arXiv - PHYS - Atmospheric and Oceanic Physics最新文献_第6页

Generative Diffusion Model-based Downscaling of Observed Sea Surface Height over Kuroshio Extension since 2000 基于生成扩散模型的 2000 年以来黑潮扩展区观测海面高度降尺度研究

arXiv - PHYS - Atmospheric and Oceanic Physics Pub Date : 2024-08-22 DOI: arxiv-2408.12632

Qiuchang Han, Xingliang Jiang, Yang Zhao, Xudong Wang, Zhijin Li, Renhe Zhang

{"title":"Generative Diffusion Model-based Downscaling of Observed Sea Surface Height over Kuroshio Extension since 2000","authors":"Qiuchang Han, Xingliang Jiang, Yang Zhao, Xudong Wang, Zhijin Li, Renhe Zhang","doi":"arxiv-2408.12632","DOIUrl":"https://doi.org/arxiv-2408.12632","url":null,"abstract":"Satellite altimetry has been widely utilized to monitor global sea surface\u0000dynamics, enabling investigation of upper ocean variability from basin-scale to\u0000localized eddy ranges. However, the sparse spatial resolution of observational\u0000altimetry limits our understanding of oceanic submesoscale variability,\u0000prevalent at horizontal scales below 0.25o resolution. Here, we introduce a\u0000state-of-the-art generative diffusion model to train high-resolution sea\u0000surface height (SSH) reanalysis data and demonstrate its advantage in\u0000observational SSH downscaling over the eddy-rich Kuroshio Extension region. The\u0000diffusion-based model effectively downscales raw satellite-interpolated data\u0000from 0.25o resolution to 1/16o, corresponding to approximately 12-km\u0000wavelength. This model outperforms other high-resolution reanalysis datasets\u0000and neural network-based methods. Also, it successfully reproduces the spatial\u0000patterns and power spectra of satellite along-track observations. Our\u0000diffusion-based results indicate that eddy kinetic energy at horizontal scales\u0000less than 250 km has intensified significantly since 2004 in the Kuroshio\u0000Extension region. These findings underscore the great potential of deep\u0000learning in reconstructing satellite altimetry and enhancing our understanding\u0000of ocean dynamics at eddy scales.","PeriodicalId":501166,"journal":{"name":"arXiv - PHYS - Atmospheric and Oceanic Physics","volume":"159 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142215443","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Extremes of Dissolved Oxygen in the California Current System 加利福尼亚洋流系统中溶解氧的极值

arXiv - PHYS - Atmospheric and Oceanic Physics Pub Date : 2024-08-22 DOI: arxiv-2408.12287

J. Xavier ProchaskaUniversity of California, Santa CruzScripps Institution of OceanographSimons Pivot Fellow, Daniel RudnickScripps Institution of Oceanograph

{"title":"Extremes of Dissolved Oxygen in the California Current System","authors":"J. Xavier ProchaskaUniversity of California, Santa CruzScripps Institution of OceanographSimons Pivot Fellow, Daniel RudnickScripps Institution of Oceanograph","doi":"arxiv-2408.12287","DOIUrl":"https://doi.org/arxiv-2408.12287","url":null,"abstract":"Dissolved oxygen (DO) is a non-conservative tracer of interactions at the\u0000air-sea interface, respiration and photosynthesis, and advection. In this\u0000manuscript, we study extremes in the degree of oxygen saturation (SO), the\u0000ratio of DO to the maximum concentration given the water's temperature,\u0000salinity, and depth with SO=1 critically saturated. We perform the analysis\u0000with the California Underwater Glider Network (CUGN), which operates gliders on\u0000four lines that extend from the California coast to several hundred kilometers\u0000offshore, profiling to 500m depth every 3km. Since ~2017, the gliders have been\u0000equipped with a Sea-Bird 63 optode sensor to measure the DO content. We find\u0000that parcels with SO>1.1, hyperoxic extrema, occur primarily near-shore in the\u0000upper 50m of the water column and during non-winter months. Along Line 90 which\u0000originates in San Diego, these hyperoxic events occur primarily in stratified\u0000waters with shallow mixed layers. We hypothesize that photosynthesis elevates\u0000DO in sub-surface water that can not rapidly ventilate with the surface. Along\u0000the three other lines, hyperoxic extrema occur almost exclusively at the\u0000surface and are correlated with elevated Chl-a fluorescence suggesting they are\u0000primarily driven by blooms of photosynthesis. We also examine hypoxic extrema,\u0000finding that parcels with SO<0.9 and z<50m occur most frequently along the\u0000northernmost line where upwelling has greatest impact.","PeriodicalId":501166,"journal":{"name":"arXiv - PHYS - Atmospheric and Oceanic Physics","volume":"9 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142215445","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Improving Typhoon Predictions by Integrating Data-Driven Machine Learning Models with Physics Models Based on the Spectral Nudging and Data Assimilation 将数据驱动的机器学习模型与基于频谱推算和数据同化的物理模型相结合，改进台风预测工作

arXiv - PHYS - Atmospheric and Oceanic Physics Pub Date : 2024-08-22 DOI: arxiv-2408.12630

Zeyi Niu, Wei Huang, Lei Zhang, Lin Deng, Haibo Wang, Yuhua Yang, Dongliang Wang, Hong Li

{"title":"Improving Typhoon Predictions by Integrating Data-Driven Machine Learning Models with Physics Models Based on the Spectral Nudging and Data Assimilation","authors":"Zeyi Niu, Wei Huang, Lei Zhang, Lin Deng, Haibo Wang, Yuhua Yang, Dongliang Wang, Hong Li","doi":"arxiv-2408.12630","DOIUrl":"https://doi.org/arxiv-2408.12630","url":null,"abstract":"With the rapid development of data-driven machine learning (ML) models in\u0000meteorology, typhoon track forecasts have become increasingly accurate.\u0000However, current ML models still face challenges, such as underestimating\u0000typhoon intensity and lacking interpretability. To address these issues, this\u0000study establishes an ML-driven hybrid typhoon model, where forecast fields from\u0000the Pangu-Weather model are used to constrain the large-scale forecasts of the\u0000Weather Research and Forecasting model based on the spectral nudging method\u0000(Pangu_SP). The results show that forecasts from the Pangu_SP experiment\u0000obviously outperform those by using the Global Forecast System as the initial\u0000field (GFS_INIT) and from the Integrated Forecasting System of the European\u0000Centre for Medium-Range Weather Forecasts (ECMWF IFS) for the track forecast of\u0000Typhoon Doksuri (2023). The predicted typhoon cloud patterns from Pangu_SP are\u0000also more consistent with satellite observations. Additionally, the typhoon\u0000intensity forecasts from Pangu_SP are notably more accurate than those from the\u0000ECMWF IFS, demonstrating that the hybrid model effectively leverages the\u0000strengths of both ML and physical models. Furthermore, this study is the first\u0000to explore the significance of data assimilation in ML-driven hybrid dynamical\u0000systems. The findings reveal that after assimilating water vapor channels from\u0000the Advanced Geostationary Radiation Imager onboard Fengyun-4B, the errors in\u0000typhoon intensity forecasts are reduced.","PeriodicalId":501166,"journal":{"name":"arXiv - PHYS - Atmospheric and Oceanic Physics","volume":"2 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142215450","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Climate Bistability at the Inner Edge of the Habitable Zone due to Runaway Greenhouse and Cloud Feedbacks 温室效应和云反作用失控导致宜居带内边缘气候不稳定性

arXiv - PHYS - Atmospheric and Oceanic Physics Pub Date : 2024-08-22 DOI: arxiv-2408.12563

Bowen Fan, Da Yang, Dorian S. Abbot

{"title":"Climate Bistability at the Inner Edge of the Habitable Zone due to Runaway Greenhouse and Cloud Feedbacks","authors":"Bowen Fan, Da Yang, Dorian S. Abbot","doi":"arxiv-2408.12563","DOIUrl":"https://doi.org/arxiv-2408.12563","url":null,"abstract":"Understanding the climate dynamics at the inner edge of the habitable zone\u0000(HZ) is crucial for predicting the habitability of rocky exoplanets. Previous\u0000studies using Global Climate Models (GCMs) have indicated that planets\u0000receiving high stellar flux can exhibit climate bifurcations, leading to\u0000bistability between a cold (temperate) and a hot (runaway) climate. However,\u0000the mechanism causing this bistability has not been fully explained, in part\u0000due to the difficulty associated with inferring mechanisms from small numbers\u0000of expensive numerical simulations in GCMs. In this study, we employ a\u0000two-column (dayside and nightside), two-layer climate model to investigate the\u0000physical mechanisms driving this bistability. Through mechanism-denial\u0000experiments, we demonstrate that the runaway greenhouse effect, coupled with a\u0000cloud feedback on either the dayside or nightside, leads to climate\u0000bistability. We also map out the parameters that control the location of the\u0000bifurcations and size of the bistability. This work identifies which mechanisms\u0000and GCM parameters control the stellar flux at which rocky planets are likely\u0000to retain a hot, thick atmosphere if they experience a hot start. This is\u0000critical for the prioritization of targets and interpretation of observations\u0000by the James Webb Space Telescope (JWST). Furthermore, our modeling framework\u0000can be extended to planets with different condensable species and cloud types.","PeriodicalId":501166,"journal":{"name":"arXiv - PHYS - Atmospheric and Oceanic Physics","volume":"11 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142215447","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Changes in anthropogenic aerosols during the first wave of COVID-19 lockdowns in the context of long-term historical trends at 51 AERONET stations 从 51 个 AERONET 台站的长期历史趋势看 COVID-19 第一波锁定期间人为气溶胶的变化

arXiv - PHYS - Atmospheric and Oceanic Physics Pub Date : 2024-08-21 DOI: arxiv-2408.11757

Robert Blaga, Delia Calinoiu, Gavrila Trif-Tordai

{"title":"Changes in anthropogenic aerosols during the first wave of COVID-19 lockdowns in the context of long-term historical trends at 51 AERONET stations","authors":"Robert Blaga, Delia Calinoiu, Gavrila Trif-Tordai","doi":"arxiv-2408.11757","DOIUrl":"https://doi.org/arxiv-2408.11757","url":null,"abstract":"A quasi-consensus has steadily formed in the scientific literature on the\u0000fact that the prevention measures implemented by most countries to curb the\u00002020 COVID-19 pandemic have led to significant reductions in pollution levels\u0000around the world, especially in urban environments. Fewer studies have looked\u0000at the how these reductions at ground level translate into variations in the\u0000whole atmosphere. In this study, we examine the columnar values of aerosols at\u000051 mainland European stations of the Aerosol Robotic Network (AERONET). We show\u0000that when considered in the context of the long-term trend over the last\u0000decade, the columnar aerosol levels for 2020, at the regional level, do not\u0000appear exceptional. Both the yearly means and the number of episodes with\u0000extreme values for this period are within the one standard deviation of the\u0000long-term trends. We conclude that the spatially and temporally very localized\u0000reductions do not add up to statistically significant reductions in the global\u0000levels of aerosols. Furthermore, considering that pandemic lockdowns can be\u0000thought of as a simulation of a climate change mitigation scenario, we conclude\u0000that such lifestyle-based changes present a very low potential as a global\u0000climate change mitigation strategy.","PeriodicalId":501166,"journal":{"name":"arXiv - PHYS - Atmospheric and Oceanic Physics","volume":"159 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142215451","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

DABench: A Benchmark Dataset for Data-Driven Weather Data Assimilation DABench：数据驱动的天气数据同化基准数据集

arXiv - PHYS - Atmospheric and Oceanic Physics Pub Date : 2024-08-21 DOI: arxiv-2408.11438

Wuxin Wang, Weicheng Ni, Tao Han, Lei Bai, Boheng Duan, Kaijun Ren

{"title":"DABench: A Benchmark Dataset for Data-Driven Weather Data Assimilation","authors":"Wuxin Wang, Weicheng Ni, Tao Han, Lei Bai, Boheng Duan, Kaijun Ren","doi":"arxiv-2408.11438","DOIUrl":"https://doi.org/arxiv-2408.11438","url":null,"abstract":"Recent advancements in deep learning (DL) have led to the development of\u0000several Large Weather Models (LWMs) that rival state-of-the-art (SOTA)\u0000numerical weather prediction (NWP) systems. Up to now, these models still rely\u0000on traditional NWP-generated analysis fields as input and are far from being an\u0000autonomous system. While researchers are exploring data-driven data\u0000assimilation (DA) models to generate accurate initial fields for LWMs, the lack\u0000of a standard benchmark impedes the fair evaluation among different data-driven\u0000DA algorithms. Here, we introduce DABench, a benchmark dataset utilizing ERA5\u0000data as ground truth to guide the development of end-to-end data-driven weather\u0000prediction systems. DABench contributes four standard features: (1) sparse and\u0000noisy simulated observations under the guidance of the observing system\u0000simulation experiment method; (2) a skillful pre-trained weather prediction\u0000model to generate background fields while fairly evaluating the impact of\u0000assimilation outcomes on predictions; (3) standardized evaluation metrics for\u0000model comparison; (4) a strong baseline called the DA Transformer (DaT). DaT\u0000integrates the four-dimensional variational DA prior knowledge into the\u0000Transformer model and outperforms the SOTA in physical state reconstruction,\u0000named 4DVarNet. Furthermore, we exemplify the development of an end-to-end\u0000data-driven weather prediction system by integrating DaT with the prediction\u0000model. Researchers can leverage DABench to develop their models and compare\u0000performance against established baselines, which will benefit the future\u0000advancements of data-driven weather prediction systems. The code is available\u0000on this Github repository and the dataset is available at the Baidu Drive.","PeriodicalId":501166,"journal":{"name":"arXiv - PHYS - Atmospheric and Oceanic Physics","volume":"180 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142215449","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Rapid Statistical-Physical Adversarial Downscaling Reveals Bangladesh's Rising Rainfall Risk in a Warming Climate 快速统计物理对抗性降尺度揭示了孟加拉国在气候变暖情况下不断上升的降雨风险

arXiv - PHYS - Atmospheric and Oceanic Physics Pub Date : 2024-08-21 DOI: arxiv-2408.11790

Anamitra Saha, Sai Ravela

{"title":"Rapid Statistical-Physical Adversarial Downscaling Reveals Bangladesh's Rising Rainfall Risk in a Warming Climate","authors":"Anamitra Saha, Sai Ravela","doi":"arxiv-2408.11790","DOIUrl":"https://doi.org/arxiv-2408.11790","url":null,"abstract":"In Bangladesh, a nation vulnerable to climate change, accurately quantifying\u0000the risk of extreme weather events is crucial for planning effective adaptation\u0000and mitigation strategies. Downscaling coarse climate model projections to\u0000finer resolutions is key in improving risk and uncertainty assessments. This\u0000work develops a new approach to rainfall downscaling by integrating statistics,\u0000physics, and machine learning and applies it to assess Bangladesh's extreme\u0000rainfall risk. Our method successfully captures the observed spatial pattern\u0000and risks associated with extreme rainfall in the present climate. It also\u0000produces uncertainty estimates by rapidly downscaling multiple models in a\u0000future climate scenario(s). Our analysis reveals that the risk of extreme\u0000rainfall is projected to increase throughout Bangladesh mid-century, with the\u0000highest risk in the northeast. The daily maximum rainfall at a 100-year return\u0000period is expected to rise by approximately 50 mm per day. However, using\u0000multiple climate models also indicates considerable uncertainty in the\u0000projected risk.","PeriodicalId":501166,"journal":{"name":"arXiv - PHYS - Atmospheric and Oceanic Physics","volume":"7 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142215448","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Satellite monitoring of annual US landfill methane emissions and trends 对美国垃圾填埋场甲烷年排放量和趋势的卫星监测

arXiv - PHYS - Atmospheric and Oceanic Physics Pub Date : 2024-08-20 DOI: arxiv-2408.10957

Nicholas Balasus, Daniel J. Jacob, Gabriel Maxemin, Carrie Jenks, Hannah Nesser, Joannes D. Maasakkers, Daniel H. Cusworth, Tia R. Scarpelli, Daniel J. Varon, Xiaolin Wang

{"title":"Satellite monitoring of annual US landfill methane emissions and trends","authors":"Nicholas Balasus, Daniel J. Jacob, Gabriel Maxemin, Carrie Jenks, Hannah Nesser, Joannes D. Maasakkers, Daniel H. Cusworth, Tia R. Scarpelli, Daniel J. Varon, Xiaolin Wang","doi":"arxiv-2408.10957","DOIUrl":"https://doi.org/arxiv-2408.10957","url":null,"abstract":"We use satellite observations of atmospheric methane from the TROPOMI\u0000instrument to estimate total annual methane emissions for 2019-2023 from four\u0000large Southeast US landfills with gas collection and control systems. The\u0000emissions are on average 6$times$ higher than the values reported by the\u0000landfills to the US Greenhouse Gas Reporting Program (GHGRP) which are used by\u0000the US Environmental Protection Agency (EPA) for its national Greenhouse Gas\u0000Inventory (GHGI). We find increasing emissions over the 2019-2023 period\u0000whereas the GHGRP reports a decrease. The GHGRP requires gas-collecting\u0000landfills to estimate their annual emissions either with a recovery-first model\u0000(estimating emissions as a function of methane recovered) or a generation-first\u0000model (estimating emissions from a first-order-decay applied to\u0000waste-in-place). All four landfills choose to use the recovery-first model,\u0000which yields emissions that are one-quarter of those from the generation-first\u0000model and decreasing over 2019-2023, in contrast with the TROPOMI observations.\u0000Our TROPOMI estimates for two of the landfills agree with the generation-first\u0000model, with increasing emissions over 2019-2023 due to increasing\u0000waste-in-place or decreasing methane recovery, and are still higher than the\u0000generation-first model for the other two landfills. Further examination of the\u0000GHGRP emissions from all reporting landfills in the US shows that the 19%\u0000decrease in landfill emissions reported by the GHGI over 2005-2022 reflects an\u0000increasing preference for the recovery-first model by the reporting landfills,\u0000rather than an actual emission decrease. The generation-first model would imply\u0000an increase in landfill emissions over 2013-2022, and this is more consistent\u0000with atmospheric observations.","PeriodicalId":501166,"journal":{"name":"arXiv - PHYS - Atmospheric and Oceanic Physics","volume":"109 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142215453","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Forecasting seasonal rainfall in SE Australia using Empirical Orthogonal Functions and Neural Networks 利用经验正交函数和神经网络预测澳大利亚东南部的季节性降雨量

arXiv - PHYS - Atmospheric and Oceanic Physics Pub Date : 2024-08-20 DOI: arxiv-2408.10550

Stjepan Marcelja

引用次数: 0

MambaDS: Near-Surface Meteorological Field Downscaling with Topography Constrained Selective State Space Modeling MambaDS：利用地形约束选择性状态空间建模进行近地表气象场降维分析

arXiv - PHYS - Atmospheric and Oceanic Physics Pub Date : 2024-08-20 DOI: arxiv-2408.10854

Zili Liu, Hao Chen, Lei Bai, Wenyuan Li, Wanli Ouyang, Zhengxia Zou, Zhenwei Shi

{"title":"MambaDS: Near-Surface Meteorological Field Downscaling with Topography Constrained Selective State Space Modeling","authors":"Zili Liu, Hao Chen, Lei Bai, Wenyuan Li, Wanli Ouyang, Zhengxia Zou, Zhenwei Shi","doi":"arxiv-2408.10854","DOIUrl":"https://doi.org/arxiv-2408.10854","url":null,"abstract":"In an era of frequent extreme weather and global warming, obtaining precise,\u0000fine-grained near-surface weather forecasts is increasingly essential for human\u0000activities. Downscaling (DS), a crucial task in meteorological forecasting,\u0000enables the reconstruction of high-resolution meteorological states for target\u0000regions from global-scale forecast results. Previous downscaling methods,\u0000inspired by CNN and Transformer-based super-resolution models, lacked tailored\u0000designs for meteorology and encountered structural limitations. Notably, they\u0000failed to efficiently integrate topography, a crucial prior in the downscaling\u0000process. In this paper, we address these limitations by pioneering the\u0000selective state space model into the meteorological field downscaling and\u0000propose a novel model called MambaDS. This model enhances the utilization of\u0000multivariable correlations and topography information, unique challenges in the\u0000downscaling process while retaining the advantages of Mamba in long-range\u0000dependency modeling and linear computational complexity. Through extensive\u0000experiments in both China mainland and the continental United States (CONUS),\u0000we validated that our proposed MambaDS achieves state-of-the-art results in\u0000three different types of meteorological field downscaling settings. We will\u0000release the code subsequently.","PeriodicalId":501166,"journal":{"name":"arXiv - PHYS - Atmospheric and Oceanic Physics","volume":"59 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142215452","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0