基于深度学习的模型排名和缩小框架的区域气候预测:适用于欧洲气候区
Parthiban Loganathan1, Elias Zea2, Ricardo Vinuesa2
1Department of Engineering Mechanics, KTH Royal Institute of Technology, Stockholm, Sweden. parthi@kth.se.
Environmental science and pollution research international
|August 15, 2025
概括
这项研究使用深度学习方法对32种气候模型进行了排名,确定了区域气候预测中表现最佳的模型. 像GeoSTANet这样的先进变压器模型提供了精确的缩小气候数据,改善了未来的影响评估.
科学领域:
- 气候科学 气候科学
- 机器学习 机器学习
- 数据科学数据科学数据科学
背景情况:
- 准确的区域气候预测需要全球气候模型 (GCM) 的高分辨率缩放.
- 通过先进的计算技术,可以改进评估和缩小GCM的现有方法.
研究的目的:
- 开发和评估一个深度学习框架,用于对联模型相互比较项目第六阶段 (CMIP6) 的GCM进行排名.
- 通过先进的深度学习模型来改进气候预测,以实现高分辨率的缩小.
- 评估不同科本-盖格尔气候区和季节的模型性能.
主要方法:
- 使用深度学习-TOPSIS (DL-TOPSIS) 机制,根据9个绩效标准对32个CMIP6模型进行了排名.
- 在四个季节分析了五个气候区 (热带,干旱,温带,大陆,极地).
- 四个深度学习模型 (视觉变压器 (ViT),GeoSTANet,CNN-LSTM,ConvLSTM) 用于将输出缩小到0.1°分辨率.
主要成果:
- 确定NorESM2-LM,GISS-E2-1-G和HadGEM3-GC31-LL是表现最好的GCM,表现优于其他GCM.
- 在捕捉极端温度 (TXx,TNn) 方面,GeoSTANet表现出最高的准确性,其中RMSE = 1.57°C,KGE = 0.89,NSE = 0.85,r = 0.92.
- 基于变压器的缩放模型通常优于传统的深度学习方法,尽管ViT在细度波动方面遇到了困难.
结论:
- 开发的多标准排名框架有效地改善了区域气候研究的GCM选择.
- 基于变压器的深度学习模型为高分辨率的气候数据缩小提供了卓越的性能.
- 这种可扩展的框架增强了气候预测,支持气候影响评估和适应计划.
相关概念视频
Global Climate Change
24.8K
Throughout its ~4.5 billion year history, the Earth has experienced periods of warming and cooling. However, the current drastic increase in global temperatures is well outside of the Earth’s cyclic norms, and evidence for human-caused global climate change is compelling. Paleoclimatology, the study of ancient climate conditions, provides ample evidence for human-caused global climate change by comparing recent conditions with those in the past.
24.8K
What is Climate?
18.8K
Climate refers to the prevailing weather conditions in a specific area over an extended period. As the saying goes, “Climate is what you expect. Weather is what you get.” Climate is influenced by geographic factors, such as latitude, terrain, and proximity to bodies of water.
18.8K
Precipitation Processes
589
The experimental conditions in a gravimetric analysis should be optimized to maximize the particle size and purity of the obtained precipitate. Ideally, the concentration of the precipitating reagent should be low with effective stirring to maintain low relative supersaturation for the growth of large crystals. In homogeneous precipitation, the precipitant is slowly generated by a chemical reaction in the solution to avoid local reagent excesses. For example, urea decomposes gradually to...
589
Precipitation and Co-precipitation
2.0K
Precipitation and coprecipitation methods can be used to separate a mixture of ions in a solution. In qualitative inorganic analysis, ions that form sparingly soluble precipitates with the same reagent are separated based on the differences in solubility products. For example, consider the separation of Cu(II) and Fe(II) ions by precipitation as insoluble sulfides. First, copper(II) sulfide is precipitated by the addition of acidic H2S, where the dissociation of H2S is suppressed. Adding H2S...
2.0K
Regional Terms
12.6K
Regional terms describe anatomy by dividing the body parts into different regions that contain structures involved in contributing similar functions. Using these terms helps increase the accurate description and identification of the particular region of interest or region affected by the disease.
Primarily, the human body has two major regions, the axial and appendicular regions. The axial region comprises regions from the head to the abdomen and makes up the central body axis. In contrast,...
Primarily, the human body has two major regions, the axial and appendicular regions. The axial region comprises regions from the head to the abdomen and makes up the central body axis. In contrast,...
12.6K
Downsampling
253
When considering a sampled sequence with zero values between sampling instants, one can replace it by taking every N-th value of the sequence. At these integer multiples of N, the original and sampled sequences coincide. This process, known as decimation, involves extracting every N-th sample from a sequence, thereby creating a more efficient sequence.
The Fourier transform of the decimated sequence reveals a combination of scaled and shifted versions of the original spectrum. This...
The Fourier transform of the decimated sequence reveals a combination of scaled and shifted versions of the original spectrum. This...
253


