Search

Abstract The sensitivity strength of air temperature (T) to surface soil temperature (sST) (namely β hereafter) constitutes a significant factor in how global climate models quantify changes in the climate. This study examines how this sensitivity is represented in the CMIP6 models. Results show regional differences and even contrasts in the β trends at interannual scales between climate models and two reanalysis products during 1980–2014. At high latitudes in the Northern Hemisphere (NH), β is stronger in the CMIP6 models than in the reanalysis data. Additionally, the β trends differ between the CMIP6 and reanalysis data, which may be related to the different precipitation (PR) and soil water availability (PR-ET) trends between the models. In the regions of increasing β intensity at high latitudes in the NH, sST is more sensitive to PR and PR-ET. Consequently, increasing PR and PR-ET leads to slower sST changes, potentially making β intensity stronger in CMIP6 models. However, in the reanalysis data, decreases in PR and PR-ET accelerate sST changes, leading to a weakening of the β intensity. A resulting implication is that β, based on high-emission scenarios, shows a stronger trend during 2015–2100, although this trend could potentially be overestimated. The findings contribute to a better understanding of the sensitivity of T to sST and facilitate the assessment of energy exchange between the land surface and the atmosphere in climate models.

This work assesses the ability of nine Coupled Model Intercomparison Project phase 6 (CMIP6) High-Resolution Model Intercomparison Project (HighResMIP) models and their ensemble mean to reproduce precipitation extremes over East Africa for the period 1995–2014. The model datasets are assessed against two observation datasets: CHIRPS and GPCC. The precipitation indices considered are CDD, CWD, R1mm, R10mm, R20mm, SDII, R95p, PRCPTOT, and Rx1day. The overall results show that HighResMIP models reproduce annual variability fairly well; however, certain consistent biases are found across HighResMIP models, which tend to overestimate CWD and R1mm and underestimate CDD and SDII. The HighResMIP models are ranked using the Taylor diagram and Taylor Skill Score. The results show that the models reasonably simulate indices, such as PRCPTOT, R1mm, R10mm, R95p, and CDD; however, the simulation of SDII CWD, SDII, and R20mm is generally poor. They are CMCC-CM2-VHR4, HadGEM31-MM, HadGEM3-GC31-HM, and GFDL-CM4. Conversely, MPI-ESM1-2-XR and MPI-ESM1-2-HR show remarkable performance in simulating the OND season while underestimating the MAM season. A comparative analysis demonstrates that the MME has better accuracy than the individual models in the simulation of the various indices. The findings of the present study are important to establish the ability of HighResMIP data to reproduce extreme precipitation events over East Africa and, thus, help in decision making. However, caution should be exercised in the interpretation of the findings based on individual CMIP6 models over East Africa given the overall weakness observed in reproducing mean precipitation.

Tropical cyclones (TCs) can cause significant economic damage and loss of life in coastal areas. Therefore, TC prediction has become a crucial topic in current research. In recent years, TC track prediction has progressed considerably, and intensity prediction remains a challenge due to the complex mechanism of TC structure. In this study, we propose a model for short-term intensity prediction based on adaptive weight learning (AWL-Net) for the evolution of the TC’s structure as well as intensity changes, exploring the multidimensional fusion of features including TC morphology, structure, and scale. Furthermore, in addition to using satellite imageries, we construct a dataset that can more comprehensively explore the degree of TC cloud organization and structure evolution. Considering the information difference between multi-source data, a multi-branch structure is constructed and adaptive weight learning (AWL) is designed. In addition, according to the three-dimensional dynamic features of TC, 3D Convolutional Gated Recurrent (3D ConvGRU) is used to achieve feature enhancement, and then 3D Convolutional Neural Network (CNN) is used to capture and learn TC temporal and spatial features. Experiments on a sample of northwest Pacific TCs and official agency TC intensity prediction records are used to validate the effectiveness of our proposed model, and the results show that our model is able to focus well on the spatial and temporal features associated with TC intensity changes, with a root mean square error (RMSE) of 10.62 kt for the TC 24 h intensity forecast.

Accurate tropical cyclone (TC) intensity and size estimation are key in disaster management and prevention. While great breakthroughs have been made in TC intensity estimation research, there is currently a lack of research on TC size reflecting TC influence radius. Therefore, we propose a lightweight multi-task learning model (TC-MTLNet) with adaptive loss balance to simultaneously estimate TC intensity and size. Adaptive loss balance is utilized to solve the problem of inconsistent convergence speed of TC intensity and size estimation tasks. The model based on four 2-D convolutions, four 3-D convolutions and three fully connected layers takes up less computational and storage space and improves the accuracy of TC intensity and size estimation by sharing knowledge among multiple tasks. In addition, due to the imbalanced distribution of TC samples, with significantly few low-intensity and high-intensity TC satellite data, this phenomenon poses a great challenge to TC intensity and size estimation. So, we utilize the influence of nearby samples to calibrate the sample density to weight the loss function to enable the model to be generalized to all samples. The result shows that the root-mean-square error (RMSE) of TC intensity estimation is $\text{8.40}\,\text{kts}$, which is 33.5% lower than that of the Advanced Dvorak Technique (ADT) and 11.4% lower than that of the deep learning method (3DAttentionTCNet). The mean absolute error (MAE) of the TC size estimation is $\text{20.89}\,\text{nmi}$, which is a 16% reduction compared to the Multi-Platform Tropical Cyclone Surface Winds Analysis (MTCSWA).

The transfer of atmospheric kinetic energy to the ocean is one of the major concerns in climate research. According to previous studies, the work of wind stress on oceanic mesoscale eddy is negative in most oceans, referred to as “eddy killing”. However, another recent work, finds that the wind work on an eddy varies with interaction time. To better understand the wind work on eddies, the present study uses satellite remote sensing wind stress data and eddy data from 2000 to 2021 to investigate the effects of wind stress on eddies in the northeast tropical Pacific (NETP). The study demonstrates that the work done by the wind stress on eddies in this region varies seasonally and that there is a strong spatiotemporal link between the work done and the wind stress curl. The work of wind stress with positive (negative) curl on the entire area of a cyclonic eddy is positive (negative), and vice versa on an anticyclonic eddy. These results indicate that wind energy input is sensitive to wind stress curl, and eddies do not always hinder wind energy input in this area.

Validation of observed gridded precipitation datasets sourced from satellites or reanalysis over Africa remains a challenge due to the dearth of in-situ products that can act as a true estimate. To address this gap, this study compares the performance of different precipitation products (gauge, reanalysis, and satellite-based) sourced from the Frequent Rainfall Observations on GridS (FROGS) database over Africa. Satellite products are classified as corrected (incorporating gauge observations into their algorithms) or uncorrected, which implies that temporal variations depend entirely on the satellite. The main aim is to identify regions where precipitation products depict minimal uncertainties, supporting the use of the datasets in understanding precipitation variability in the specific regions. This is achieved by applying the triple collocation approach, which takes advantage of three collocated datasets of the same variable to derive the mean square error without requiring knowledge of the true value. The results show that light precipitation (1–5 mm d ^−1 ) was prevalent in most regions of Africa during the study duration (2001–2016). Estimating the spatial distribution of daily precipitation greater than the 90th percentiles suggests that extreme precipitation is mainly detected over the Central Africa region and coastal regions of West Africa, where the majority of uncorrected satellite products show consistent performance. The satellite product CMORPH_V1_RAW shows higher estimates of 90th percentile precipitation among the uncorrected satellite products. The ability of precipitation products to detect rainy or non-rainy days shows that corrected satellite products depict notable agreement for probability of detection and false alarm ratio over most regions of Africa. Overall, better performance is demonstrated by the IMERG products, ARCv2, CHIRPSv2 and PERSIANN_CDRv1r1 (corrected), and GPCC, CPC_v1.0 and REGEN_ALL (gauge) during the study period. Among the reanalysis products, ERA5 datasets shows good performance in estimating daily precipitation over Africa. The optimal maps that show the classification of products in regions where they depict reliable performance can be recommended for various usage by different stakeholders.

Radar echo extrapolation provides important information for precipitation nowcasting. Existing mainstream radar echo extrapolation methods are based on the Single-Input-Single-Output (SISO) architecture. These approaches of recursively predicting the predictive echo image with the current echo image as input often results in error accumulation, leading to severe performance degradation. In addition, the echo motion variations are extremely complex. Different regions of strong or weak echoes should receive different degrees of attention. Previous methods have not been specifically designed for this aspect. This paper proposes a new radar echo extrapolation network based entirely on a convolutional neural network (CNN). The network uses a Multi-Input-Multi-Output (MIMO) architecture to mitigate cumulative errors. It incorporates a multi-scale, large kernel convolutional attention module that enhances the extraction of both local and global information. This design results in improved performance while significantly reducing training costs. Experiments on dual-polarization radar echo datasets from Shijiazhuang and Nanjing show that the proposed fully CNN-based model can achieve better performance while reducing computational cost.

Differentiating between snow and clouds presents a formidable challenge in the context of mapping snow cover over the Qinghai–Tibetan Plateau (QTP). The frequent presence of cloudy conditions severely complicates the discrimination of snow cover from satellite imagery. To accurately monitor the spatiotemporal evolution of snow cover, it is imperative to address these challenges and enhance the segmentation schemes employed for snow cover assessment. In this study, we devised a pixel-wise classification algorithm based on Support Vector Machine (SVM) called the 3-D Orientation Gradient algorithm (3-D OG), which captures the variations of the gradient direction of snow and clouds in spatiotemporal dimensions based on geostationary satellite “Fengyun-4A” (FY-4A) multi-spectral and multi-temporal optical imagery. This algorithm assumes that the speed and direction of clouds and snow are different in the process of movement leading to their discrepancy of gradient characteristics in time and space. Therefore, in this algorithm, the gradient of the images in the spatiotemporal dimensions is calculated first, and then the movement angle and trend are obtained based on that. Finally, the feature space is composed of the multi-spectral image, gradient image, and movement feature maps, which are used as the input of the SVM. Our results demonstrate that the proposed algorithm can identify snow and clouds more accurately during snowfall by utilizing the FY-4A’s high temporal resolution image. Weather station data, which was collected during snowstorms in the QTP, were used for evaluating the accuracy of our algorithm. It is demonstrated that the overall accuracy of snow cover segmentation by using the 3-D OG algorithm is improved by at least 12% and 10% as compared to snow products of Fengyun-2 and MODIS, respectively. Overall, the proposed algorithm has overcome the axial swing errors existing in Geostationary satellites and is successfully applied to cloud and snow segmentation in QTP. Furthermore, our study underscores that the visible and near-infrared bands of Fengyun-4A can be used for near real-time snow cover monitoring with high performance using the 3-D OG algorithm.

The Qinghai–Tibet Plateau is one of the regions with the highest snow accumulation in China. Although the Fengyun-4A (FY4A) satellite is capable of monitoring snow-covered areas in real time and on a wide scale at high temporal resolution, its spatial resolution is low. In this study, the Qinghai–Tibet Plateau, which has a harsh climate with few meteorological stations, was selected as the study area. We propose a deep learning model called the Dual-Branch Super-Resolution Semantic Segmentation Network (DSRSS-Net), in which one branch focuses with super resolution to obtain high-resolution snow distributions and the other branch carries out semantic segmentation to achieve accurate snow recognition. An edge enhancement module and coordinated attention mechanism were introduced into the network to improve the classification performance and edge segmentation effect for cloud versus snow. Multi-task loss is also used for optimization, including feature affinity loss and edge loss, to obtain fine structural information and improve edge segmentation. The 1 km resolution image obtained by coupling bands 1, 2, and 3; the 2 km resolution image obtained by coupling bands 4, 5, and 6; and the 500 m resolution image for a single channel, band 2, were inputted into the model for training. The accuracy of this model was verified using ground-based meteorological station data. Snow classification accuracy, false detection rate, and total classification accuracy were compared with the MOD10A1 snow product. The results show that, compared with MOD10A1, the snow classification accuracy and the average total accuracy of DSRSS-Net improved by 4.45% and 5.1%, respectively. The proposed method effectively reduces the misidentification of clouds and snow, has higher classification accuracy, and effectively improves the spatial resolution of FY-4A satellite snow cover products.

Rapid and accurate identification of precipitation clouds from satellite observations is essential for the research of quantitative precipitation estimation and precipitation nowcasting. In this study, we proposed a novel Convolutional Neural Network (CNN)-based algorithm for precipitation cloud identification (PCINet) in the daytime, nighttime, and nychthemeron. High spatiotemporal and multi-spectral information from the Fengyun-4A (FY-4A) satellite is utilized as the inputs, and a multi-scale structure and skip connection constraint strategy are presented in the framework of the algorithm to improve the precipitation cloud identification. Moreover, the effectiveness of visible/near-infrared spectral information in improving daytime precipitation cloud identification is explored. To evaluate this algorithm, we compare it with five other deep learning models used for image segmentation and perform qualitative and quantitative analyses of long-time series using data from 2021. In addition, two heavy precipitation events are selected to analyze the spatial distribution of precipitation cloud identification. Statistics and visualization of the experiment results show that the proposed model outperforms the baseline models in this task, and adding visible/near-infrared spectral information in the daytime can effectively improve model performance. More importantly, the proposed model can provide accurate and near-real-time results, which has important application in observing precipitation clouds.

Water bodies extraction is important in water resource utilization and flood prevention and mitigation. Remote sensing images contain rich information, but due to the complex spatial background features and noise interference, problems such as inaccurate tributary extraction and inaccurate segmentation occur when extracting water bodies. Recently, using a convolutional neural network (CNN) to extract water bodies is gradually becoming popular. However, the local property of CNN limits the extraction of global information, while Transformer, using a self-attention mechanism, has great potential in modeling global information. This paper proposes the MU-Net, a hybrid MixFormer architecture, as a novel method for automatically extracting water bodies. First, the MixFormer block is embedded into Unet. The combination of CNN and MixFormer is used to model the local spatial detail information and global contextual information of the image to improve the ability of the network to capture semantic features of the water body. Then, the features generated by the encoder are refined by the attention mechanism module to suppress the interference of image background noise and non-water body features, which further improves the accuracy of water body extraction. The experiments show that our method has higher segmentation accuracy and robust performance compared with the mainstream CNN- and Transformer-based semantic segmentation networks. The proposed MU-Net achieves 90.25% and 76.52% IoU on the GID and LoveDA datasets, respectively. The experimental results also validate the potential of MixFormer in water extraction studies.

Precipitation nowcasting is an important tool for economic and social services, especially for forecasting severe weather. The crucial and challenging part of radar echo image prediction is the focus of radar-based precipitation nowcasting. Recently, a number of deep learning models have been designed to solve the problem of extrapolating radar images. Although these methods can generate better results than traditional extrapolation methods, the issue of error accumulation in precipitation forecasting is exacerbated by using only the mean square error (MSE) and mean absolute error (MAE) as loss functions. In this paper, we approach the problem from the perspective of the loss function and propose dynamic weight loss (DWL), a simple but effective loss function for radar echo extrapolation. The method adds model self-adjusted dynamic weights to the weighted loss function and structural similarity index measures. Radar echo extrapolation experiments are performed on four models, ConvLSTM, ConvGRU, PredRNN, and PredRNN++. Radar reflectivity is predicted using Nanjing University C-band Polarimetric (NJU-CPOL) weather radar data. The quantitative statistics show that using the DWL method reduces the MAE of the four models by up to 10.61%, 5.31%, 14.8%, and 13.63%, respectively, over a 1 h prediction period. The results show that the DWL approach is effective in reducing the accumulation of errors over time, improving the predictive performance of currently popular deep learning models.

The classification of land use information is important for land resource management. With the purpose of extracting precise spatial information, we present a novel land use classification model based on a mixed attention module and adjustable feature enhancement layer (MAAFEU-net). Our unique design, the mixed attention module, allows the model to concentrate on target-specific discriminative features and capture class-related features within different land use types. In addition, an adjustable feature enhancement layer is proposed to further enhance the classification ability of similar types. We assess the performance of this model using the publicly available GID dataset and the self-built Gwadar dataset. Six semantic segmentation deep networks are used for comparison. The experimental results show that the F1 score of MAAFEU-net is 2.16% and 2.3% higher than the next model and that MIoU is 3.15% and 3.62% higher than the next model. The results of the ablation experiments show that the mixed attention module improves the MIoU by 5.83% and the addition of the adjustable feature enhancement layer can further improve it by 5.58%. Both structures effectively improve the accuracy of the overall land use classification. The validation results show that MAAFEU-net can obtain land use classification images with high precision.

Drought is one of the major global natural disasters, and appropriate monitoring systems are essential to reveal drought trends. In this regard, deep learning is a very promising approach for characterizing the non-linear nature of drought factors. We used multi-source remote sensing data such as the Moderate Resolution Imaging Spectroradiometer (MODIS) and Climate Hazards Group Infrared Precipitation with Station (CHIRPS) data to integrate drought impact factors such as precipitation, vegetation, temperature, and soil moisture. The application of convolutional long short-term memory (ConvLSTM) to construct an integrated drought monitoring model was proposed and tested, using the Xinjiang Uygur Autonomous Region as an example. To better compare the monitoring performance of ConvLSTM models, three other classical deep learning models and three classical machine learning models were also used for comparison. The results show that the composite drought index (CDI) output by the ConvLSTM model had a consistent high correlation with the drought rating of the multi-scale standardized precipitation evapotranspiration index (SPEI). The correlation coefficients between the CDI and the multi-scale standardized precipitation index (SPI) were all above 0.5 (p < 0.01), which was highly significant, and the correlation coefficient between CDI-1 and the monthly soil relative humidity at a 10 cm depth was above 0.45 (p < 0.01), which was well correlated. In addition, the spatial distribution of the CDI-6 simulated by the model was highly correlated with the degree of drought expressed by the SPEI-6 observations at the stations. This study provides a new approach for integrated regional drought monitoring.

This study adopts a two-way approach to CORDEX-CORE RegCM4-7 seasonal precipitation simulations’ Added Value (AV) analysis over Africa, which aims to quantify potential improvements introduced by the downscaling approach at high and low resolution, using satellite-based observational products. The results show that RegCM4-7 does add value to its driving Global Climate Models (GCMs) with a positive Added Value Coverage (AVC) ranging between 20 and 60% at high resolution, depending on the season and the boundary conditions. At low resolution, the results indicate an increase in the positive AVC by up to 20% compared to the high-resolution results, with an up to 8% decrease for instances where an increase is not observed. Typical climate zones such as West Africa, Central Africa, and Southern East Africa, where improvements by Regional Climate Models (RCMs) are expected due to strong dependence on mesoscale and fine-scale features, show positive AVC greater than 20%, regardless of the season and the driving GCM. These findings provide more evidence for confirming the hypothesis that the RCMs AV is influenced by their internal physics rather than being the product of a mere disaggregation of large-scale features provided by GCMs. Although the results show some dependencies to the driving GCMs relating to their equilibrium climate sensitivity nature, the findings at low resolutions similar to the native GCM resolutions make the influence of internal physics more important. The findings also feature the CORDEX-CORE RegCM4-7 precipitation simulations’ potential in bridging the quality and resolution gap between coarse GCMs and high-resolution remote sensing datasets. Even if further post-processing activities, such as bias correction, may still be needed to remove persistent biases at high resolution, using upscaled RCMs as an alternative to GCMs for large-scale precipitation studies over Africa can be insightful if the AV and other performance statistics are satisfactory for the intended application.

In a warming climate, drought events are projected to increase in many regions across the world, which would have detrimental impacts on water resources for agriculture activity and human life. Thus, projecting drought changes, especially the frequency of future drought events, is very important for the African continent. This study investigates the future changes in drought events based on the France Centre National de Recherches Météorologiques (CNRM-CM6) model in the Coupled Model Intercomparison Project phase six (CMIP6) datasets for four shared socio-economic pathways (SSP): SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5; and three time slices: near future (2020–2039), mid-century (2050–2069), and end-of-century (2080–2099), relative to a historical baseline period (1995–2014). The interannual variability and trends of the self-calibrating Palmer Drought Severity Index (scPDSI) based on the Penman–Monteith methods for measuring potential evapotranspiration (PET) are used to estimate future droughts. The temporal analysis shows that the drought frequency, intensity, and affected area will increase throughout the 21st century. Among the scenarios, SSP3-7.0 and SSP5-8.5 project a larger upward trend in drought characteristics than SSP1-2.6 and SSP2-4.5. The spatial pattern shows drought frequency decreases in humid regions and increases in non-humid regions across Africa. For all SSP scenarios, the projected wetting trend per decade ranges from 0.05 to 0.25, while the drying trend per decade ranges from −0.05 to 0.25. A regional trend analysis revealed key differences in spatial pattern, with varied trend projections of wetter and drier conditions in humid and non-humid regions under all SSP scenarios. Drier conditions are expected to intensify in Southern Africa under all SSP scenarios but are projected to be more intense under either SSP3-7.0 and SSP5-8.5. In general, the projected wetter trends in humid areas may favor agricultural production and ecological conservation, and drier trends in non-humid regions may call for the possible adoption of tailor-made drought adaptation strategies and development programmes to minimize impacts.

The future planning and management of water resources ought to be based on climate change projections at relevant temporal and spatial scales. This work uses the new regional demarcation for Southern Africa (SA) to investigate the spatio-temporal precipitation variability and trends of centennial-scale observation and modeled data, based on datasets from the sixth phase of the Coupled Model Intercomparison Project (CMIP6). The study employs several statistical methods to rank the models according to their precipitation simulation ability. The Theil–Sen slope estimator is used to assess precipitation trends, with a Student’s t-test for the significance test. The comparison of observation and model historical data enables identification of the best-performing global climate models (GCMs), which are then employed in the projection analysis under two Shared Socioeconomic Pathways (SSPs): SSP2-4.5 and SSP5-8.5. The GCMs adequately capture the annual precipitation variation but with a general overestimation, especially over high-elevation areas. Most of the models fail to capture precipitation over the Lesotho-Eswatini area. The three best-performing GCMs over SA are FGOALS-g3, MPI-ESM1-2-HR and NorESM2-LM. The sub-regions demonstrate that precipitation trends cannot be generalized and that localized studies can provide more accurate findings. Overall, precipitation in the wet and dry seasons shows an initial increase during the near future over western and eastern SA, followed by a reduction in precipitation during the mid- and far future under both projection scenarios. Madagascar is expected to experience a decrease in precipitation amount throughout the twenty-first century.