| 22 Jul 2025
Greenland Monthly Accumulation Maps (1960–2022): A Statistical Semi-Empirical Bias-Adjustment Model
Abstract. Accurate estimates of snow accumulation over the Greenland Ice Sheet (GrIS) are essential for reliable projections of sea-level rise. These are typically obtained from Regional Climate Models (RCMs), which carry substantial temporal and spatially variable biases, contributing to the metre-scale uncertainties in sea-level rise projections. While numerous studies have evaluated RCM bias using select in-situ observational datasets, many assessments are deduced from comparison to reanalysis datasets, which too carry substantial uncertainties. Such biases stem partly from the inability of RCMs and reanalysis products to assimilate point-based in-situ precipitation measurements directly. As a result, the rich network of observations from radar, ice cores, snow pits and stake networks remains under-utilised in systematic bias-correction of model accumulation.
In this study, we present a novel statistical-semi-empirical model for bias-correcting gridded accumulation output from any RCM or reanalysis product, utilising two million observational data points from the SUMup surface mass balance dataset. The method applies an empirical orthogonal function (EOF) decomposition to model accumulation output and adjusts the mean, climatology, EOFs and corresponding principle components (PCs) through a set of coefficients. The coefficients are calibrated by using a least squares optimisation that minimises the misfit between each component of the model accumulation and the in-situ observations. This allows us to reconstruct spatially complete bias-corrected accumulation maps. Here we apply this method to monthly accumulation output from HIRHAM5 (1960–2022), RACMO 2.4p1 (1980–2022), and CARRA reanalysis (1991–2022), identifying initial mean biases of -8.7 % (HIRHAM), +0.4 % (RACMO) and +10.9 % (CARRA). After adjustment, these are reduced to -0.1 %, -0.1 % and -0.2 %, respectively. Resulting bias-corrected mean annual accumulation rates over the ice sheet are estimated at 321 mm yr−1 (HIRHAM, 1960–2022), 375 mm yr−1 (RACMO, 1980–2022) and 384 mm yr−1 (CARRA, 1991–2022).
The framework outlined in this study offers a scalable, transferable solution for enhancing accumulation estimates, applicable to other climate models, variables, regions and observational datasets. The resulting bias-corrected accumulation fields offer an improved input to ice-sheet models, with the potential to reduce uncertainties in future sea-level rise projections through enhanced integration of observational data.
