Titre

Predicting years with extremely low gross primary production from daily weather data using Convolutional Neural Networks

Type
article
Institution
UNIL/CHUV/Unisanté + institutions partenaires
Périodique
Auteur(s)
Marcolongo, Aris
Auteure/Auteur
Vladymyrov, Mykhailo
Auteure/Auteur
Lienert, Sebastian
Auteure/Auteur
Peleg, Nadav
Auteure/Auteur
Haug, Sigve
Auteure/Auteur
Zscheischler, Jakob
Auteure/Auteur
Liens vers les personnes
Liens vers les unités
ISSN
2634-4602
Statut éditorial
Publié
Date de publication
2022
Volume
1
Peer-reviewed
Oui
Langue
anglais
Résumé
Understanding the meteorological drivers of extreme impacts in social or environmental systems is important to better quantify current and project future climate risks. Impacts are typically an aggregated response to many different interacting drivers at various temporal scales, rendering such driver identification a challenging task. Machine learning–based approaches, such as deep neural networks, may be able to address this task but require large training datasets. Here, we explore the ability of Convolutional Neural Networks (CNNs) to predict years with extremely low gross primary production (GPP) from daily weather data in three different vegetation types. To circumvent data limitations in observations, we simulate 100,000 years of daily weather with a weather generator for three different geographical sites and subsequently simulate vegetation dynamics with a complex vegetation model. For each resulting vegetation distribution, we then train two different CNNs to classify daily weather data (temperature, precipitation, and radiation) into years with extremely low GPP and normal years. Overall, prediction accuracy is very good if the monthly or yearly GPP values are used as an intermediate training target (area under the precision-recall curve AUC 0.9). The best prediction accuracy is found in tropical forests, with temperate grasslands and boreal forests leading to comparable results. Prediction accuracy is strongly reduced when binary classification is used directly. Furthermore, using daily GPP during training does not improve the predictive power. We conclude that CNNs are able to predict extreme impacts from complex meteorological drivers if sufficient data are available.
Sujets

Carbon cycle

Convolutional Neural ...

extreme events

extreme impacts

PID Serval
serval:BIB_5DFA6967F996
DOI
Permalien
Open Access
Oui
Date de création
2022-04-15T06:40:24.023Z
Date de création dans IRIS
2025-05-20T20:02:22Z
Fichier(s)
En cours de chargement...
Vignette d'image
Nom

Marcolongo 2022.pdf

Version du manuscrit

published

Licence

https://creativecommons.org/licenses/by/4.0

Taille

2.6 MB

Format

Adobe PDF

PID Serval

serval:BIB_5DFA6967F996.P001

URN

urn:nbn:ch:serval-BIB_5DFA6967F9965

Somme de contrôle

(MD5):db09109ca65ffdf8bc4fc1cf0d286b5a