Remote sensing for sustainable river management: Estimating riverscape vulnerability for Ganga, the world's most densely populated river basin

Surface water mixed with wastewater creates serious environmental concerns, particularly in densely populated urban areas with inadequate infrastructure. Such contamination threatens to cause major public health crises in the Ganga Basin where monsoonal flooding converges with 6 billion liters of untreated sewage that is discharged daily into the basin by 650 million people. GIS-based analytic hierarchy process (AHP) with remote sensing data was conducted to highlight areas of vulnerability along a 20-km wide riverscape. Analytic network process (ANP), Nested AHP, fuzzy AHP, and 1-N AHP (novel variant of AHP) were used to constrain AHP model uncertainties, and composites of these analyses were utilized to define the vulnerability of the river Ganga to pollution. AHP categorized 83.7% of the area as having extremely low or low vulnerability and 3.5% of the area as having highly or extremely high vulnerability. ANP and Nested AHP produced focused, yet dampened, vulnerability-score maps compared to AHP. Fuzzy AHP and 1-N AHP detected sensitivities to factor variability and potential unknown acute and chronic factors. While fuzzy AHP identified quintile-level changes in vulnerability based on scenario parameters, vulnerability scores of 1-N AHP and AHP showed no major differences. Normalized composite vulnerability (\geq)2 standard deviations highlighted particularly vulnerable locations and identified instances where network effects were greater than factor class and vice versa. Together, these analyses located areas of extreme vulnerability at the nexus of river Ganga and urban landscapes as well as regions of low vulnerability potentially suitable for conservation efforts or sustainable development practices to prevent their degradation.

💡 Research Summary

This study presents a comprehensive spatial vulnerability assessment of the Ganga River basin, focusing on a 20‑km‑wide corridor (10 km on each side of the main channel) that stretches 1,330 km from Haridwar to the confluence with the Ghaghara River. The authors integrate satellite‑derived remote sensing products (Landsat‑8 Level 2 surface reflectance, NASADEM‑derived DEM, and ancillary climate and demographic datasets) with a suite of multi‑criteria decision‑making (MCDM) techniques to quantify the susceptibility of riverine environments to pollution originating from densely populated urban areas.

The core analytical framework is the Analytic Hierarchy Process (AHP). Six key factors are selected: population density (PD), land‑use/land‑cover (LULC), annual rainfall, drainage density (DD), slope, and land‑surface temperature (LST). Each factor is classified into five or six natural‑break categories using the Jenks algorithm, and a pair‑wise comparison matrix is constructed based on expert judgment. The resulting Saaty‑scale weights (PD = 1, LULC = 2, rainfall = 4, DD = 5, slope = 7, LST = 9) are multiplied by the class scores on a per‑pixel basis within Google Earth Engine, producing a continuous vulnerability index that is later discretized into five risk levels. The baseline AHP model classifies 83.7 % of the study area as low or very low vulnerability, 12.8 % as moderate, and only 3.5 % as high or very high.

Recognizing well‑documented limitations of AHP—particularly the assumption of criterion independence and sensitivity to weight perturbations—the authors implement four complementary variants:

1. Analytic Network Process (ANP): By constructing a super‑matrix that captures feedback loops among criteria, ANP relaxes the strict hierarchy of AHP. The resulting vulnerability map shows a dampened distribution of high‑risk scores, indicating that network effects moderate the influence of any single factor.

2. Nested AHP: Each criterion is re‑evaluated against itself using non‑linear transformations (e.g., logarithmic, quadratic) to model diminishing or accelerating impacts of extreme class values. This approach accentuates hotspots where population density and urban land‑use are simultaneously high, sharpening the spatial contrast between vulnerable and resilient zones.

3. Fuzzy AHP: Triangular fuzzy numbers replace crisp pair‑wise judgments, allowing the authors to conduct a sensitivity analysis on weight uncertainty. The fuzzy analysis reveals that rainfall and drainage density are most prone to rank reversal when their fuzzy bounds are widened, underscoring the importance of robust climate data for reliable vulnerability estimates.

4. 1‑N AHP (Novel Variant): Building on the fuzzy framework, 1‑N AHP introduces a stochastic “unknown” variable (N) that models hidden or unmeasured influences as a probability distribution over the weight vector. While the overall spatial pattern mirrors the classic AHP, localized spikes in the 1‑N output flag potential acute pollution sources that may be missed by deterministic weighting.

All five models are normalized and combined into a composite vulnerability index. Areas exceeding the mean ± 2 standard deviations are flagged as “extremely vulnerable.” These zones cluster at the interface of dense urban settlements and the river channel—particularly around Haridwar, the industrial belt near Kanpur, and the confluence region at Sitab Diara—where both factor‑level effects (high PD, urban LULC) and network‑level effects (high DD, steep slopes) reinforce each other. Conversely, low‑vulnerability patches correspond to sparsely populated upland sections with low drainage density and predominantly natural land cover, suggesting they are suitable candidates for conservation or sustainable development initiatives.

The paper contributes methodologically by (i) demonstrating how remote sensing data can be seamlessly integrated with advanced MCDM techniques for large‑scale riverine risk mapping, (ii) providing a systematic comparative evaluation of AHP’s variants to diagnose and mitigate model uncertainty, and (iii) delivering actionable spatial intelligence for policymakers tasked with prioritizing remediation, infrastructure upgrades, and habitat protection in one of the world’s most populated watersheds. The authors argue that the workflow is transferable to other densely populated river basins (e.g., the Yangtze, Mekong) and can support the broader goals of water security and sustainable urban planning.