Shifting Green-to-Blue Water Footprints Partitioning in Irrigated Croplands across the Canadian Prairies: A Remote Sensing and Machine Learning Assessment

Reliable estimates of irrigated agricultural extent and associated water consumption are critical for sustainable water resources management, especially in farming regions facing rising climatic variability and water scarcity. This study developes an integrated remote sensing and machine learning framework to map irrigated croplands and quantify field-scale green, blue, and consumptive water footprints across the Canadian Prairie Provinces (Alberta, Saskatchewan, and Manitoba) for 2010-2024. We employ 30-m reoslution Landsat imagery, vegetation indices, and climate drivers as predictors for five machine learning classifiers, including Random Forest (RF), Support Vector Machine, K-Nearest Neighbors, Gradient Tree Boost (GTB), Classification and Regression Trees (CART), and Multi-Model Ensemble Mapping approach for irrigated cropland detection. MODIS-derived evapotranspiration products are subsequently used to compute water footprints at the field scale. RF yields the highest overall accuracy (~90%), followed by GTB and CART. While total agricultural extent remained largely stable from 2015 to 2019, the partitioning of agricultural water consumption underwent marked reorganization. Between 2010 and 2024, the green water footprint declined by 16%, whereas the blue water footprint increased more than fivefold (from 2.34 to 12.84 million m³), concentrated primarily in southern Alberta and southwestern Saskatchewan. Accordingly, blue water's share of total consumptive water use rose from 5% in 2010 to 31% in 2020 before falling to 25% by 2024. The total consumptive water footprint increased only moderately (6.5%), signaling a substantive shift in the water sources sustaining regional crop production. These spatially explicit findings support the identification of agricultural water-use hotspots and advance irrigation management and drought adaptation planning for the Canadian Prairies.

Authors

Institutions

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-16
DOI
https://doi.org/10.5281/zenodo.24273603
Primary Topic
Remote Sensing in Agriculture
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Shifting Green-to-Blue Water Footprints Partitioning in Irrigated Croplands across the Canadian Prairies: A Remote Sensing and Machine Learning Assessment

Shahbaz Nasir Khan, Miaohua Mao, Muhammad Umar Akbar, M. I. Abdul Aleem et al.
Zenodo (CERN European Organization for Nuclear Research)
Remote Sensing in Agriculture
article

Shifting Green-to-Blue Water Footprints Partitioning in Irrigated Croplands across the Canadian Prairies: A Remote Sensing and Machine Learning Assessment

Shahbaz Nasir Khan, Miaohua Mao, Muhammad Umar Akbar, M. I. Abdul Aleem, Hammad Ul Hussan, Hailong He, Muhammad Ahmad Akram, Muhammad Maaz Tanveer
article en

Abstract

Reliable estimates of irrigated agricultural extent and associated water consumption are critical for sustainable water resources management, especially in farming regions facing rising climatic variability and water scarcity. This study developes an integrated remote sensing and machine learning framework to map irrigated croplands and quantify field-scale green, blue, and consumptive water footprints across the Canadian Prairie Provinces (Alberta, Saskatchewan, and Manitoba) for 2010-2024. We employ 30-m reoslution Landsat imagery, vegetation indices, and climate drivers as predictors for five machine learning classifiers, including Random Forest (RF), Support Vector Machine, K-Nearest Neighbors, Gradient Tree Boost (GTB), Classification and Regression Trees (CART), and Multi-Model Ensemble Mapping approach for irrigated cropland detection. MODIS-derived evapotranspiration products are subsequently used to compute water footprints at the field scale. RF yields the highest overall accuracy (~90%), followed by GTB and CART. While total agricultural extent remained largely stable from 2015 to 2019, the partitioning of agricultural water consumption underwent marked reorganization. Between 2010 and 2024, the green water footprint declined by 16%, whereas the blue water footprint increased more than fivefold (from 2.34 to 12.84 million m³), concentrated primarily in southern Alberta and southwestern Saskatchewan. Accordingly, blue water's share of total consumptive water use rose from 5% in 2010 to 31% in 2020 before falling to 25% by 2024. The total consumptive water footprint increased only moderately (6.5%), signaling a substantive shift in the water sources sustaining regional crop production. These spatially explicit findings support the identification of agricultural water-use hotspots and advance irrigation management and drought adaptation planning for the Canadian Prairies.

Zenodo (CERN European Organization for Nuclear Research)
Oklahoma State University (US), Peoples' Friendship University of Russia (RU), Chongqing University (CN), Chinese Academy of Sciences (CN), Yantai Institute of Coastal Zone Research (CN), Aerospace Information Research Institute (CN), University of Chinese Academy of Sciences (CN), University of Manitoba (CA), University of Agriculture Faisalabad (PK)
Remote Sensing in Agriculture
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon