Convolutional conditional neural processes for local climate downscaling

Abstract A new model is presented for multisite statistical downscaling of temperature and precipitation using convolutional conditional neural processes (convCNPs). ConvCNPs are a recently developed class of models that allow deep-learning techniques to be applied to off-the-grid spatio-temporal data. In contrast to existing methods that map from low-resolution model output to high-resolution predictions at a discrete set of locations, this model outputs a stochastic process that can be queried at an arbitrary latitude–longitude coordinate. The convCNP model is shown to outperform an ensemble of existing downscaling techniques over Europe for both temperature and precipitation taken from the VALUE intercomparison project. The model also outperforms an approach that uses Gaussian processes to interpolate single-site downscaling models at unseen locations. Importantly, substantial improvement is seen in the representation of extreme precipitation events. These results indicate that the convCNP is a robust downscaling model suitable for generating localised projections for use in climate impact studies.

Location
Deutsche Nationalbibliothek Frankfurt am Main
Extent
Online-Ressource
Language
Englisch

Bibliographic citation
Convolutional conditional neural processes for local climate downscaling ; volume:15 ; number:1 ; year:2022 ; pages:251-268 ; extent:18
Geoscientific model development ; 15, Heft 1 (2022), 251-268 (gesamt 18)

Creator
Vaughan, Anna
Tebbutt, Will
Hosking, J. Scott
Turner, Richard E.

DOI
10.5194/gmd-15-251-2022
URN
urn:nbn:de:101:1-2022012004280165579581
Rights
Open Access; Der Zugriff auf das Objekt ist unbeschränkt möglich.
Last update
15.08.2025, 7:20 AM CEST

Data provider

This object is provided by:
Deutsche Nationalbibliothek. If you have any questions about the object, please contact the data provider.

Associated

  • Vaughan, Anna
  • Tebbutt, Will
  • Hosking, J. Scott
  • Turner, Richard E.

Other Objects (12)