Evaluation of Reanalysis Temperature and Precipitation for the Andean Altiplano and Adjacent Cordilleras

This study compares temperature, precipitation, and other climate variables from six widely used climate reanalysis products to inform ice‐core climate proxy record calibration in the Altiplano region of the central Andes. The reanalyzes are the European Reanalysis version 5 (ERA5), European Reanalysis Interim, Modern‐Era Retrospective analysis for Research and Applications (MERRA2), Japanese 55‐year Reanalysis, Climate Forecast System Reanalysis and version 2 extension, and NCEP/NCAR Reanalysis version 1. These data products are validated against observations from automatic weather stations on the Quelccaya Ice Cap, Peru (5,650 m a.s.l) and Chacaltaya, Bolivia (5,238 m a.s.l), in addition to lower sites ranging in elevation 2,500–4,900 m a.s.l. Our results suggest that ERA5 provides the most robust overall depiction of temperature and precipitation across the study domain, and the data set is particularly useful for its back‐extension to 1950. However, MERRA2 produces lower precipitation error scores owing to a gaged‐based bias correction. An examination of ERA5 vertical atmospheric profiles for a latitudinal transect over Quelccaya shows considerable variability, including across major El Niño events, suggesting the need for caution when interpreting isotopic signatures in ice cores. Key Points European Reanalysis version 5 (ERA5) affords a robust overall climate depiction for a domain spanning the central Andes in comparison to other reanalyzes Modern‐Era Retrospective analysis for Research and Applications version 2 produces lower error scores than ERA5 against precipitation observations, but improvement is due to a gauge‐based bias correction ERA5 vertical profiles over Quelccaya for major El Niños suggest need for caution interpreting El Niño Southern Oscillation signatures in ice cores