Atmospheric photochemistry is essential for simulating atmospheric composition, impacting air quality and climate change. However, conventional numerical schemes of photochemistry within atmospheric models are computationally expensive, leading to simplifications or omissions of critical processes in weather and climate models. Previous attempts to leverage artificial intelligence (AI) scheme to reduce computational costs have faced obstacles such as the curse of dimensionality and error propagation, and most have been limited to box models without coupling into numerical models. Here, we develop an innovative AI PhotoChemistry (AIPC) scheme coupled into an atmospheric model (WRF-Chem). With Multi-Head Self-Attention algorithm (MHSA), we simulate 74 chemical species and 229 reactions following the SAPRC-99 mechanism. This marks the first implementation of a sophisticated photochemical mechanism within one unified AI model, enabling fast, accurate, and stable simulations without needing individual AI model for each species as previous works. Comparative analysis reveals that the AIPC scheme outperforms previous AI schemes using Multi-Layer Perceptron and Residual Neural Network algorithms, offering superior accuracy and computational efficiency. Moreover, fine-tuning learning rate and broadening network width within the MHSA algorithm are more effective for improving the AIPC scheme’s performance than adjusting batch size or increasing network depth. When coupling AIPC into WRF-Chem, this hybrid model with both physics and AI schemes reproduces the spatiotemporal distributions of various species on monthly time scale, and achieves substantial speed enhancement with ~8 times faster than conventional scheme. This advancement lays the groundwork for future development of weather and climate models with sophisticated chemical processes.

Moisture transport in summer induces annual precipitation peak over the Tibetan Plateau (TP) thus being one crucial sustentation of water cycle between the TP and its surrounding areas. Simulating moisture transport accurately over the TP remains uncertain for current numerical models with one important influencing factor as horizontal resolution. In this study, in order to investigate the difference in moisture transport at resolutions from hydrostatic to non-hydrostatic scales, three experiments are conducted for summer of 2015 using a global variable-resolution model, including one with a globally quasi-uniform resolution of 60 km (U60km) and two with regional refinements over the TP at resolutions of 16 km (V16km) and 4 km (V4km), respectively. The differences in moisture transport among three simulations are significantly influenced by the changes in wind fields through the Himalayas and eastern TP at two layers, 700~600 and 600~400 hPa, which is largely modulated by their difference in large-scale circulations particularly monsoon depression. At hydrostatic scale (from 60 km to 16 km), the monsoon depression is slightly stronger and shifts northward along with the mid-latitude westerlies, which is due to the combination of the sensitivity of convection scheme to integrating timestep and different extents of resolved dynamical processes at different resolutions. With horizontal resolution increasing to convection-permitting scale (from 16 km to 4 km), the resolved moist convection along with its associated less latent heat leads to weaker monsoon depression over the south of TP, which is much larger than the resolution induced difference at hydrostatic scale.

Mars, characterized as a “desert” planet with little water vapor, primarily relies on dry deposition processes for dust removal. Although dry deposition processes include gravitational sedimentation, turbulent transfer, Brownian diffusion, impaction, and interception, gravitational sedimentation is considered the only way for dust removal in most current models. To have a more comprehensive understanding of the effects of Martian dust removal processes, a physics-based scheme of dry deposition processes (e.g., turbulent transfer, Brownian diffusion, impaction, and interception) with resolved dust particle sizes representing the lifting dust size distribution is implemented in the MarsWRF general circulation model in this study. The model results reveal that the dry deposition velocity increases significantly with the decrease in dust size, especially for small dust particles. This enhancement in the removal efficiency of small particles leads to an increase in the effective particle radius of airborne dust and a decrease in dust opacity, particularly in the high latitudes of the northern hemisphere during the period of high dust loading. In these latitudes, the atmospheric temperature rises from the surface up to an altitude of 55 km, with a peak temperature difference of about 3.8 K, driven by dynamical warming from the strengthened descending branch of the upper meridional circulation. In addition, the sublimation of CO2 surface ice in the high latitudes of the northern hemisphere is increased, and the condensation of the gas phase is decreased.

MarsWRF, the general circulation model of Mars, is one of the most commonly used models to study the dust cycle in the Martian atmosphere. It has been widely used to study the mechanisms of dust storms and their effects on the Martian atmosphere. To better understand the ability of MarsWRF to simulate the dust cycle on Mars, this study assesses the current dust lifting schemes in the model, specifically the convective lifting and wind stress schemes. It is found that, by tuning lifting efficiency, the model with the convective lifting scheme can generally reproduce the seasonal variation of the mid-level atmospheric temperature (T15) but cannot reproduce the observed spatial distribution of dust devils, which exhibits non-homogeneous (uniform) distribution in the northern (southern) hemisphere. The model with the wind stress lifting scheme can generally capture the observed magnitude of T15 and column dust optical depth (CDOD) with properly tuned lifting efficiency and threshold drag velocity. There is a discrepancy in the assessment of modeling seasonal variations of dust with T15 and CDOD, which may be partly due to the observational uncertainties related to T15 and CDOD and the empirical modeling methods of Martian dust optical properties and radiative effect. For the spatial distribution of dust, there are significant simulation biases regardless of the tuning, which may be caused by the biases in the dust lifting process and large-scale atmospheric circulation. The analysis highlights that dust lifting schemes need further improvement to better represent the dust cycle and their impacts on Mars.

The dust aerosols are a major type of aerosol over the Tibetan Plateau (TP) and influence climate at local to regional scales through their effects on thermal radiation and snow-albedo feedback. Based on the Modern-Era Retrospective Analysis for Research and Applications, Version 2 (MERRA-2) aerosol dataset, we report an increase of 34% in the atmospheric dust in the high troposphere over the TP during the spring season in the 2000s in comparison to the 1990s. This result is supported by an increase of 157% (46%) in the dust deposition flux in the Mugagangqiong (Tanggula) ice cores and an increase of 69% in the Aerosol Index (AI) from Earth Probe (EP) Total Ozone Mapping Spectrometer (TOMS), as well as by increases of simulated dust aerosols over the TP derived from the Community Earth System Model (CESM) and models from the Coupled Model Intercomparison Project Phase 6 (CMIP6). The increased atmospheric dust over the TP is caused in two aspects: (1) there was a higher dust emission over the Middle East during the 2000s than during the 1990s, which is explained by less precipitation and 25.8% higher in cyclone frequency over the Middle East. The increased cyclones uplift more dust from the surface over the Middle East to the central Asia in the middle troposphere. (2) Enhanced mid-latitude zonal winds help transport more dust in the middle troposphere from the central Asia to the Northwest China and thereafter an increase in northerly winds over Northwest China propels dust southward to the TP.