Vasala et al. (2026) Spatially explicit and regionalized quantification of blue and green crop water footprints in a water-stressed river basin

Identification

• Journal: Water Science & Technology
• Year: 2026
• Date: 2026-01-13
• Authors: Saicharan Vasala, Shwetha H. R.
• DOI: 10.2166/wst.2026.198

Research Groups

• Department of Water Resources and Ocean Engineering, National Institute of Technology Karnataka, Surathkal, India.
• Department of Civil Engineering, Sharda University, Greater Noida, India.

Short Summary

This study develops a regionalized remote sensing framework to quantify blue and green water footprints for major crops in the Upper Cauvery Basin, revealing that while most crops rely on green water, summer paddy and rabi ragi exert significant pressure on blue water resources. The research demonstrates that replacing generalized crop coefficients with real-time vegetation coefficients significantly improves the accuracy of agricultural water management data.

Objective

• To quantify the seasonal blue (BWF), green (GWF), and total water footprints (TWF) of key water-intensive crops in the Upper Cauvery River Basin (UCB) using a spatially explicit remote sensing methodology.
• To evaluate the sustainability of different crop variants by distinguishing between rainfed and irrigated water consumption at a fine regional scale.

Study Configuration

• Spatial Scale: Upper Cauvery River Basin, Karnataka, India (approximately 10,958 $km^2$), analyzed at the Gram Panchayat (local administrative) level.
• Temporal Scale: 2015–2019, capturing seasonal variations (Kharif, Rabi, and Summer) and including severe drought years (2016).

Methodology and Data

• Models used:
  • Penman-Monteith (FAO-56): For calculating reference evapotranspiration ($ETo$).
  • Stacked Ensemble (Random Forest + XGBoost): Used to downscale $ETo$ from 50 km to 500 m resolution.
  • Vegetation Coefficient ($Kv$) Method: Derived from Green Vegetation Moisture Index (GVMI), Temperature Vegetation Dryness Index (TVDI), and fractional vegetation cover ($fc$) to estimate actual evapotranspiration ($ETa$).
  • USDA Soil Conservation Service (SCS) Method: For calculating effective rainfall ($P{eff}$).
  • Random Forest Algorithm: For 10 m resolution seasonal crop-type classification using Sentinel-2 imagery.
• Data sources:
  • Satellite: MODIS (NDVI, LST, Surface Reflectance), GPM IMERG (Rainfall), SRTM DEM (Topography), and Sentinel-2 (Crop mapping).
  • Meteorological: NASA POWER (Temperature, wind speed, dew point).
  • Observational: Crop yield data from the Directorate of Economics and Statistics, Government of Karnataka.

Main Results

• Water Footprint Extremes: Ragi (Rabi-RF) recorded the highest TWF at 4,212.2 $m^3/ton$, while Maize (Kharif-IRR) was the most efficient with a TWF of 258.075 $m^3/ton$.
• Irrigation Pressure: Paddy (Summer-IRR) exhibited a high BWF (436.9 $m^3/ton$), reflecting heavy reliance on blue water due to minimal summer rainfall.
• Drought Impact: The 2016 drought caused a massive spike in the BWF of Rabi ragi, peaking at 8,202.6 $m^3/ton$, highlighting systemic water use inefficiency during water-stressed periods.
• Model Performance: The stacked ensemble downscaling model achieved an $R^2$ of 0.784, and the crop classification model maintained an accuracy exceeding 84% across all seasons.
• Sustainability: Crops like Kharif maize and Kharif ragi were found to be more sustainable due to their dominance in green water usage (rainwater) compared to blue water (irrigation).

Contributions

• Methodological Innovation: Introduces a two-stacked ensemble machine learning approach (RF + XGBoost) for high-resolution $ET_o$ downscaling.
• Regionalization: Replaces static, generalized FAO $Kc$ tables with dynamic, remote sensing-derived vegetation coefficients ($Kv$) tailored to local physiological and climatic conditions.
• Granularity: Provides the first Gram Panchayat-level (county-level) assessment of water footprints in the Cauvery Basin, offering a blueprint for localized water policy and irrigation scheduling.

Funding

• The research utilized data and frameworks associated with the National Hydrological Project (NHP) and the National Institute of Technology Karnataka. No specific grant reference codes were explicitly listed in the text.

Citation

@article{Vasala2026Spatially,
  author = {Vasala, Saicharan and R., Shwetha H.},
  title = {Spatially explicit and regionalized quantification of blue and green crop water footprints in a water-stressed river basin},
  journal = {Water Science & Technology},
  year = {2026},
  doi = {10.2166/wst.2026.198},
  url = {https://doi.org/10.2166/wst.2026.198}
}

Generated by BiblioAssistant using gemini-3-flash-preview (Google API)