Sustainable Agriculture and Ecosystem Health: New Research Offers Insights for Efficient Nutrient Management

Researchers have made a significant discovery in the field of sustainable agriculture and ecosystem health, presenting a new approach to nutrient management that could have a substantial impact on farming practices and environmental sustainability. A recent study, sponsored by the World Bank and the Karnataka Watershed Development Department, used Visible-Near-Infrared (Vis-NIR) spectroscopy to predict soil properties and classify nutrient levels for improved agricultural decision-making.

Key Takeaways:

• The study, published in Land Degradation & Development, analyzed a dataset of 216 soil samples from diverse land uses in the Gummlapalli subwatershed, Karnataka, India, for 11 soil properties, including pH, soil organic carbon (SOC), macronutrients, and micronutrients.
• The research used Partial Least Squares Regression (PLSR) model enhanced with Savitzky-Golay (SG) smoothing and Standard Normal Variate (SNV) transformation, which showed variable prediction performance for key properties, including pH (R2: 0.70), SOC (R2: 0.61), P2O5 (R2: 0.84), and K2O (R2: 0.50).
• Two approaches for nutrient classification were evaluated: indirect classification based on predicted soil properties and direct classification using Partial Least Squares Discriminant Analysis (PLS-DA). Direct classification outperformed the indirect approach, achieving higher overall accuracy for key properties, including pH (OA: 0.72), SOC (OA: 0.61), P2O5 (OA: 0.84), K2O (OA: 0.50), and Cu (OA: 0.96).
• The study highlights the potential of Vis-NIR spectroscopy as a robust tool for soil nutrient classification, enabling precise fertilizer recommendations and promoting environmental sustainability.
• The research has been peer-reviewed, and additional information on the study can be obtained from the Indian Council of Agricultural Research (ICAR) National Bureau of Soil Survey and Land Use Planning.

Statistics:

• 216 soil samples were analyzed for 11 soil properties.
• The PLSR model, enhanced with SG smoothing and SNV transformation, showed variable prediction performance for key properties, with R2 values ranging from 0.04 to 0.70.
• Direct classification outperformed indirect classification, achieving higher overall accuracy (OA) for key properties, including pH (0.72), SOC (0.61), P2O5 (0.84), K2O (0.50), and Cu (0.96).
• The study was sponsored by the World Bank and the Karnataka Watershed Development Department.

Sources:

• NewsRx. New Sustainable Food and Agriculture Findings from Indian Council of Agricultural Research (ICAR) National Bureau of Soil Survey and Land Use Planning Reported (Predicting Soil Nutrient Classes Using Vis-nir Spectroscopy To Support Sustainable ...). Ecology, Environment & Conservation. August 8, 2025; p 563.
• Predicting Soil Nutrient Classes Using Vis-nir Spectroscopy To Support Sustainable Farming Decisions. Land Degradation & Development, 2025.