Using Explainable Machine Learning Forecasts to Discover Subseasonal Drivers of High Summer Temperatures in Western and Central Europe
Reliable subseasonal forecasts of high summer temperatures would be very valuable for society. Although state-of-the-art numerical weather prediction (NWP) models have become much better in representing the relevant sources of predictability like land and sea surface states, the subseasonal potentia...
Published in: | Monthly Weather Review |
---|---|
Main Authors: | , , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
2022
|
Subjects: | |
Online Access: | https://research.vu.nl/en/publications/5c896f4b-8725-4d9e-9071-71726a039e77 https://doi.org/10.1175/MWR-D-21-0201.1 https://hdl.handle.net/1871.1/5c896f4b-8725-4d9e-9071-71726a039e77 |
id |
ftvuamstcris:oai:research.vu.nl:publications/5c896f4b-8725-4d9e-9071-71726a039e77 |
---|---|
record_format |
openpolar |
spelling |
ftvuamstcris:oai:research.vu.nl:publications/5c896f4b-8725-4d9e-9071-71726a039e77 2024-09-15T18:35:35+00:00 Using Explainable Machine Learning Forecasts to Discover Subseasonal Drivers of High Summer Temperatures in Western and Central Europe van Straaten, Chiem Whan, Kirien Coumou, Dim van den Hurk, Bart Schmeits, Maurice 2022-05-20 https://research.vu.nl/en/publications/5c896f4b-8725-4d9e-9071-71726a039e77 https://doi.org/10.1175/MWR-D-21-0201.1 https://hdl.handle.net/1871.1/5c896f4b-8725-4d9e-9071-71726a039e77 eng eng https://research.vu.nl/en/publications/5c896f4b-8725-4d9e-9071-71726a039e77 info:eu-repo/semantics/openAccess van Straaten , C , Whan , K , Coumou , D , van den Hurk , B & Schmeits , M 2022 , ' Using Explainable Machine Learning Forecasts to Discover Subseasonal Drivers of High Summer Temperatures in Western and Central Europe ' , Monthly Weather Review , vol. 150 , no. 5 , pp. 1115–1134 . https://doi.org/10.1175/MWR-D-21-0201.1 Statistical forecasting Subseasonal variability Machine learning Model interpretation and visualization /dk/atira/pure/sustainabledevelopmentgoals/life_below_water name=SDG 14 - Life Below Water article 2022 ftvuamstcris https://doi.org/10.1175/MWR-D-21-0201.1 2024-09-05T00:23:23Z Reliable subseasonal forecasts of high summer temperatures would be very valuable for society. Although state-of-the-art numerical weather prediction (NWP) models have become much better in representing the relevant sources of predictability like land and sea surface states, the subseasonal potential is not fully realized. Complexities arise because drivers depend on the state of other drivers and on interactions over multiple time scales. This study applies statistical modeling to ERA5 data, and explores how nine potential drivers, interacting on eight time scales, contribute to the subseasonal predictability of high summer temperatures in western and central Europe. Features and target temperatures are extracted with two variations of hierarchical clustering, and are fitted with a machine learning (ML) model based on random forests. Explainable AI methods show that the ML model agrees with physical understanding. Verification of the forecasts reveals that a large part of predictability comes from climate change, but that reliable and valuable subseasonal forecasts are possible in certain windows, like forecasting monthly warm anomalies with a lead time of 15 days. Contributions of each driver confirm that there is a transfer of predictability from the land and sea surface state to the atmosphere. The involved time scales depend on lead time and the forecast target. The explainable AI methods also reveal surprising driving features in sea surface temperature and 850 hPa temperature, and rank the contribution of snow cover above that of sea ice. Overall, this study demonstrates that complex statistical models, when made explainable, can complement research with NWP models, by diagnosing drivers that need further understanding and a correct numerical representation, for better future forecasts. Article in Journal/Newspaper Sea ice Vrije Universiteit Amsterdam (VU): Research Portal Monthly Weather Review 150 5 1115 1134 |
institution |
Open Polar |
collection |
Vrije Universiteit Amsterdam (VU): Research Portal |
op_collection_id |
ftvuamstcris |
language |
English |
topic |
Statistical forecasting Subseasonal variability Machine learning Model interpretation and visualization /dk/atira/pure/sustainabledevelopmentgoals/life_below_water name=SDG 14 - Life Below Water |
spellingShingle |
Statistical forecasting Subseasonal variability Machine learning Model interpretation and visualization /dk/atira/pure/sustainabledevelopmentgoals/life_below_water name=SDG 14 - Life Below Water van Straaten, Chiem Whan, Kirien Coumou, Dim van den Hurk, Bart Schmeits, Maurice Using Explainable Machine Learning Forecasts to Discover Subseasonal Drivers of High Summer Temperatures in Western and Central Europe |
topic_facet |
Statistical forecasting Subseasonal variability Machine learning Model interpretation and visualization /dk/atira/pure/sustainabledevelopmentgoals/life_below_water name=SDG 14 - Life Below Water |
description |
Reliable subseasonal forecasts of high summer temperatures would be very valuable for society. Although state-of-the-art numerical weather prediction (NWP) models have become much better in representing the relevant sources of predictability like land and sea surface states, the subseasonal potential is not fully realized. Complexities arise because drivers depend on the state of other drivers and on interactions over multiple time scales. This study applies statistical modeling to ERA5 data, and explores how nine potential drivers, interacting on eight time scales, contribute to the subseasonal predictability of high summer temperatures in western and central Europe. Features and target temperatures are extracted with two variations of hierarchical clustering, and are fitted with a machine learning (ML) model based on random forests. Explainable AI methods show that the ML model agrees with physical understanding. Verification of the forecasts reveals that a large part of predictability comes from climate change, but that reliable and valuable subseasonal forecasts are possible in certain windows, like forecasting monthly warm anomalies with a lead time of 15 days. Contributions of each driver confirm that there is a transfer of predictability from the land and sea surface state to the atmosphere. The involved time scales depend on lead time and the forecast target. The explainable AI methods also reveal surprising driving features in sea surface temperature and 850 hPa temperature, and rank the contribution of snow cover above that of sea ice. Overall, this study demonstrates that complex statistical models, when made explainable, can complement research with NWP models, by diagnosing drivers that need further understanding and a correct numerical representation, for better future forecasts. |
format |
Article in Journal/Newspaper |
author |
van Straaten, Chiem Whan, Kirien Coumou, Dim van den Hurk, Bart Schmeits, Maurice |
author_facet |
van Straaten, Chiem Whan, Kirien Coumou, Dim van den Hurk, Bart Schmeits, Maurice |
author_sort |
van Straaten, Chiem |
title |
Using Explainable Machine Learning Forecasts to Discover Subseasonal Drivers of High Summer Temperatures in Western and Central Europe |
title_short |
Using Explainable Machine Learning Forecasts to Discover Subseasonal Drivers of High Summer Temperatures in Western and Central Europe |
title_full |
Using Explainable Machine Learning Forecasts to Discover Subseasonal Drivers of High Summer Temperatures in Western and Central Europe |
title_fullStr |
Using Explainable Machine Learning Forecasts to Discover Subseasonal Drivers of High Summer Temperatures in Western and Central Europe |
title_full_unstemmed |
Using Explainable Machine Learning Forecasts to Discover Subseasonal Drivers of High Summer Temperatures in Western and Central Europe |
title_sort |
using explainable machine learning forecasts to discover subseasonal drivers of high summer temperatures in western and central europe |
publishDate |
2022 |
url |
https://research.vu.nl/en/publications/5c896f4b-8725-4d9e-9071-71726a039e77 https://doi.org/10.1175/MWR-D-21-0201.1 https://hdl.handle.net/1871.1/5c896f4b-8725-4d9e-9071-71726a039e77 |
genre |
Sea ice |
genre_facet |
Sea ice |
op_source |
van Straaten , C , Whan , K , Coumou , D , van den Hurk , B & Schmeits , M 2022 , ' Using Explainable Machine Learning Forecasts to Discover Subseasonal Drivers of High Summer Temperatures in Western and Central Europe ' , Monthly Weather Review , vol. 150 , no. 5 , pp. 1115–1134 . https://doi.org/10.1175/MWR-D-21-0201.1 |
op_relation |
https://research.vu.nl/en/publications/5c896f4b-8725-4d9e-9071-71726a039e77 |
op_rights |
info:eu-repo/semantics/openAccess |
op_doi |
https://doi.org/10.1175/MWR-D-21-0201.1 |
container_title |
Monthly Weather Review |
container_volume |
150 |
container_issue |
5 |
container_start_page |
1115 |
op_container_end_page |
1134 |
_version_ |
1810478769921589248 |