Open Collections

Moore, Ruth Elizabeth Mary

University of British Columbia

Considerable progress has been made in understanding recent Arctic temperature changes, but precipitation changes have been less documented. This is partly due to the lack of in-situ observations, and consequently, most studies of precipitation changes are based on atmospheric reanalyses (assimilated observational datasets). We evaluate how well ERA5, a widely used reanalysis, reproduces observed precipitation (AHCCD dataset) in the Canadian Arctic, and offer recommendations for bias adjustment. We specifically evaluate precipitation occurrence (number of precipitation days over a period) and total precipitation accumulation (total accumulation of liquid water from precipitation days over a period). ERA5 exhibits a precipitation occurrence bias, generating excess precipitation inconsistent with observations. Literature to date removes daily amounts below 1 mm/day which account for almost 50% of annual precipitation occurrence in the region, motivating the need for new adjustment methods. A threshold of 0.63 mm/day allows for a better match with observations for August (+9% bias) and annual (-2%) precipitation occurrence, but February is underestimated (-23%). A more elaborate adjustment based on precipitation temperature improves February precipitation occurrence (-2%). Adjusted ERA5 is unable to reproduce spatial variability. It cannot capture observed trends but shows some agreement with interannual differences (R2 = 0.32). The annual increase of 0.86 ± 0.14 days/year/year (1979-2006) is absent. Daily precipitation accumulation which is adjusted for drizzle bias reproduces annual (+7%), August (+19%) and February (-19%) precipitation accumulation. Accumulation adjustment methods were developed but should be calibrated on a recent subset of years (i.e., 10 years) since annual precipitation accumulation has increased. The spatial variability of precipitation accumulation is represented (R2 0.64) in addition to some detrended interannual variability (R2 = 0.34). The increase in precipitation accumulation of 1.50 ± 0.42 mm/year/year (1979-2006) is severely underestimated, questioning the reliability of ERA5 for investigating climate change in the Canadian Arctic. This study informs the Arctic research community of the validity of using ERA5. We evaluate and improve bias adjustment methods for precipitation occurrence and accumulation, enhancing ERA5’s representation of climatological mean precipitation in the region.

Text

2025-06-30

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/88546

Geophysics

University of British Columbia

UBCV

http://creativecommons.org/licenses/by-nc-nd/4.0/