Evaluating the Role of Internal Climate Variability and Bias Adjustment Methods on Decadal Glacier Projections

Abstract

Glacier mass loss is one of the main contributors to sea-level rise and poses challenges for future water resources. Refining glacier projections and sources of uncertainty thus supports climate adaptation and mitigation. Here we explicitly quantify the impact of internal climate variability and climate data bias adjustment methods on regional and global glacier projections through 2100 for various emissions scenarios. Uncertainty from internal climate variability is comparable to climate model structural uncertainty (i.e., arising from physical representations and parameter settings) in the coming decades at the regional level, but is not a major source of uncertainty in centennial global glacier projections. Bias adjustment options (method and time period) moderately impact projections at regional and glacier scales, but have a smaller impact (∼2% of global glacier mass at 2100, relative to 2020) at global scales. In some regions, the uncertainty due to internal climate variability is larger than climate model structural uncertainty for the entirety of the 21st century, and bias adjustment options can nearly double the regional uncertainty by 2100. At the glacier scale, bias adjustments can lead to differences in projected decadal and centennial mass loss of up to 30%, although these greatest differences are associated with the smallest (<1 km²) glaciers. Overall, internal climate variability and climate data bias adjustment methods are important to consider, especially in regional applications, to better estimate uncertainty in future sea-level rise and water resources availability.