FLASH FLOOD VULNERABILITY, SUSCEPTIBILITY & MODELING FOR HIMALAYAN MOUNTAIN RIVER BASINS IN UTTARAKHAND

Abstract

Flash floods are one of the most destructive natural hazards, capable of causing widespread damage. The mountain catchments are among the most vulnerable areas experiencing frequent flash floods. Predicting flash floods with a considerable buffer period in terms of time duration is impractical. Therefore, to minimize the damage-causing potential of flash floods, it is necessary to assess areas vulnerable and susceptible to flash floods before any mitigative actions are taken. The mountain watersheds in the Uttarakhand State, India are also subjected to rapid change due to various human-induced factors like economic expansion, globalization, infrastructural improvement, shifting land use patterns, migration, and urbanization. Thus, intense efforts are required to preserve the high-mountain Himalayan ecosystem to combat the challenges mentioned above. However, significant challenges are posed against countering these problems due to the sparse weather monitoring infrastructure and hydrometeorological data gap/sharing issues. Keeping the above-mentioned problems in context, flash flood prioritization has been carried out to generate vulnerability zonation maps. Morphometric parameters were derived for 29 sub-watersheds of the Upper Ganga Basin (UGB). The sub-watersheds were prioritized for flash flood vulnerability by employing Weighted Sum Approach (WSA), Principal Component Analysis (PCA), and an Integrated Approach (IA). A comparative assessment of the three approaches revealed that the results obtained by PCA and IA are relatively analogous in terms of the number of sub-watersheds classified under very high, high, and very low vulnerability zones. The IA methodology was validated through spatial analysis of previously occurred flash flood event locations in 2018 and 2019. To assess the vulnerable zones, remote sensing-based population distribution analysis and geospatial distribution analysis of 6360 towns and villages were carried out. The investigation revealed that 5 densely populated sub-watersheds fall under very high and high vulnerability zones. The medium vulnerability zone also emerged as a critical zone. Furthermore, Mandakini River Basin, which emerged as a critical sub-watershed, was considered for the flash flood susceptibility assessment. Five bivariate statistical models, namely Frequency Ratio (FR), Fuzzy Membership Value (FMV), Weights of Evidence (WOE), Statistical Index (SI), and Information Value (IV), were individually integrated with the Index of Entropy (IOE). These models were employed to calculate the flash flood potential index and identify susceptible zones in the Mandakini River Basin. 39 flash flood locations, 39 non -flash flood locations, and 15 flash flood conditioning factors were utilized for training and testing the models with 70% and 30% of the dataset, respectively. The model performances were examined using receiver operating characteristics curves. This best prediction rate performance was featured by SI-IOE and IV-IOE with an Area Under the Curve (AUC)= 0.896 followed by WOEIOE (AUC = 0.889). The results revealed that the areas with high and very high susceptibility cover approximately 40% of the basin area. Further, the Chamoli flash flood event of 7th February 2021 has been modeled from the source up to 0.8 km downstream using a hydrodynamic modeling approach. Hydrologic Engineering Center - River Analysis System (HEC-RAS) software was employed to model a hypothesized storage breach of 26.4 × 106 m3 at the source location, generating a peak inflow of 12761.88 m3/s. The breach was simulated as an unsteady flow with a computational interval of 5 seconds and a mapping interval of 2 minutes for 6 hours. The model-generated peak discharge values range between 7908.8 and 7975.26 m3/s near Rishiganga Hydro-Electric Project (HEP) and between 5779.53 and 5957.46 m3/s near Tapovan HEP. Also, flow depths at the above locations were 19.85 and 18.15 m, respectively. The flow velocities were 6.92 and 3.86 m/s, respectively. The model output shows good agreement with the extent and height of actual debris assessed using systematic pre and post-event analysis of high-resolution satellite datasets. Also, the same event was modeled for debris flow using the Rapid Mass Movements-Debris Flow (RAMMS-DF) model to derive important flow characteristics like velocity, height, and pressure. The model is carefully calibrated after several iterations, and finally, the optimal friction values of μ and ε were defined as 0.1 and 300 m/s2, respectively. The vulnerability and susceptibility maps can be vital input for stakeholders and decision-makers for future planning. The modeling and damage assessment approach may be applied in dynamic mountain ecosystems where population and infrastructure growth require continuous evaluation of hazards. Finally, based on the influence of changing climate on occurrences of disasters and infrastructure, with particular emphasis on the future climate projections in Uttarakhand, recommendations are proposed for a way forward toward resilience and long-term sustainability of the infrastructure and mountain communities in the region.