Deep learning downscaled high-resolution daily near surface meteorological datasets over East Asia

Abstract

U-Net, a deep-learning convolutional neural network, is used to downscale coarse meteorological data. Based on 19 models from the Coupled Model Intercomparison Project Phase 6 and the Multi-Source Weather (MSWX) dataset, bias correction and UNet downscaling approaches are used to develop high resolution dataset over the East Asian region, referred to as Climate Change for East Asia with Bias corrected UNet Dataset (CLIMEA-BCUD). CLIMEA-BCUD provides nine meteorological variables including 2-m air temperature, 2-m daily maximum air temperature, 2-m daily minimum air temperature, precipitation, 10-m wind speed, 2-m relative humidity, 2-m specific humidity, downward shortwave radiation and downward longwave radiation with 0.1° horizontal resolution at daily intervals over the historical period of 1950–2014 and three future scenarios (SSP1-2.6, SSP2-4.5 and SSP5-8.5) of 2015–2100. Validation against MSWX indicates that CLIMEA-BCUD shows reasonable performance in terms of climatology, and it is capable of simulating seasonal cycles and future changes well. It is suggested that CLIMEA-BCUD can promote the application of deep learning in climate research in the areas of climate change, hydrology, etc.

Background & Summary

Climate change exerts tremendous influence on water resources^1,37 dataset at 1.0° × 1.0° spacing resolution³⁸, then the UNet is trained for downscaling the biased corrected GCM dataset. The BC-UNet archive is applied to the historical simulations (1950–2014) and three future (2015–2100) scenarios of SSP1-2.6, SSP2-4.5 and SSP5-8.5. There are nine near-surface meteorological variables including 2-m air temperature (tas), 2-m daily maximum air temperature (tasmax), 2-m daily minimum air temperature (tasmin), precipitation (pr), 10-m wind speed (sfcWind), downward longwave radiation (rlds), downward shortwave radiation (rsds), 2-m relative humidity (hurs) and 2-m specific humidity (huss) (Table 1). CLIMEA-BCUD provides high-resolution large-scale DL downscaling in East Asia, which we suggest will be helpful for assessing climate change under global warming.

Table 1 Variables included in the CLIMEA-BCUD.

Methods

Data acquisition

The MSWX gridded high-resolution bias-corrected meteorological dataset is used as observations. Based on ERA5, MSWX produces 10 widely used near-surface meteorological variables with 0.1° horizontal resolution and 3-hour temporal resolution. The study area covers the whole of East Asia from 4.95°N to 60.05°N and 64.75°E to 150.25°E (Fig. 1). In order to construct the bias correction and a UNet downscaling model, the high-resolution MSWX datasets are averaged to coarse resolution at 1.0° × 1.0° as MSWX_LR using the area average method.

Fig. 1

Study area in CLIMEA-BCUD.

For climate change downscaling, we use the CMIP6 data, which provides the latest GCM simulations including voluminous global gridded model data over the historical period of 1950–2014 and four Shared Socioeconomic Pathways (SSPs) scenarios with 2015–2100 period. There are 19 GCM outputs for historical simulations and three representative future scenarios (SSP1-2.6, SSP2-4.5, and SSP5-8.5) (Table 2). As shown in Table 2, the original CMIP6 GCMs outputs have coarse spacing resolution. All CMIP6 data can be downloaded at https://esgf-node.llnl.gov/projects/cmip6/.

Table 2 CMIP6 modes included in downscaled archive.

BC-UNet

The framework to construct the CLIMEA-BCUD, called BC-UNet is demonstrated in Fig. 2. BC-UNet takes GCM simulation datasets and observation as input. It has two main steps: (1) bias correction and (2) UNet downscaling. The details of the two steps are as below.

Fig. 2

BC-UNet framework flow chart.

In the first step, the bias correction method using QDM is applied, which can reduce the bias between observations and GCMs outputs and preserves the change of model projection in quantile^39,40. When applying bias correction, the GCMs outputs are interpolated to 1° × 1° coarse horizontal resolution to match the MSWX_LR with bi-linear interpolation algorithm. Then QDM is used to correct the biases between GCMs and MSWX_LR at coarse resolution, and to calculate the bias corrected GCM results (GCM_BC).

In the second step, UNet with 3 layers neural network, known for its exceptional performance in super-resolution and downscaling tasks, is used for climate downscaling⁴¹. Every convolution and downsampling operation lead to a feature map, which captures the spatial features. The UNet with 3 layers represent 3 downsampling and 3 upsampling. A convolution operation of each layer will generate a feature map, and the number of convolution channels represents the number of feature maps extracted by this layer. The downsampling component of UNet captures crucial spatial features, while the upsampling counterpart generates high-resolution data, effectively facilitating the downscaling process. The convolution channel numbers to capture the spatial features in UNet are {64, 96, 128, 160} for precipitation and {56, 112, 224, 448} for the other variables (Fig. 3).

Fig. 3

UNet framework in this study. Blue block indicates the convolution, batch normalization and ReLu operation. Yellow block corresponds to the max pooling operation to downsampling. Red block means the transparent convolution operation to upsampling. Grey arrow and grey block means the skip connection to merge the feature maps.

As the goal of training stage, the loss function^{https://doi.org/10.57760/sciencedb.07718)44. While CLIMEA-BCUD has a wonderful performance in producing the overall patterns of climate mean, seasonal cycle, frequency, and future changes, some limitations must be acknowledged. Firstly, data users should be aware of underestimation when using CLIMEA-BCUD due to its underestimation in representing observations. Secondly, despite displaying good performance in reproducing seasonal variability and extreme events, the bias-corrected products may contain inherent uncertainties, and obscure some fundamental deficiencies presented by the climate models.}

Numerous studies have extensively researched methods to enhance model performance in the field of super-resolution, and these advancements are expected to be applicable to downscaling tasks as well. Among them, image enhancement techniques including adaptive gamma correction with weighting distribution⁴⁶ (AGCWD), adaptive gamma correction with color preserving framework⁴⁷ (AGCCPF), range limited Bi-histogram equalization^48,49 (RLBHE), and region adaptive contrast limited adapted histogram equalization⁵⁰ (RACLAHE) are common and powerful tools for improving the performance of DL model. It is valuable to explore its effectiveness in the context of climate downscaling. Furthermore, several studies have explored improved models based on UNet such as UNet++⁵¹, UNet3+⁵², ResUNet⁵³ and USE-NET⁵⁴, which have demonstrated significant potential in various applications. Additionally, models that combine technologies such as generative adversarial network⁵⁵ (GAN) and Transformer⁵⁶ have also shown great potential for further improvement.