Optimizing Seasonal‐To‐Decadal Analog Forecasts With a Learned Spatially‐Weighted Mask
Abstract Seasonal‐to‐decadal climate prediction is crucial for decision‐making in a number of industries, but forecasts on these timescales have limited skill. Here, we develop a data‐driven method for selecting optimal analogs for seasonal‐to‐decadal analog forecasting. Using an interpretable neura...
Published in: | Geophysical Research Letters |
---|---|
Main Authors: | , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
Wiley
2023
|
Subjects: | |
Online Access: | https://doi.org/10.1029/2023GL104983 https://doaj.org/article/dea19bb40b7247f685e88cf4ccabb771 |
id |
ftdoajarticles:oai:doaj.org/article:dea19bb40b7247f685e88cf4ccabb771 |
---|---|
record_format |
openpolar |
spelling |
ftdoajarticles:oai:doaj.org/article:dea19bb40b7247f685e88cf4ccabb771 2024-09-09T19:55:52+00:00 Optimizing Seasonal‐To‐Decadal Analog Forecasts With a Learned Spatially‐Weighted Mask Jamin K. Rader Elizabeth A. Barnes 2023-12-01T00:00:00Z https://doi.org/10.1029/2023GL104983 https://doaj.org/article/dea19bb40b7247f685e88cf4ccabb771 EN eng Wiley https://doi.org/10.1029/2023GL104983 https://doaj.org/toc/0094-8276 https://doaj.org/toc/1944-8007 1944-8007 0094-8276 doi:10.1029/2023GL104983 https://doaj.org/article/dea19bb40b7247f685e88cf4ccabb771 Geophysical Research Letters, Vol 50, Iss 23, Pp n/a-n/a (2023) interpretable machine learning analog forecasting seasonal‐to‐decadal climate prediction El Niño Southern Oscillation North Atlantic sources of predictability Geophysics. Cosmic physics QC801-809 article 2023 ftdoajarticles https://doi.org/10.1029/2023GL104983 2024-08-05T17:49:23Z Abstract Seasonal‐to‐decadal climate prediction is crucial for decision‐making in a number of industries, but forecasts on these timescales have limited skill. Here, we develop a data‐driven method for selecting optimal analogs for seasonal‐to‐decadal analog forecasting. Using an interpretable neural network, we learn a spatially‐weighted mask that quantifies how important each grid point is for determining whether two climate states will evolve similarly. We show that analogs selected using this weighted mask provide more skillful forecasts than analogs that are selected using traditional spatially‐uniform methods. This method is tested on two prediction problems using the Max Planck Institute for Meteorology Grand Ensemble: multi‐year prediction of North Atlantic sea surface temperatures, and seasonal prediction of El Niño Southern Oscillation. This work demonstrates a methodical approach to selecting analogs that may be useful for improving seasonal‐to‐decadal forecasts and understanding their sources of skill. Article in Journal/Newspaper North Atlantic Directory of Open Access Journals: DOAJ Articles Geophysical Research Letters 50 23 |
institution |
Open Polar |
collection |
Directory of Open Access Journals: DOAJ Articles |
op_collection_id |
ftdoajarticles |
language |
English |
topic |
interpretable machine learning analog forecasting seasonal‐to‐decadal climate prediction El Niño Southern Oscillation North Atlantic sources of predictability Geophysics. Cosmic physics QC801-809 |
spellingShingle |
interpretable machine learning analog forecasting seasonal‐to‐decadal climate prediction El Niño Southern Oscillation North Atlantic sources of predictability Geophysics. Cosmic physics QC801-809 Jamin K. Rader Elizabeth A. Barnes Optimizing Seasonal‐To‐Decadal Analog Forecasts With a Learned Spatially‐Weighted Mask |
topic_facet |
interpretable machine learning analog forecasting seasonal‐to‐decadal climate prediction El Niño Southern Oscillation North Atlantic sources of predictability Geophysics. Cosmic physics QC801-809 |
description |
Abstract Seasonal‐to‐decadal climate prediction is crucial for decision‐making in a number of industries, but forecasts on these timescales have limited skill. Here, we develop a data‐driven method for selecting optimal analogs for seasonal‐to‐decadal analog forecasting. Using an interpretable neural network, we learn a spatially‐weighted mask that quantifies how important each grid point is for determining whether two climate states will evolve similarly. We show that analogs selected using this weighted mask provide more skillful forecasts than analogs that are selected using traditional spatially‐uniform methods. This method is tested on two prediction problems using the Max Planck Institute for Meteorology Grand Ensemble: multi‐year prediction of North Atlantic sea surface temperatures, and seasonal prediction of El Niño Southern Oscillation. This work demonstrates a methodical approach to selecting analogs that may be useful for improving seasonal‐to‐decadal forecasts and understanding their sources of skill. |
format |
Article in Journal/Newspaper |
author |
Jamin K. Rader Elizabeth A. Barnes |
author_facet |
Jamin K. Rader Elizabeth A. Barnes |
author_sort |
Jamin K. Rader |
title |
Optimizing Seasonal‐To‐Decadal Analog Forecasts With a Learned Spatially‐Weighted Mask |
title_short |
Optimizing Seasonal‐To‐Decadal Analog Forecasts With a Learned Spatially‐Weighted Mask |
title_full |
Optimizing Seasonal‐To‐Decadal Analog Forecasts With a Learned Spatially‐Weighted Mask |
title_fullStr |
Optimizing Seasonal‐To‐Decadal Analog Forecasts With a Learned Spatially‐Weighted Mask |
title_full_unstemmed |
Optimizing Seasonal‐To‐Decadal Analog Forecasts With a Learned Spatially‐Weighted Mask |
title_sort |
optimizing seasonal‐to‐decadal analog forecasts with a learned spatially‐weighted mask |
publisher |
Wiley |
publishDate |
2023 |
url |
https://doi.org/10.1029/2023GL104983 https://doaj.org/article/dea19bb40b7247f685e88cf4ccabb771 |
genre |
North Atlantic |
genre_facet |
North Atlantic |
op_source |
Geophysical Research Letters, Vol 50, Iss 23, Pp n/a-n/a (2023) |
op_relation |
https://doi.org/10.1029/2023GL104983 https://doaj.org/toc/0094-8276 https://doaj.org/toc/1944-8007 1944-8007 0094-8276 doi:10.1029/2023GL104983 https://doaj.org/article/dea19bb40b7247f685e88cf4ccabb771 |
op_doi |
https://doi.org/10.1029/2023GL104983 |
container_title |
Geophysical Research Letters |
container_volume |
50 |
container_issue |
23 |
_version_ |
1809926131564937216 |