediction Accuracy.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Guo%2C+Jin%22">Guo, Jin</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Zhang%2C+Yuhong%22">Zhang, Yuhong</searchLink><relatesTo>1,2</relatesTo> (AUTHOR)<i> zhangyh@hrbnu.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Kong%2C+Depiao%22">Kong, Depiao</searchLink><relatesTo>2,3</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Zhao%2C+Xueyao%22">Zhao, Xueyao</searchLink><relatesTo>1,4</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Li%2C+Xue%22">Li, Xue</searchLink><relatesTo>1,3</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Luo%2C+Chong%22">Luo, Chong</searchLink><relatesTo>2,4</relatesTo> (AUTHOR)
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Remote+Sensing%22">Remote Sensing</searchLink>. May2026, Vol. 18 Issue 10, p1485. 26p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Sampling+%28Process%29%22">Sampling (Process)</searchLink><br /><searchLink fieldCode="DE" term="%22Digital+soil+mapping%22">Digital soil mapping</searchLink><br /><searchLink fieldCode="DE" term="%22Remote-sensing+images%22">Remote-sensing images</searchLink><br /><searchLink fieldCode="DE" term="%22Organic+compound+content+of+soils%22">Organic compound content of soils</searchLink><br /><searchLink fieldCode="DE" term="%22Forecasting%22">Forecasting</searchLink><br /><searchLink fieldCode="DE" term="%22Environmental+indicators%22">Environmental indicators</searchLink><br /><searchLink fieldCode="DE" term="%22Statistical+sampling%22">Statistical sampling</searchLink><br /><searchLink fieldCode="DE" term="%22Boosting+algorithms%22">Boosting algorithms</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Highlights: What are the main findings? Integrating Sentinel-2 imagery from the April–May bare-soil period with selected environmental variables significantly improves the accuracy of the XGBoost prediction model for Soil Organic Matter (SOM). Using a soil-type-based Stratified Random (SR) sampling strategy allows the training dataset to be reduced to just 434 surface samples while maintaining stable predictive performance (R2 = 0.499, RMSE = 5.319 g/kg) and spatial pattern consistency. What are the implications of the main findings? Integrating environmental constraints with stratified thinning offers a cost-effective approach to maintain reliable mapping accuracy at significantly reduced sampling densities. This framework provides actionable guidance for optimizing surface soil monitoring networks and achieving low-cost, high-precision agricultural management, particularly in highly heterogeneous black-soil regions. High sampling density in digital soil mapping improves accuracy but increases costs. This study optimized sampling density for soil organic matter (SOM) mapping in Longjiang County using an XGBoost model, integrating Sentinel-2 imagery (April–May) and environmental variables. We evaluated the impact of reducing a dataset of 2059 surface soil samples using a soil-type-based stratified random (SR) sampling strategy. The optimal scale was determined using a cost–performance index restricting RMSE increases to under 10%. Results showed that integrating April–May environmental variables significantly improved model accuracy. Crucially, the SR strategy allowed training samples to be reduced to just 434 while maintaining stable predictive performance (R2 = 0.499, RMSE = 5.319 g/kg). Furthermore, SOM spatial prediction patterns at this optimized, lower density aligned closely with high-density results, showing no systematic bias. In conclusion, combining environmental constraints with stratified thinning allows regional SOM mapping to maintain reliable accuracy at significantly reduced sampling densities. This offers a cost-effective approach for optimizing soil monitoring networks in black-soil regions. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of Remote Sensing is the property of MDPI and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.)
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=194141010
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.3390/rs18101485
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 26
        StartPage: 1485
    Subjects:
      – SubjectFull: Sampling (Process)
        Type: general
      – SubjectFull: Digital soil mapping
        Type: general
      – SubjectFull: Remote-sensing images
        Type: general
      – SubjectFull: Organic compound content of soils
        Type: general
      – SubjectFull: Forecasting
        Type: general
      – SubjectFull: Environmental indicators
        Type: general
      – SubjectFull: Statistical sampling
        Type: general
      – SubjectFull: Boosting algorithms
        Type: general
    Titles:
      – TitleFull: Optimization of Sampling Density in Regional-Scale Soil Organic Matter Mapping and Its Impact on Prediction Accuracy.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Guo, Jin
      – PersonEntity:
          Name:
            NameFull: Zhang, Yuhong
      – PersonEntity:
          Name:
            NameFull: Kong, Depiao
      – PersonEntity:
          Name:
            NameFull: Zhao, Xueyao
      – PersonEntity:
          Name:
            NameFull: Li, Xue
      – PersonEntity:
          Name:
            NameFull: Luo, Chong
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 15
              M: 05
              Text: May2026
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-print
              Value: 20724292
          Numbering:
            – Type: volume
              Value: 18
            – Type: issue
              Value: 10
          Titles:
            – TitleFull: Remote Sensing
              Type: main
ResultId 1