Probabilistic precipitation downscaling for ungauged mountain sites: a pilot study for the Hindu Kush Himalaya

Girona-Mata, Marc ORCID: https://orcid.org/0000-0001-7773-0333; Orr, Andrew ORCID: https://orcid.org/0000-0001-5111-8402; Widmann, Martin ORCID: https://orcid.org/0000-0001-5447-5763; Bannister, Daniel ORCID: https://orcid.org/0000-0002-2982-3751; Dars, Ghulam Hussain ORCID: https://orcid.org/0000-0002-6986-1461; Hosking, Scott ORCID: https://orcid.org/0000-0002-3646-3504; Norris, Jesse; Ocio, David ORCID: https://orcid.org/0000-0002-4900-4606; Phillips, Tony ORCID: https://orcid.org/0000-0002-3058-9157; Steiner, Jakob ORCID: https://orcid.org/0000-0002-0063-0067; Turner, Richard E.. 2025 Probabilistic precipitation downscaling for ungauged mountain sites: a pilot study for the Hindu Kush Himalaya. Hydrology and Earth System Sciences, 29 (14). 3073-3100. 10.5194/hess-29-3073-2025

Before downloading, please read NORA policies.

Preview

Text (Open Access) © Author(s) 2025. hess-29-3073-2025.pdf - Published Version Available under License Creative Commons Attribution 4.0. Download (7MB) | Preview

Abstract/Summary

This study introduces a novel approach to post-processing (i.e. downscaling and bias-correcting) reanalysis-driven regional climate model daily precipitation outputs that can be generalised to ungauged mountain locations by leveraging sparse in situ observations and a probabilistic regression framework. We call this post-processing approach generalised probabilistic regression (GPR) and implement it using both generalised linear models and artificial neural networks (i.e. multi-layer perceptrons). By testing the GPR post-processing approach across three Hindu Kush Himalaya (HKH) basins with varying hydro-meteorological characteristics and four experiments, which are representative of real-world scenarios, we find it performs consistently much better than both raw regional climate model output and deterministic bias correction methods for generalising daily precipitation post-processing to ungauged locations. We also find that GPR models are flexible and can be trained using data from a single region or multiple regions combined together, without major impacts on model performance. Additionally, we show that the GPR approach results in superior skill for post-processing entirely ungauged regions, by leveraging data from other regions as well as ungauged high-elevation ranges. This suggests that GPR models have potential for extending post-processing of daily precipitation to ungauged areas of HKH. Whilst multi-layer perceptrons yield marginally improved results overall, generalised linear models are a robust choice, particularly for data-scarce scenarios, i.e. post-processing extreme precipitation events and generalising to completely ungauged regions.

Item Type:	Publication - Article
Digital Object Identifier (DOI):	10.5194/hess-29-3073-2025
ISSN:	10275606
Date made live:	12 Aug 2025 13:22 +0 (UTC)
URI:	https://nora.nerc.ac.uk/id/eprint/538909

Actions (login required)

View Item

Document Downloads

More statistics for this item...