Evaluation of reanalysis datasets for studying high-elevation meteorological conditions in the Langtang Catchment, Nepal

Climate reanalysis datasets are widely used in hydro-meteorological and cryospheric research across High Mountain Asia (HMA), where in situ observations are scarce. Here, we evaluate ERA5L, High Asian Refined Analysis (HARv2), Modern-Era Retrospective Analysis for Research and Applications (MERRA2), and Climatologies at High Resolution for the Earth’s Land Surface Areas (CHELSA) against near-surface meteorological observations between 3,865 and 5,350 m a.s.l., in the upper Langtang Catchment, Nepal, using statistical metrics and Taylor Skill Score ( TSS ). We further assess their suitability for glacier surface energy balance (SEB) and mass balance modelling. No reanalysis product consistently outperforms the others; rather, their performance varies by meteorological variable, site, surface conditions, and timescale, with generally better performance at glacier sites, likely because their relatively homogeneous surface conditions are more readily represented by reanalysis products. Near-surface air temperature ( T 2m ) and humidity (specific: q and relative: RH ) are well reproduced ( TSS > 0.77) across products, although cold temperature and both dry and moist humidity biases are pervasive across sites. Wind speed ( u ) is poorly represented across products, sites, and timescales, reflecting their limited ability to resolve boundary layer processes and local katabatic and anabatic flows. Precipitation also exhibits large discrepancies, with ERA5L and HARv2 overestimating precipitation totals by ∼89–232% and MERRA2 and CHELSA underestimating by 30%–60%. They also generate frequent wet days and early-afternoon convective rainfall, in contrast to observed late-afternoon and nocturnal-to-early morning maxima. Hourly and daily SEB simulations at the Yala-on-Glacier site driven by reanalysis forcings yield statistically significant surface melt relative to reference estimates driven by in situ datasets; however, substantial differences in magnitude exist among products. ERA5L and HARv2 exhibit relatively small errors, largely due to compensating errors between radiative and turbulent fluxes, whereas MERRA2 and CHELSA show larger melt discrepancies due to their large radiative flux errors and underestimated precipitation that affects the surface albedo. Overall, our results underscore both the usefulness and limitations of reanalysis products for cryospheric applications in complex mountainous catchments and emphasize the need for assimilation of high-elevation observations in reanalysis products, alongside improved downscaling approaches, to better resolve local-scale atmospheric processes.

Authors

Institutions

Publication Details

Journal
Frontiers in Earth Science
Published
2026-11-19
DOI
https://doi.org/10.3389/feart.2026.1950767
Primary Topic
Cryospheric studies and observations
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Evaluation of reanalysis datasets for studying high-elevation meteorological conditions in the Langtang Catchment, Nepal

Gunjan Silwal, Bethan J. Davies, Owen King, Rachel Joanne Carr et al.
Frontiers in Earth Science
Cryospheric studies and observations
article

Evaluation of reanalysis datasets for studying high-elevation meteorological conditions in the Langtang Catchment, Nepal

Gunjan Silwal, Bethan J. Davies, Owen King, Rachel Joanne Carr, Sammie Buzzard
article en

Abstract

Climate reanalysis datasets are widely used in hydro-meteorological and cryospheric research across High Mountain Asia (HMA), where in situ observations are scarce. Here, we evaluate ERA5L, High Asian Refined Analysis (HARv2), Modern-Era Retrospective Analysis for Research and Applications (MERRA2), and Climatologies at High Resolution for the Earth’s Land Surface Areas (CHELSA) against near-surface meteorological observations between 3,865 and 5,350 m a.s.l., in the upper Langtang Catchment, Nepal, using statistical metrics and Taylor Skill Score ( TSS ). We further assess their suitability for glacier surface energy balance (SEB) and mass balance modelling. No reanalysis product consistently outperforms the others; rather, their performance varies by meteorological variable, site, surface conditions, and timescale, with generally better performance at glacier sites, likely because their relatively homogeneous surface conditions are more readily represented by reanalysis products. Near-surface air temperature ( T 2m ) and humidity (specific: q and relative: RH ) are well reproduced ( TSS > 0.77) across products, although cold temperature and both dry and moist humidity biases are pervasive across sites. Wind speed ( u ) is poorly represented across products, sites, and timescales, reflecting their limited ability to resolve boundary layer processes and local katabatic and anabatic flows. Precipitation also exhibits large discrepancies, with ERA5L and HARv2 overestimating precipitation totals by ∼89–232% and MERRA2 and CHELSA underestimating by 30%–60%. They also generate frequent wet days and early-afternoon convective rainfall, in contrast to observed late-afternoon and nocturnal-to-early morning maxima. Hourly and daily SEB simulations at the Yala-on-Glacier site driven by reanalysis forcings yield statistically significant surface melt relative to reference estimates driven by in situ datasets; however, substantial differences in magnitude exist among products. ERA5L and HARv2 exhibit relatively small errors, largely due to compensating errors between radiative and turbulent fluxes, whereas MERRA2 and CHELSA show larger melt discrepancies due to their large radiative flux errors and underestimated precipitation that affects the surface albedo. Overall, our results underscore both the usefulness and limitations of reanalysis products for cryospheric applications in complex mountainous catchments and emphasize the need for assimilation of high-elevation observations in reanalysis products, alongside improved downscaling approaches, to better resolve local-scale atmospheric processes.

Frontiers in Earth ScienceVol. 14
Northumbria University (GB), Centre for Polar Observation and Modelling (GB), Newcastle University (GB)
Cryospheric studies and observations
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon