Adjustment of global gridded precipitation for systematic bias

TLDR

Gauge‐based precipitation measurements suffer substantial systematic biases, especially from wind‐induced undercatch of solid precipitation. The study develops a methodology to produce gridded mean monthly catch ratios for adjusting wind‐induced undercatch and wetting losses in global precipitation products. The adjustments were estimated using gauge type‐specific regression equations from the WMO Solid Precipitation Measurement Intercomparison, liquid precipitation and wetting losses estimated with prior global bias adjustment methods, and Canadian adjustments derived from more detailed network data, producing gridded gauge adjustment products applicable to climatology and individual years 1979–1998. Applying the CRs increased mean annual global terrestrial precipitation by 11.7%, with wind‐induced undercatch estimates 1.6–7.9% higher than recent localized studies, and producing slightly larger warm‑season and smaller cold‑season precipitation increases, especially over North America, compared to previous global adjustments.

Abstract

Systematic biases in gauge‐based measurement of precipitation can be substantial. Of the sources of bias, wind‐induced undercatch of solid precipitation is by far the largest. A methodology for producing gridded mean monthly catch ratios (CRs) for the adjustment of wind‐induced undercatch and wetting losses is developed, which is suitable for application to continental or global gridded precipitation products. The adjustments for wind‐induced solid precipitation were estimated using gauge type‐specific regression equations from the recent World Meteorological Organization Solid Precipitation Measurement Intercomparison. Wind‐induced undercatch of liquid precipitation and wetting losses were estimated using methods employed in previous global bias adjustment efforts. Due to the unique nature of Canada's precipitation measurement network, the Canadian adjustments were determined using more detailed information than for the rest of the domain, and are therefore expected to be more reliable. The gridded gauge adjustment products are designed to be applicable both to climatological estimates and to individual years during the 1979 through 1998 reference period. Application of the CRs to an existing precipitation product yielded an increase in mean annual global terrestrial precipitation of 11.7%. As compared with recent (but more localized) studies that used a similar method to account for wind‐induced catch deficiencies, our estimates of wind‐induced undercatch are 1.6–7.9% higher on a mean annual basis. Compared to a previous global precipitation bias adjustment effort, our adjusted data set results on average in slightly greater warm season and lower cold season precipitation increases, greater precipitation increases over North America, and lower precipitation increases over Eurasia.