Majeed, Israr and Roy, S and Reddy, N N and Budama, N and Akuraju, V R and Anantha, K H and Niranjan, P and Singh, R and Jat, M L and Das, B S and Garg, K K (2025) Assessing Soil Degradation in Agricultural Landscapes of Semi-Arid Tropics Using Proximal and Remote Sensing-Based Diffuse Reflectance Spectroscopy. Soil Use and Management (TSI), 41 (2). pp. 1-18. ISSN 0266-0032
![[img]](http://oar.icrisat.org/style/images/fileicons/application_pdf.png) |
PDF
- Published Version
Available under License ["licenses_description_cc_attribution" not defined]. Download (3MB) |
Abstract
Monitoring soil degradation using the soil degradation index (SDI) is a complex process. Typically, multiple soil parameters are measured under laboratory conditions to create such a composite parameter. Because conventional soil testing methods are tedious and time-consuming, frequent monitoring of soil degradation through SDI continues to be a challenging task. With diffuse reflectance spectroscopy (DRS) emerging as a rapid soil testing method, the major objective of this study is to examine the DRS approach for estimating SDIs in a degradation-prone dryland landscape of Maharashtra, India. Accordingly, surface soil samples were collected from 141 locations and 20 different soil parameters were measured in these samples. Six key parameters were identified to formulate the SDI following a minimum dataset (MDS) approach: soil organic carbon content (SOC), soil erodibility index (eMCR), available S, available Mn, the ratio between exchangeable Ca to Mg and silt content. Spectral reflectance data collected under laboratory conditions and those extracted from multispectral imaging data from Sentinel-2 L2A over the visible to infrared (VNIR) region were used to estimate SDIs and its six indicators by calibrating two popular chemometric models: support vector regression (SVR) and feature selection-based partial-least-squares regression (PLSRFS). Results showed that the SDI values could be estimated from the laboratory-measured DRS data with the coefficient of determination (R2) value of 0.81 and root-mean-squared error (RMSE) value of 0.03. Similarly, chemometric models also performed well for the MSI data (R2 = 0.52; RMSE = 0.04). Although the laboratory-based DRS approach provided greater estimation accuracy, low RMSE values associated with the MSI data showed that SDI may be effectively mapped for the entire study area at high spatial resolution (~10 m for Sentinel-2 L2A data). Correlation analyses between mapped SDI and crop yield further showed yield declines with increasing soil degradation for different rainfed crops, while no such trends were observed for the irrigated crops, suggesting that irrigation management in dryland areas may circumvent land degradation challenges.
| Item Type: | Article |
|---|---|
| Divisions: | Global Research Program - Resilient Farm and Food Systems |
| CRP: | UNSPECIFIED |
| Uncontrolled Keywords: | chemometric models, crop yield, soil degradation index |
| Subjects: | Others > Remote Sensing Others > Semi-Arid Tropics Others > Crop Yield |
| Depositing User: | Mr Nagaraju T |
| Date Deposited: | 17 Jun 2025 05:48 |
| Last Modified: | 17 Jun 2025 05:48 |
| URI: | http://oar.icrisat.org/id/eprint/13164 |
| Official URL: | https://bsssjournals.onlinelibrary.wiley.com/doi/f... |
| Projects: | UNSPECIFIED |
| Funders: | UNSPECIFIED |
| Acknowledgement: | The authors gratefully acknowledge the funding support received from Groundwater Survey and Development Agency Government of Maharashtra under the Atal Bhujal Yojana for establishing seven benchmarking sites during 2023–2026. Support of CGIAR- science program of Multifunctional Landscape (MFL) is also acknowledged for partially covering ICRISAT scientists’ time for undertaking this study. Authors are also thankful to the field and laboratory staff for their meticulous efforts in collecting soil samples and their analysis. |
| Links: |
Actions (login required)
 |
View Item |
Altmetric
Altmetric