Use of Multi-Criteria Decision Analysis (MCDA) for Identification of Hazard Prone Lakes in Chitral River Basin

Abstract

Glacial Lake Outburst Floods (GLOFs) are a pressing concern in Pakistan and other parts of the world, posing an immediate threat to downstream communities, infrastructure, and agriculture. Urgent action is required to mitigate the devastating impact of these floods and safeguard the lives and livelihoods of those at risk. The far-reaching and hazardous consequences of Glacial Lake Outburst Floods (GLOFs) are a sobering reminder of the urgent need for effective measures to prevent and mitigate the impact of these catastrophic events. Up till today, no work has been done on Chitral lakes while using Multi-Criteria Decision Analysis (MCDA) to identify hazard-prone lakes. The study was carried out to identify the hazardous lakes in the Chitral basin with the help of MCDA. The study provides valuable insights into all the risk factors related to GLOF and informs the decision-making process while keeping in view the threat linked to a specific lake. Two software ArcMap 10.5 and Stochastic Multi-Criteria Acceptability Analysis-SMAA (a multi-criteria decision-making tool) were used for digitization, criteria assessment, and sensitivity analysis, respectively. Various criteria were chosen from the literature to study the GLOF events Chitral region. Remote sensing data of Sentinel 2 imagery was used to digitize the maps. Seismic data was taken from USGS, and reanalysis data for temperature and precipitation from ERA-5 and GPCC was used. Following lake selection criteria (lake >0.02 km2, distance <1000m) was taken. All other criteria were defined using ArcMap, Bing Maps, and Google Earth. Around 130 lakes were identified, and 13 were found to be completing our criteria. The study used 10 criteria to study lake behaviour and after determining threshold values, 6 out of 13 lakes were found to be at the highest risk, and one lake was found at medium risk. A sensitivity analysis was performed to predict the future impacts of lakes. Criteria weight sensitivity analysis in SMAA-2 shows the most hazardous criteria in each lake. SMAA-TRI performed a sensitivity analysis to show how previously found lakes changed their risk classes from low/medium to high. The results showed that the developed model is effective in identifying hazard-prone lakes, and the study is relevant to areas (lakes) with no history of GLOF (Glacial Lake Outburst Floods). Further investigation is suggested for high risk posing lakes and continuous monitoring is also required for low-risk lakes along with some other criteria and field investigation.