Search

Recent advancements in artificial intelligence (AI) have notably enhanced global weather forecasting, yet accurately predicting typhoon intensity remains challenging. This is largely due to constraints inherent in regression algorithm properties including deep neural networks and inability of coarse resolution to capture the finer-scale weather processes. To address these insufficiencies in typhoon intensity forecasting, we propose an attractive approach by initiating regional Weather Research and Forecasting (WRF) model with Pangu-weather, a state-of-the-art AI weather forecasting system (AI-Driven WRF), whose forecasting power can be further augmented by the implementation of dynamic vortex initialization. The results highlight limitations in Pangu-Weather's capability to accurately forecast typhoon intensity. In contrast, the AI-Driven WRF model demonstrated notable advancements over Pangu-Weather, achieving more reliable and accurate predictions of typhoon intensity. Furthermore, the AI-Driven WRF model demonstrated promising results in predicting typhoon intensity and wind details, showing commendable performance to traditional global numerical model-driven WRF models. Our analysis underscores the potential of AI weather forecasting models as a viable alternative for driving regional models, suggesting a promising avenue for future research in meteorology. [ABSTRACT FROM AUTHOR]

Given the important role of Atmospheric River precipitation (ARP) in the global hydrological cycle, accurate representation of ARP is significant. However, general circulation models (GCMs) demonstrate bias in simulating ARP. The target of this study is to quantify the performance of ARP intensity/frequency for CMIP6 simulations, and further to improve ARP estimation of CMIP6 using Cycle‐Consistent Generative Adversarial Networks (CycleGAN) with highlighting the more accurate ARP features under the global warming background. The findings of this study are as follows: (a) although ARP intensity/frequency in reserved‐optimal CMIP6 overall reproduces the observation, it is still underestimated at the stronger Atmospheric river (AR) scales, particularly for the AR highly active mid‐latitude regions. (b) The CycleGAN‐based bias correction approach markedly diminishes the bias of the CMIP6 simulations within most of the AR scales among both global and the four AR highly active regions. Moreover, the performance of the ARP in AR highly active regions is significant improvement, which is mainly due to the reduction of the bias at the strongest scale. (c) Relative to reference period (1986–2005), ARP intensity/frequency at the strongest scale increase notably under 3°C warming level, with an average value of 373.3% in intensity and 415.9% in frequency for global and the four key regions before correction, and the value is 451.9% and 492.5% after bias correction. The results illustrate that CycleGAN can effectively improve the ARP simulations of GCMs, and an early warning implies that future strong extreme ARP should potentially surpass the current expected. Plain Language Summary: Despite the important role of Atmospheric River precipitation (ARP) in global hydrological cycle, accurately simulating global ARP remains a great challenge. Here, the content of our study consists of three main parts. Firstly, we quantify the performance of CMIP6 general circulation models (GCMs) on ARP intensity/frequency simulations and find that CMIP6 GCMs underestimates ARP at the Atmospheric river (AR) highly active regions, especially in the strongest AR intensity scale. Secondly, we utilize a deep learning method named CycleGAN to improve global ARP intensity/frequency simulations of CMIP6 GCMs. The results indicate that CycleGAN effectively corrects the ARP intensity/frequency bias within most of the AR intensity scales among both global and the AR highly active regions. Moreover, the performance of ARP in the AR highly active regions is significant improvement, which is mainly due to the reduction of the bias in the strongest scale. Finally, the trained CycleGAN model is used to evaluate ARP under global warming of 1.5, 2, and 3°. The bias correction results show that ARP in the strongest AR intensity scale surpasses the current expected, especially under 3° warming level. Key Points: CMIP6 models underestimates Atmospheric River precipitation (ARP) at the Atmospheric river (AR) highly active regions, especially in the strongest AR scaleThe CycleGAN‐based deep learning method reduces model bias notably for ARP intensity and frequency, respectivelyARP in the strongest AR scale potentially surpasses the current GCM‐based projections expected in different degrees of warming after bias correction [ABSTRACT FROM AUTHOR]

Cloud‐related latent heat is critical to the development and maintenance of atmospheric circulation. However, it is unclear how sensitive the key large‐scale weather systems of an extreme rainfall event could be to the variation in cloud‐related latent heat. In a companion study, by investigating the indirect effects of different microphysics schemes, we hypothesized that the latent heat of condensation (CT) would play a role in the ability to simulate the extreme rainfall event that took place in Henan Province, central China, on July 19–21, 2021. To examine the proposed hypothesis, the impacts of varying the CT and the collection of cloud water by rainwater (ACW) in the Thompson, WDM6, and Morrison microphysics schemes on the simulation of the southerly flow and its indirect effect on this extreme rainfall case were studied using the Advanced Research Weather Research and Forecasting (ARW‐WRF) model. It was found that directly changing the CT and ACW had significant effects on the southerly flow and precipitation. An increase in CT generally led to a reduced temperature gradient, a smaller pressure gradient, a stronger southerly flow, and a precipitation surge of 32%–50%. Similar outcomes resulted from increasing ACW in the WDM6 scheme, which effectively diminished the cloud mixing ratio, generated higher CT, and consequently intensified the southerly flow and precipitation by 31%. This study confirmed that the accurate simulation of microphysical processes, along with latent heating, is desirable for more realistically reproducing the key weather systems in the simulation of extreme rainfall. [ABSTRACT FROM AUTHOR]

Tropical cyclones (TCs) constitute one of the major types of meteorological disasters in China, causing extensive damage. Based on the latest quality‐controlled provincial‐level TC disaster losses data over China from 2001 to 2020, the spatiotemporal variation in TC disaster and its link with TC activity in the form of prevailing track, precipitation, and intensity were investigated. Results showed that the areas impacted by TCs expanded significantly in the past decade (2011–2020) compared with the previous decade (2001–2010) throughout China, while the direct economic losses resulting from TCs did not increase significantly. Regionally, TC disaster losses increased in Northeast China but decreased in Fujian province. Causal analysis indicated that the increased influential TC (ITC) frequency and TC precipitation together led to the increase in damage for northern China. Also, more westward ITCs resulted in the increase in TC disasters for southwestern China. For southeastern China, the reduction in TC disasters was related to the northward shift in the northwestward ITC track. Although TC disasters increase with exposure, they are heavily influenced by TC track, precipitation, and intensity. Our findings suggest that more attention should be paid to TC prediction and early warning in northern China, especially Northeast China, where TC disasters are growing but there is greater vulnerability and insufficient capacity for TC prevention and mitigation. Plain Language Summary: China is one of the most frequently affected countries by tropical cyclones (TCs) in the world, suffering huge losses in term of human life and property. The average number of influential TCs varies little from year to year, but TC disasters vary broadly. These TC disasters involve not only the features of TCs themselves, such as precipitation, winds, and storm surges, but also socioeconomic development, the capacity of disaster prevention and preparedness. Utilizing the latest data on TC disasters, this study's results reveal that TC disasters have been aggravated in low‐impact areas, such as Northeast China. Understanding these changes in the characteristics of TC disasters is essential for TC disaster mitigation and climate change adaptation. Key Points: Tropical cyclone (TC) impacted areas expanded significantly in the past decade (2011–2020) compared with the previous decade (2001–2010) throughout ChinaTC disasters increased in Northeast China, where there is greater vulnerability and insufficient capacity for TC prevention and mitigationTC disasters are heavily influenced by TC track, precipitation, and intensity [ABSTRACT FROM AUTHOR]

Previous numerical studies have focused on the direct impact of microphysics schemes on multiscale atmospheric systems that produce rainfall. However, the indirect impact of microphysics schemes on the key large‐scale circulation around areas of extreme rainfall has not yet been examined systematically. We used the ARW‐WRF (Advanced Research WRF (Weather Research and Forecasting) Model) to simulate the extreme rainfall event in Henan province, China from 19 to 21 July 2021. Experiments were conducted to investigate the sensitivity of the simulation to the three popular double‐moment microphysics parameterizations: the Thompson, Morrison and WDM6 (WRF Double‐Moment 6‐Class Microphysics Schemes). We found significant sensitivity to the microphysics parameterization, with the maximum precipitation varying by up to 400 mm and the area‐averaged precipitation by 33 mm. The Thompson and Morrison microphysics schemes produced the largest amount of precipitation, whereas the WDM6 scheme produced the smallest amount of precipitation. The simulated southerly flow varied substantially between the different microphysics schemes. This is due to the ability of the microphysics schemes to produce latent heat, which enhances the southerly flow, leading to more intense precipitation. The higher the parameterized latent heat, the stronger and more obvious the southerly flow over southern Henan province, with the WDM6 scheme simulating a much weaker southerly flow and the Thompson scheme producing the strongest southerly flow. These results indicate that the indirect effects on the key circulation patterns can be just as sensitive to the formulation of the microphysics scheme as the direct effects on extreme rainfall. These results may help to improve the prediction of extreme rainfall events. Plain Language Summary: Complex microphysical processes have crucial impacts on the accurate prediction of the intensity and areal extent of extreme rainfall. However, previous studies have focused on rain amount related to the microphysical schemes caused by intrinsic microphysical processes or direct influencing systems—such as typhoons, squall lines and convective systems—and little attention has been paid to increase of rain amount related to the microphysical schemes caused by the surrounding key circulation of extreme rainfall. Through numerical simulations of the extreme rainfall event in Henan province, China from 19 to 21 July 2021 using three popular microphysics schemes, we show that indirect effect of microphysical schemes on the extreme rainfall can effectively regulate the southerly flow over southern China. The microphysics schemes influence the southerly flow through latent heating. A greater amount of latent heat can produce a stronger southerly flow, which favors the transport of water vapor and energy to the Henan region and produce more intense precipitation. Key Points: The simulated southerly flow and precipitation vary substantially between different microphysics schemesThe microphysics schemes influence the southerly flow through latent heatingThe Thompson and Morrison schemes produce more latent heat, and therefore a stronger southerly flow, than the WDM6 (WRF Double‐Moment 6‐Class Microphysics scheme) [ABSTRACT FROM AUTHOR]