Please use this identifier to cite or link to this item: http://hdl.handle.net/20.500.11765/13613
A critical view on the suitability of machine learning techniques to downscale climate change projections : illustration for temperature with a toy experiment
Title: A critical view on the suitability of machine learning techniques to downscale climate change projections : illustration for temperature with a toy experiment
Authors: Hernanz Lázaro, Alfonso ORCID Autor AEMETGarcía Valero, Juan Andrés ORCID RESEARCHERID Autor AEMETDomínguez Alonso, Marta ORCID Autor AEMETRodríguez Camino, Ernesto ORCID Autor AEMET
Keywords: Climate projections; Evaluation; Extrapolation; Neural networks; Machine learning; Statistical downscaling
Issue Date: 2022
Publisher: Royal Meteorological Society; Wiley
Citation: Atmospheric Science Letters. 2022, e1087
Publisher version: https://doi.org/10.1002/asl.1087
Abstract: Machine learning is a growing field of research with many applications. It provides a series of techniques able to solve complex nonlinear problems, and that has promoted their application for statistical downscaling. Intercomparison exercises with other classical methods have so far shown promising results. Nevertheless, many evaluation studies of statistical downscaling methods neglect the analysis of their extrapolation capability. In this study, we aim to make a wakeup call to the community about the potential risks of using machine learning for statistical downscaling of climate change projections. We present a set of three toy experiments, applying three commonly used machine learning algorithms, two different implementations of artificial neural networks and a support vector machine, to downscale daily maximum temperature, and comparing them with the classical multiple linear regression. We have tested the four methods in and out of their calibration range, and have found how the three machine learning techniques can perform poorly under extrapolation. Additionally, we have analysed the impact of this extrapolation issue depending on the degree of overlapping between the training and testing datasets, and we have found very different sensitivities for each method and specific implementation.
URI: http://hdl.handle.net/20.500.11765/13613
ISSN: 1530-261X
Appears in Collections:Artículos científicos 2019-2022


Files in This Item:
  File Description SizeFormat 
ASL2022_Hernanz.pdf
1,59 MBAdobe PDFThumbnail
View/Open
Show full item record



Items in Arcimís are protected by Creative Commons License, unless otherwise indicated.

Arcimís Repository
Nota Legal Contacto y sugerencias