Deep Learning based Extreme Heatwave Forecast
- URL: http://arxiv.org/abs/2103.09743v1
- Date: Wed, 17 Mar 2021 16:10:06 GMT
- Title: Deep Learning based Extreme Heatwave Forecast
- Authors: Val\'erian Jacques-Dumas, Francesco Ragone, Freddy Bouchet, Pierre
Borgnat, Patrice Abry
- Abstract summary: Using 1000 years of state-of-the-art PlaSim Planete Simulator Climate Model data, it is shown that Convolutional Neural Network-based Deep Learning frameworks, with large-class undersampling and transfer learning achieve significant performance in forecasting the occurrence of extreme heatwaves.
- Score: 8.975667614727648
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Forecasting the occurrence of heatwaves constitutes a challenging issue, yet
of major societal stake, because extreme events are not often observed and
(very) costly to simulate from physics-driven numerical models. The present
work aims to explore the use of Deep Learning architectures as alternative
strategies to predict extreme heatwaves occurrences from a very limited amount
of available relevant climate data. This implies addressing issues such as the
aggregation of climate data of different natures, the class-size imbalance that
is intrinsically associated with rare event prediction, and the potential
benefits of transfer learning to address the nested nature of extreme events
(naturally included in less extreme ones). Using 1000 years of state-of-the-art
PlaSim Planete Simulator Climate Model data, it is shown that Convolutional
Neural Network-based Deep Learning frameworks, with large-class undersampling
and transfer learning achieve significant performance in forecasting the
occurrence of extreme heatwaves, at three different levels of intensity, and as
early as 15 days in advance from the restricted observation, for a single time
(single snapshoot) of only two spatial fields of climate data, surface
temperature and geopotential height.
Related papers
- Evaluation of deep learning models for Australian climate extremes: prediction of streamflow and floods [0.17999333451993949]
In recent years, climate extremes such as floods have created significant environmental and economic hazards for Australia.
Deep learning methods have been promising for predicting small to medium-sized climate extreme events over a short time horizon.
We present an ensemble-based machine learning approach that addresses large-scale extreme flooding challenges.
arXiv Detail & Related papers (2024-07-20T23:45:04Z) - ClimODE: Climate and Weather Forecasting with Physics-informed Neural ODEs [14.095897879222676]
We present ClimODE, a continuous-time process that implements key principle of statistical mechanics.
ClimODE models precise weather evolution with value-conserving dynamics, learning global weather transport as a neural flow.
Our approach outperforms existing data-driven methods in global, regional forecasting with an order of magnitude smaller parameterization.
arXiv Detail & Related papers (2024-04-15T06:38:21Z) - Comparing Data-Driven and Mechanistic Models for Predicting Phenology in
Deciduous Broadleaf Forests [47.285748922842444]
We train a deep neural network to predict a phenological index from meteorological time series.
We find that this approach outperforms traditional process-based models.
arXiv Detail & Related papers (2024-01-08T15:29:23Z) - Climate Intervention Analysis using AI Model Guided by Statistical
Physics Principles [6.824166358727082]
We propose a novel solution by utilizing a principle from statistical physics known as the Fluctuation-Dissipation Theorem (FDT)
By leveraging, we are able to extract information encoded in a large dataset produced by Earth System Models.
Our model, AiBEDO, is capable of capturing the complex, multi-timescale effects of radiation perturbations on global and regional surface climate.
arXiv Detail & Related papers (2023-02-07T05:09:10Z) - ClimaX: A foundation model for weather and climate [51.208269971019504]
ClimaX is a deep learning model for weather and climate science.
It can be pre-trained with a self-supervised learning objective on climate datasets.
It can be fine-tuned to address a breadth of climate and weather tasks.
arXiv Detail & Related papers (2023-01-24T23:19:01Z) - Multi-scale Digital Twin: Developing a fast and physics-informed
surrogate model for groundwater contamination with uncertain climate models [53.44486283038738]
Climate change exacerbates the long-term soil management problem of groundwater contamination.
We develop a physics-informed machine learning surrogate model using U-Net enhanced Fourier Neural Contaminated (PDENO)
In parallel, we develop a convolutional autoencoder combined with climate data to reduce the dimensionality of climatic region similarities across the United States.
arXiv Detail & Related papers (2022-11-20T06:46:35Z) - Spatiotemporal modeling of European paleoclimate using doubly sparse
Gaussian processes [61.31361524229248]
We build on recent scale sparsetemporal GPs to reduce the computational burden.
We successfully employ such a doubly sparse GP to construct a probabilistic model of paleoclimate.
arXiv Detail & Related papers (2022-11-15T14:15:04Z) - Probabilistic forecasts of extreme heatwaves using convolutional neural
networks in a regime of lack of data [6.972317847755389]
We develop a methodology to build forecasting models for extreme heatwaves.
These models are based on convolutional neural networks, trained on extremely long 8,000-year climate model outputs.
We demonstrate that deep neural networks are suitable for this purpose for long lasting 14-day heatwaves over France.
arXiv Detail & Related papers (2022-08-01T16:19:38Z) - Forecasting large-scale circulation regimes using deformable
convolutional neural networks and global spatiotemporal climate data [86.1450118623908]
We investigate a supervised machine learning approach based on deformable convolutional neural networks (deCNNs)
We forecast the North Atlantic-European weather regimes during extended boreal winter for 1 to 15 days into the future.
Due to its wider field of view, we also observe deCNN achieving considerably better performance than regular convolutional neural networks at lead times beyond 5-6 days.
arXiv Detail & Related papers (2022-02-10T11:37:00Z) - DeepClimGAN: A High-Resolution Climate Data Generator [60.59639064716545]
Earth system models (ESMs) are often used to generate future projections of climate change scenarios.
As a compromise, emulators are substantially less expensive but may not have all of the complexity of an ESM.
Here we demonstrate the use of a conditional generative adversarial network (GAN) to act as an ESM emulator.
arXiv Detail & Related papers (2020-11-23T20:13:37Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.