Adhvaryu et al. (2025) Multi-Objective Optimization of Irrigation Canal Network Using Geospatial Computing: A Case Study of the Kadi Narmada Main Canal, Gujarat

Identification

• Journal: ISPRS annals of the photogrammetry, remote sensing and spatial information sciences
• Year: 2025
• Date: 2025-12-19
• Authors: Janki Adhvaryu, Parul Patel
• DOI: 10.5194/isprs-annals-x-5-w2-2025-1-2025

Research Groups

• Civil Engineering Department, NIRMA University, Ahmedabad, India.
• Accionland Private Limited (Collaborative support for UAV surveys).

Short Summary

This study integrates multi-temporal satellite imagery, high-resolution UAV data, and geophysical measurements into a multi-objective optimization framework to mitigate canal seepage. The findings demonstrate that selective lining of 15% of the canal length can reduce seepage losses by approximately 20% without compromising irrigation delivery.

Objective

• To develop a geospatially driven computational framework using the Non-Dominated Sorting Genetic Algorithm II (NSGA-II) to minimize seepage losses and waterlogging while sustaining irrigation reliability.

Study Configuration

• Spatial Scale: Kadi branch of the Narmada Main Canal, Gujarat, India (23°18′06″ N, 72°19′35″ E); analysis conducted within a 1 km corridor of the canal alignment with 250 m segment resolution.
• Temporal Scale: Long-term historical analysis spanning 34 years (1990–2024) using multi-temporal satellite datasets.

Methodology and Data

• Models used: Non-Dominated Sorting Genetic Algorithm II (NSGA-II) for multi-objective optimization; Mann–Kendall trend test and Sen’s slope estimator for temporal trend analysis; Getis–Ord Gi* statistic for spatial hotspot clustering.
• Data sources:
  • Satellite: Landsat 5 TM, 7 ETM+, and 8 OLI (30 m resolution) for NDVI and NDWI indices.
  • UAV: DJI Mavic Pro ortho-mosaics (3 cm ground sampling distance) for high-resolution surface characterization.
  • Geophysical: Electromagnetic induction and Vertical Electrical Sounding (VES) for subsurface moisture and conductivity (up to 15 m depth).
  • Ancillary: Differential GPS (DGPS) for positional accuracy (< 3 cm), rainfall records, soil maps, and canal discharge data.

Main Results

• Seepage Correlation: Established a strong positive correlation (r ≈ 0.65, p < 0.01) between the Normalized Difference Water Index (NDWI) and apparent electrical conductivity, validating NDWI as a proxy for subsurface seepage.
• Vegetation Dynamics: NDVI values in seepage-affected zones increased from approximately 0.2 in the 1990s to 0.4 in the 2000s-2020s, indicating persistent moisture-induced growth.
• Optimization Efficiency: The Pareto-optimal "knee-point" solution identified that lining 15% of the canal segments and implementing 10 drainage clusters achieves a 20% reduction in seepage and a 17% reduction in waterlogged area.
• Temporal Trends: Analysis revealed a progressive expansion of waterlogged patches since 2000, coinciding with increased canal flow and aging infrastructure.

Contributions

• Methodological Integration: Combines multi-scale sensing (satellite, UAV, and ground geophysics) with evolutionary algorithms to move beyond traditional manual canal inspections.
• Targeted Intervention: Provides a quantitative basis for "selective lining," offering a more cost-effective alternative to uniform canal rehabilitation.
• Decision Support: Develops a scalable, data-driven tool for irrigation managers to balance the trade-offs between water conservation, environmental degradation (salinization/waterlogging), and infrastructure costs.

Funding

• NIRMA University, Ahmedabad, India.
• Accionland Private Limited (Technical support and drone survey equipment).

Citation

@article{Adhvaryu2025MultiObjective,
  author = {Adhvaryu, Janki and Patel, Parul},
  title = {Multi-Objective Optimization of Irrigation Canal Network Using Geospatial Computing: A Case Study of the Kadi Narmada Main Canal, Gujarat},
  journal = {ISPRS annals of the photogrammetry, remote sensing and spatial information sciences},
  year = {2025},
  doi = {10.5194/isprs-annals-x-5-w2-2025-1-2025},
  url = {https://doi.org/10.5194/isprs-annals-x-5-w2-2025-1-2025}
}

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