On the reliability of global seasonal forecasts: sensitivity to ensemble size, hindcast length and region definition

This work is protected by the corresponding intellectual and industrial property rights. Except where otherwise noted, its contents are licensed under a Creative Commons license : Attribution-NonCommercial 4.0 International

ProjectAfriCultuReS - Enhancing Food Security in AFRIcan AgriCULTUral Systems with the Support of REmote Sensing (EC-H2020-774652)
FOCUS-Africa - Full-value chain Optimised Climate User-centric Services for Southern Africa: FOCUS-Africa (EC-H2020-869575)

Abstract

One of the key quality aspects in a probabilistic prediction is its reliability. However, this property is difficult to estimate in the case of seasonal forecasts due to the limited size of most of the hindcasts that are available nowadays. To shed light on this issue, this work presents a detailed analysis of how the ensemble size, the hindcast length and the number of points pooled together within a particular region affect the resulting reliability estimates. To do so, we build on 42 land reference regions recently defined for the IPCC-AR6 and assess the reliability of global seasonal forecasts of temperature and precipitation from the European Center for Medium Weather Forecasts SEAS5 prediction system, which is compared against its predecessor, System4. Our results indicate that whereas longer hindcasts and larger ensembles lead to increased reliability estimates, the number of points that are pooled together within a homogeneous climate region is much less relevant.

DatasetData Availability Statement SEAS5 data was retrieved from the MARS archive (https://confluence.ecmwf.int/display/COPSRV/MARS+archive) following the ECMWF data policy. CRU TS v4.04 was downloaded from https://catalogue.ceda.ac.uk/uuid/89e1e34ec3554dc98594a5732622bce9. Differently, System4 and EWEMBI were obtained from the User Data Gateway (UDG), a THREDDS-based service from the Santander Climate Data Service that provides access to a wide catalogue of popular climate datasets: http://meteo.unican.es/udg-tap/home. See Cofiño et al. (2018) for further information. Finally, reliability categories can be computed (and plotted) with the function reliabilityCategories included in the visualizeR package (Frías et al., 2018), which forms part of climate4R (Iturbide et al., 2019), a bundle of R packages developed by the Santander Meteorology Group for transparent climate data access, post-processing and visualization.

CitationManzanas, R. [et al.]. On the reliability of global seasonal forecasts: sensitivity to ensemble size, hindcast length and region definition. "Geophysical Research Letters", Setembre 2022, vol. 49, núm. 17, e2021GL094662.

URIhttp://hdl.handle.net/2117/373553

DOI10.1029/2021GL094662

ISSN0094-8276
1944-8007

Publisher versionhttps://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021GL094662

Collections