Hard-Constrained Deep Learning for Climate Downscaling
- URL: http://arxiv.org/abs/2208.05424v9
- Date: Thu, 29 Feb 2024 20:31:03 GMT
- Title: Hard-Constrained Deep Learning for Climate Downscaling
- Authors: Paula Harder, Alex Hernandez-Garcia, Venkatesh Ramesh, Qidong Yang,
Prasanna Sattigeri, Daniela Szwarcman, Campbell Watson, David Rolnick
- Abstract summary: High-resolution climate and weather data is important to inform long-term decisions on climate adaptation and mitigation.
Forecasting models are limited by computational costs and, therefore, often generate coarse-resolution predictions.
Statistical downscaling, including super-resolution methods from deep learning, can provide an efficient method of upsampling low-resolution data.
- Score: 30.280862393706542
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: The availability of reliable, high-resolution climate and weather data is
important to inform long-term decisions on climate adaptation and mitigation
and to guide rapid responses to extreme events. Forecasting models are limited
by computational costs and, therefore, often generate coarse-resolution
predictions. Statistical downscaling, including super-resolution methods from
deep learning, can provide an efficient method of upsampling low-resolution
data. However, despite achieving visually compelling results in some cases,
such models frequently violate conservation laws when predicting physical
variables. In order to conserve physical quantities, here we introduce methods
that guarantee statistical constraints are satisfied by a deep learning
downscaling model, while also improving their performance according to
traditional metrics. We compare different constraining approaches and
demonstrate their applicability across different neural architectures as well
as a variety of climate and weather data sets. Besides enabling faster and more
accurate climate predictions through downscaling, we also show that our novel
methodologies can improve super-resolution for satellite data and natural
images data sets.
Related papers
- Towards Physically Consistent Deep Learning For Climate Model Parameterizations [46.07009109585047]
We propose an efficient supervised learning framework for deep learning-based parameterizations.
We show that our method robustly identifies a small subset of the inputs as actual physical drivers.
Our framework represents a crucial step in addressing a major challenge in data-driven climate model parameterizations.
arXiv Detail & Related papers (2024-06-06T10:02:49Z) - Generative Adversarial Models for Extreme Downscaling of Climate
Datasets [0.0]
We describe a conditional GAN-based geospatial downscaling method for extreme downscaling of climate datasets.
The method explicitly considers the uncertainty inherent to the downscaling process that tends to be ignored in existing methods.
We demonstrate the performances of the framework in downscaling tasks with very high scaling factors.
arXiv Detail & Related papers (2024-02-21T18:25:04Z) - Weather Prediction with Diffusion Guided by Realistic Forecast Processes [49.07556359513563]
We introduce a novel method that applies diffusion models (DM) for weather forecasting.
Our method can achieve both direct and iterative forecasting with the same modeling framework.
The flexibility and controllability of our model empowers a more trustworthy DL system for the general weather community.
arXiv Detail & Related papers (2024-02-06T21:28:42Z) - FengWu-GHR: Learning the Kilometer-scale Medium-range Global Weather
Forecasting [56.73502043159699]
This work presents FengWu-GHR, the first data-driven global weather forecasting model running at the 0.09$circ$ horizontal resolution.
It introduces a novel approach that opens the door for operating ML-based high-resolution forecasts by inheriting prior knowledge from a low-resolution model.
The hindcast of weather prediction in 2022 indicates that FengWu-GHR is superior to the IFS-HRES.
arXiv Detail & Related papers (2024-01-28T13:23:25Z) - Precipitation Downscaling with Spatiotemporal Video Diffusion [19.004369237435437]
This work extends recent video diffusion models to precipitation super-resolution.
We use a deterministic downscaler followed by a temporally-conditioned diffusion model to capture noise characteristics and high-frequency patterns.
Our analysis, capturing CRPS, MSE, precipitation distributions, and qualitative aspects using California and the Himalayas, establishes our method as a new standard for data-driven precipitation downscaling.
arXiv Detail & Related papers (2023-12-11T02:38:07Z) - Learning Robust Precipitation Forecaster by Temporal Frame Interpolation [65.5045412005064]
We develop a robust precipitation forecasting model that demonstrates resilience against spatial-temporal discrepancies.
Our approach has led to significant improvements in forecasting precision, culminating in our model securing textit1st place in the transfer learning leaderboard of the textitWeather4cast'23 competition.
arXiv Detail & Related papers (2023-11-30T08:22:08Z) - Deep Ensembles to Improve Uncertainty Quantification of Statistical
Downscaling Models under Climate Change Conditions [0.0]
We propose deep ensembles as a simple method to improve the uncertainty quantification of statistical downscaling models.
Deep ensembles allow for a better risk assessment, highly demanded by sectoral applications to tackle climate change.
arXiv Detail & Related papers (2023-04-27T19:53:18Z) - An evaluation of deep learning models for predicting water depth
evolution in urban floods [59.31940764426359]
We compare different deep learning models for prediction of water depth at high spatial resolution.
Deep learning models are trained to reproduce the data simulated by the CADDIES cellular-automata flood model.
Our results show that the deep learning models present in general lower errors compared to the other methods.
arXiv Detail & Related papers (2023-02-20T16:08:54Z) - Diffusion Models for High-Resolution Solar Forecasts [0.0]
Score-based diffusion models offer a new approach to modeling probability distributions over many dependent variables.
We apply the technique to day-ahead solar irradiance forecasts by generating many samples from a diffusion model trained to super-resolve numerical weather predictions.
arXiv Detail & Related papers (2023-02-01T01:32:25Z) - 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) - Downscaling Extreme Rainfall Using Physical-Statistical Generative
Adversarial Learning [0.0]
We develop a data-driven downscaling (super-resolution) method that incorporates physics and statistics in a generative framework to learn the fine-scale spatial details of rainfall.
Our method transforms coarse-resolution ($0.25circ times 0.25circ$) climate model outputs into high-resolution ($0.01circ times 0.01circ$) rainfall fields while efficaciously quantifying uncertainty.
arXiv Detail & Related papers (2022-12-02T21:04:32Z)
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.