Chen et al. (2026) Remote sensing-based mapping of flood irrigation extent and depth during fallow periods in arid irrigation district

Identification

• Journal: Agricultural Water Management
• Year: 2026
• Date: 2026-07-16
• Authors: Xinyi Chen, Songhao Shang
• DOI: 10.1016/j.agwat.2026.110644

Research Groups

• Department of Hydraulic Engineering, Tsinghua University, Beijing 100084, China

Short Summary

This study developed and applied an improved optical trapezoid model (OPTRAM) with Sentinel-2 data to map the extent and estimate the depth of autumn and spring flood irrigation during fallow periods (2016-2023) in the Hetao Irrigation District, China, providing crucial information for water management.

Objective

• To map irrigation extent and irrigation depth in fallow periods and to understand the spatiotemporal patterns of autumn and spring irrigation applications in an arid irrigation district in Northwest China.

Study Configuration

• Spatial Scale: Hetao Irrigation District (HID), Northwest China, covering a total land area of 12,000 km² and an irrigated area of 7,700 km², divided into five sub-irrigation districts. Remote sensing data resolution of 10-20 meters.
• Temporal Scale:
  • Spring irrigation extent mapping: 2016–2023 (April to May).
  • Autumn and spring irrigation depth estimation: 2019–2023.
  • Field surveys: Autumn 2020 and Spring 2021.

Methodology and Data

• Models used:
  • Improved Optical Trapezoid Model (OPTRAM) for identifying spring irrigation extent.
  • Linear relationship between the ratio of Sentinel-2 Green (B3s) and Near-infrared (B8s) band reflectances and shallow water depth for irrigation depth estimation.
• Data sources:
  • Satellite: Sentinel-2 Level-1C (for extent mapping) and Level-2A (for depth simulation) images from Google Earth Engine (GEE).
  • Observation: Field surveys for spring irrigation extent (2485 spring irrigated, 15 non-spring irrigated points) and actual water depth (102 sampling points).
  • Auxiliary: Autumn irrigation extent maps (2016–2023) from Chen et al. (2024), autumn irrigation depth maps (2019–2020) from Qian et al. (2024) (MODIS 500 m resolution), and statistical data of spring irrigation area from the Bayannur Institute of Water Conservancy Research.

Main Results

• The improved OPTRAM achieved correct ratios of 0.94 for spring irrigated and 1.00 for non-spring irrigated validation points.
• Simulated irrigation depth showed a coefficient of determination (R²) of 0.82 and a Root Mean Square Error (RMSE) of 0.046 meters.
• Comparison with existing MODIS-based autumn irrigation depth maps showed a total Normalized RMSE (NRMSE) of approximately 15%.
• Spring irrigation cropland is mainly located in the north of Jiefangzha, north of Yongji, south of Yichang, and north of Wulate sub-irrigation districts. Yichang had the largest average spring irrigated area of 1091.91 km².
• The inter-annual trend of spring irrigation area generally increased in most sub-irrigation districts, while autumn irrigation area showed a predominantly decreasing trend.
• Irrigation depths in Yichang and Wulate sub-irrigation districts tend to be larger (often exceeding 0.2 meters, up to 0.6 meters) compared to other sub-districts (generally within 0.2 meters).
• Most croplands are irrigated once during fallow periods, but 15.41% to 29.02% of the total autumn/spring irrigated area from 2016 to 2022 was double irrigated (both autumn and spring), primarily in Jiefangzha.
• Autumn irrigation depth is generally greater than spring irrigation depth, except in Wulate, where spring irrigation depth is consistently larger.
• In 2023, autumn irrigation depth sharply decreased across all sub-irrigation districts due to water use restrictions, while spring irrigation depth remained relatively stable.
• The calculated canal efficiency for autumn irrigation in 2020 was 0.46, close to the statistical value of 0.50.

Contributions

• Provides the first comprehensive investigation of flood irrigation extent and depth for both autumn and spring fallow periods with enhanced spatial details (10-20 meter resolution Sentinel-2 data).
• Validates and applies an improved OPTRAM that effectively identifies irrigation extent across different fallow seasons using the same set of parameters, reducing reliance on precise cropland maps.
• Delivers detailed spatiotemporal patterns of irrigation extent and depth, including the identification of double-irrigated areas and quantitative comparisons between autumn and spring irrigation practices.
• Offers updated flood irrigation maps and critical information for improved irrigation water management, particularly in the context of decreasing water availability and the need to balance water conservation with soil salinization control.
• Highlights the influence of changing crop varieties and planting patterns on irrigation water requirements and practices.

Funding

• National Natural Science Foundation of China (grant number 52279038)

Citation

@article{Chen2026Remote,
  author = {Chen, Xinyi and Shang, Songhao},
  title = {Remote sensing-based mapping of flood irrigation extent and depth during fallow periods in arid irrigation district},
  journal = {Agricultural Water Management},
  year = {2026},
  doi = {10.1016/j.agwat.2026.110644},
  url = {https://doi.org/10.1016/j.agwat.2026.110644}
}