CoChat
Sign Up
Cited by 50Climate variability and modelsMeteorological Phenomena and SimulationsForecasting Techniques and Applications

Using explainable machine learning forecasts to discover subseasonal drivers of high summer temperatures in Western and Central Europe

Open access
van Straaten, Chiem, Whan, Kirien, Coumou, Dim, van den Hurk, Bart, Schmeits, Maurice·2022·Monthly Weather Review
DOI 10.1175/mwr-d-21-0201.1
Open this paper in CoChat — free accountOpen PDF

Abstract

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.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1175/mwr-d-21-0201.1

Cites

104 of 106 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Decision-tree-based ensemble methods like RF and XGBoost are often chosen for their robustness against overfitting.

    “Decision-tree-based ensemble methods like RF and XGBoost (He et al., 2021;Kiefer et al., 2023;van Straaten et al., 2022;Weirich-Benet et al., 2023) are often chosen for their robustness against overfitting.”

    From Artificial intelligence for climate prediction of extremes: State of the art, challenges, and future perspectives (2024) · cites this paper as van Straaten et al., 2022

    The Machine learning model section says random forests are suited to a task with a low observation-to-feature ratio, where many statistical models risk overfitting. It also says the authors limited tree depth and minimum node size to avoid overfitting. The paper does not establish the same rationale for XGBoost or the broader claim that both methods are often chosen for this reason; pages are not available in our copy. (LLM, conf=0.86)

    Experimental — accuracy not yet measured

  • Little research has applied XAI to deep learning models in the context of extreme heat prediction.

    “While much work has been done to integrate XAI with machine learning models [26, 37], there has been little research into applying XAI to DL models, particularly in the context of extreme heat prediction.”

    From Extreme heat prediction through deep learning and explainable AI (2025) · cites this paper as [26]

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

1 retracted

Flagged first, then by citations

Research works printed but not linked: 2

  1. Dorrington, J., I. Finney, T. Palmer, and A. Weisheimer, 2020: Beyond skill scores: Exploring sub-seasonal forecast value through a case-study of French month-ahead energy predic- tion. Quart. J. Roy. Meteor. Soc., 146, 3623-3637, https://doi. org/10.1002/qj.3863. · DOI
  2. D. Coumou, B. van den Hurk, and M. Schmeits, 2020: The influence of aggregation and statistical post-processing on the subseasonal predictability of European temperatures. · DOI

Cited by

Sorted by citations

Related

Authors
van Straaten, C., Whan, K., Coumou, D., van den Hurk, B., & Schmeits, M.
Affiliations
  • a Royal Netherlands Meteorological Institute (KNMI), De Bilt, Netherlands
  • Royal Netherlands Meteorological Institute (NL)
  • d Deltares, Delft, Netherlands
ORCID
Venue
Monthly Weather Review
Publisher
American Meteorological Society
Published
2022-05
Volume
150
Issue
5
Pages
1115-1134
ISSN
  • 0027-0644 (ISSN-L)
  • 1520-0493
DOI
10.1175/mwr-d-21-0201.1
OpenAlex
W4212808191
Cited by
50
References
105
Citations by year
2026: 7 · 2025: 8 · 2024: 15 · 2023: 18 · 2022: 2
FWCI
4.28
Type
journal article
Language
en
Open access
Licence
Locations
  • Monthly Weather Review (publishedVersion, cc-by)
  • Publication Database PIK (Potsdam Institute for Climate Impact Research (PIK)) (submittedVersion, cc-by)
  • VU Research Portal (submittedVersion, other-oa)
  • Data Archiving and Networked Services (DANS) (submittedVersion, other-oa)
Funders
nederlandse organisatie voor wetenschappelijk onderzoek (ALWOP.395)
Topics
  • Physical Sciences › Environmental Science › Global and Planetary Change › Climate variability and models
  • Physical Sciences › Earth and Planetary Sciences › Atmospheric Science › Meteorological Phenomena and Simulations
  • Social Sciences › Decision Sciences › Management Science and Operations Research › Forecasting Techniques and Applications
Keywords
high summer temperatures, subseasonal predictability, explainable AI, random forest, hierarchical clustering, ERA5 reanalysis, climate change, sea surface temperature, snow cover, sea ice, 850 hPa temperature, Western and Central Europe

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review