Climate scenarios in CMIP6 multi-model projections for climate resilience planning in the upper stream of Tri An reservoir

Abstract

Climate change poses significant and multifaceted challenges to global hydrological systems and ecosystem integrity. Regions of particular ecological and economic importance, such as the upper stream of Tri An reservoir, face heightened vulnerability to these changes. While the Coupled Model Intercomparison Project Phase 6 (CMIP6) provides an advanced suite of Global Climate Models (GCMs) for projecting future conditions, significant challenges persist regarding the quantification of inter-model variability and the reduction of projection uncertainty. This research systematically evaluates and compares a selection of 5 GCMs from the CMIP6 archive. Utilizing the LARS-WG statistical downscaling platform, coarse-resolution model outputs are processed to generate high-resolution, localized climate change scenarios for the study area under a range of Shared Socioeconomic Pathways (SSPs). The study extends beyond trend analysis of core climatic elements, such as temperature and precipitation, across temporal and spatial dimensions. It specifically investigates the uncertainty spectrum inherent in the multi-model ensemble, assessing the degree of agreement and divergence among individual models. This evaluation identifies model-specific biases and performances in replicating regional climate dynamics, informing a more nuanced approach to ensemble construction and model confidence weighting. The outcomes of this analysis provide a critical evidence base for climate risk assessment and strategic planning. They are intended to support policymakers in developing robust, evidence-based adaptation and mitigation measures to enhance resilience and promote sustainable development within the area.