Search

Abstract Humid heat waves pose significant risks to human health and the ecosystem. Intuitively, rainfall often alleviates extreme humid heat. However, here we show that light rain often accompanies extreme humid heat, exacerbating its frequency and intensity, especially over arid and semi-arid regions compared to no rain and moderate-to-heavy rain cases. This is because light rain does not dramatically reduce solar radiation but increases near-surface humidity through enhanced surface evaporation. The water replenishment from light rain as well as a shallower planetary boundary layer is crucial for consecutive extremes where there are commonly sporadic drizzle days amidst several rain-free days. These extremes last longer than rain-free extremes. Current global climate models (GCMs) overestimate light rain. After reducing this bias in a GCM, underestimations of humid heat waves in energy-limited regions and overestimations in water-limited regions are largely alleviated. These findings underscore the underappreciated impact of light rain on extreme humid heat.

Abstract Numerous Earth system models exhibit excessive aerosol effective forcing at the top of the atmosphere (TOA), including the Department of Energy's Energy Exascale Earth System Model (E3SM). Here, in the context of the E3SM version 3 effort, the predicted particle property (P3) stratiform cloud microphysics scheme and an enhanced deep convection parameterization suite (ZM_plus) are implemented into E3SM. The ZM_plus includes a convective cloud microphysics scheme, a multi‐scale coherent structure parameterization for mesoscale convective systems, and a revised cloud base mass flux formulation considering impacts of the large‐scale environment. The P3 scheme improved cloud and radiation particularly over the Northern Hemisphere and the frequency of heavy precipitation over the tropics, and the ZM_plus improved clouds in the tropics. P3 decreases aerosol effective forcing by 0.15 W m−2, while the ZM_plus increases it by 0.27 W m−2, resulting from excessive direct (0.31 W m−2) and indirect forcing (−1.79 W m−2). The excessive aerosol forcings are due to aerosol overestimation associated with insufficient aerosol wet removal. By improving the physical treatments in the aerosol wet removal, we effectively mitigate anthropogenic aerosol overestimation and thus attenuate direct (0.09 W m−2) and indirect aerosol forcing (−1.52 W m−2). Adjustment to primary organic matter hygroscopicity reduces direct and indirect forcing to more reasonable values: −0.13 W m−2 and −1.31 W m−2, respectively. On climatology, improved aerosol treatments mitigate overestimation of aerosol optical depth.

Abstract Current global climate models (GCMs), limited to grid‐scale land‐atmosphere coupling, cannot represent subgrid urban‐rural precipitation contrasts. This study develops an innovative two‐way subgrid land‐atmosphere coupling framework in the National Center for Atmospheric Research (NCAR) Community Earth System Model version 2 (CESM2) to explicitly resolve land‐atmosphere interaction over subgrid individual land units. Results show that urban heat island (UHI) leads to the urban rainfall effect (URE), which in turn alleviates overestimated UHI over China in CESM2. The URE manifests as a shift toward more heavy precipitation and less light precipitation in world urban areas than in surrounding rural counterparts. This feature is consistent with available observations. In heavy precipitation situations, the UHI promotes atmospheric instability and enhances atmospheric water vapor holding capacity, resulting in more heavy precipitation in urban areas. Conversely, in light precipitation situations, the UHI and decreased evaporation from urban impermeable surfaces diminish atmospheric relative humidity, suppressing light precipitation.

Trends in surface air temperature (SAT) are a common metric for global warming. Using observations and observationally driven models, we show that a more comprehensive metric for global warming and weather extremes is the trend in surface equivalent potential temperature (Thetae_sfc) since it also accounts for the increase in atmospheric humidity and latent energy. From 1980 to 2019, while SAT increased by 0.79[Formula: see text], Thetae_sfc increased by 1.48[Formula: see text] globally and as much as 4[Formula: see text] in the tropics. The increase in water vapor is responsible for the factor of 2 difference between SAT and Thetae_sfc trends. Thetae_sfc increased more uniformly (than SAT) between the midlatitudes of the southern hemisphere and the northern hemisphere, revealing the global nature of the heating added by greenhouse gases (GHGs). Trends in heat extremes and extreme precipitation are correlated strongly with the global/tropical trends in Thetae_sfc. The tropical amplification of Thetae_sfc is as large as the arctic amplification of SAT, accounting for the observed global positive trends in deep convection and a 20% increase in heat extremes. With unchecked GHG emissions, while SAT warming can reach 4.8[Formula: see text] by 2100, the global mean Thetae_sfc can increase by as much as 12[Formula: see text], with corresponding increases of 12[Formula: see text] (median) to 24[Formula: see text] (5% of grid points) in land surface temperature extremes, a 14- to 30-fold increase in frequency of heat extremes, a 40% increase in the energy available for tropical deep convection, and an up to 60% increase in extreme precipitation.

Abstract With the recent advances in data science, machine learning has been increasingly applied to convection and cloud parameterizations in global climate models (GCMs). This study extends the work of Han et al. (2020, https://doi.org/10.1029/2020MS002076) and uses an ensemble of 32‐layer deep convolutional residual neural networks, referred to as ResCu‐en, to emulate convection and cloud processes simulated by a superparameterized GCM, SPCAM. ResCu‐en predicts GCM grid‐scale temperature and moisture tendencies, and cloud liquid and ice water contents from moist physics processes. The surface rainfall is derived from the column‐integrated moisture tendency. The prediction uncertainty inherent in deep learning algorithms in emulating the moist physics is reduced by ensemble averaging. Results in 1‐year independent offline validation show that ResCu‐en has high prediction accuracy for all output variables, both in the current climate and in a warmer climate with +4K sea surface temperature. The analysis of different neural net configurations shows that the success to generalize in a warmer climate is attributed to convective memory and the 1‐dimensional convolution layers incorporated into ResCu‐en. We further implement a member of ResCu‐en into CAM5 with real world geography and run the neural‐network‐enabled CAM5 (NCAM) for 5 years without encountering any numerical integration instability. The simulation generally captures the global distribution of the mean precipitation, with a better simulation of precipitation intensity and diurnal cycle. However, there are large biases in temperature and moisture in high latitudes. These results highlight the importance of convective memory and demonstrate the potential for machine learning to enhance climate modeling.

Historical land use impacts climate by biogeophysical and biogeochemical effects. Their combined effects on mean and extreme temperature may harm economically disadvantaged countries but benefit those in rich countries, raising questions of equality.

We use 3-D cloud-resolving model (CRM) simulations of two mesoscale convective systems at mid-latitudes and a simple statistical ensemble method to diagnose the scale dependency of convective momentum transport (CMT) and CMT-related properties, and evaluate a parameterization scheme for convection-induced pressure gradient (CIPG) developed by Gregory et al. (GKI97). GKI97 relates CIPG to a constant coefficient multiplied by mass flux and vertical mean wind shear. CRM results show that mass fluxes and CMT exhibit strong scale dependency in temporal evolution and vertical structure. The prevalent understandings of CMT characteristics in terms of upgradient/downgradient transport are applicable to updrafts but not downdrafts across a wide range of grid spacings (4-512 km). For the small-to-median grid spacings (4~64 km), GKI97 reproduces some aspects of CIPG scale dependency except for underestimating the variations of CIPG as grid spacing decreases. However, for large grid spacings (128~512 km), GKI97 might even less adequately parameterize CIPG because it omits the contribution from either the nonlinear shear or buoyancy forcings. Further diagnosis of CRM results suggests that inclusion of nonlinear shear forcing in GKI97 is needed for the large grid spacings, and use of the three-updraft and one downdraft approach proposed in an earlier study may help a modified GKI97 capture more variations of CIPG as grid spacing decreases for the small-to-median grid spacings. Further, the optimal coefficients used in GKI97 seems insensitive to grid spacings, but they might be different for updrafts and downdrafts, for different MCS types, and for zonal and meridional components.

In this study, we evaluate mesoscale convective system (MCS) simulations in the second version of U.S. Department of Energy (DOE) Energy Exascale Earth System Model (E3SMv2). E3SMv2 atmosphere model (EAMv2) is run at the uniform 0.25° horizontal resolution. We track MCSs consistently in the model and observations using PyFLEXTRKR algorithm, which defines MCSs based on both cloud top brightness temperature (Tb) and surface precipitation. Results from using only Tb to define MCSs are also discussed to understand the impact of different MCS tracking algorithms on MCS evaluation and provide additional insights into model errors in simulating MCSs. Our results show that EAMv2 simulated MCS precipitation is largely underestimated in tropical and extratropical regions. This is mainly attributed to the underestimated MCS genesis and underestimated precipitation intensity in EAMv2. Comparing the two MCS tracking methods, simulated MCS precipitation is increased if MCSs are defined with only cloud top Tb. The Tb‐based MCS tracking method, however, includes cloud systems with very weak precipitation. This illustrates the model issues in simulating heavy precipitation even though the convective cloud shield is overall well simulated from the moist convective processes. Furthermore, sensitivity experiments are performed to examine the impact of new cloud and convection parameterizations developed for EAMv3 on simulated MCSs. The new physics parameterizations help increase the relative contribution of convective precipitation to total precipitation in the tropics, but the simulated MCS properties are not significantly improved. This suggests that simulating MCSs still remain a challenge for the next version of E3SM. Plain Language Summary: Mesoscale convective systems (MCSs) are one of the largest forms of deep convective storms, which play an important role in the earth system. It is imperative for global climate models to reasonably simulate MCS properties. This study aims to evaluate simulated MCS properties in the second version of U.S. Department of Energy (DOE) Energy Exascale Earth System Model (E3SMv2). We utilized two different approaches to define MCSs in the model and observations for consistent comparisons. Our results show that the E3SMv2 model underestimates MCS precipitation in the tropical and subtropical regions. The too few MCSs and overly weak precipitation intensity in individual MCSs are the primary reasons for this MCS precipitation bias. The simulated MCS precipitation becomes more comparable to the observations when surface precipitation is not included in the MCS definition. However, many cloud systems with very weak precipitation characteristics are included. This comparison illustrates the model issues in precipitation formation while convective cloud structures are overall well simulated. In addition, by examining the impact of new physics parameterizations that are developed for the next generation of E3SM model on the MCS simulation, we find simulating MCSs will remain a challenge for the next version of E3SM model. Key Points: Simulated mesoscale convective system (MCS) precipitation is substantially underestimated in E3SMv2 due to insufficient MCS genesis and rain rate in individual MCSsUtilizing different MCS tracking methods provides a more complete picture about the model capability in simulating MCSsMCS properties in E3SMv2 are not significantly improved with the new cloud and convection parameterizations developed for E3SMv3 [ABSTRACT FROM AUTHOR]

Numerous Earth system models exhibit excessive aerosol effective forcing at the top of the atmosphere (TOA), including the Department of Energy's Energy Exascale Earth System Model (E3SM). Here, in the context of the E3SM version 3 effort, the predicted particle property (P3) stratiform cloud microphysics scheme and an enhanced deep convection parameterization suite (ZM_plus) are implemented into E3SM. The ZM_plus includes a convective cloud microphysics scheme, a multi‐scale coherent structure parameterization for mesoscale convective systems, and a revised cloud base mass flux formulation considering impacts of the large‐scale environment. The P3 scheme improved cloud and radiation particularly over the Northern Hemisphere and the frequency of heavy precipitation over the tropics, and the ZM_plus improved clouds in the tropics. P3 decreases aerosol effective forcing by 0.15 W m−2, while the ZM_plus increases it by 0.27 W m−2, resulting from excessive direct (0.31 W m−2) and indirect forcing (−1.79 W m−2). The excessive aerosol forcings are due to aerosol overestimation associated with insufficient aerosol wet removal. By improving the physical treatments in the aerosol wet removal, we effectively mitigate anthropogenic aerosol overestimation and thus attenuate direct (0.09 W m−2) and indirect aerosol forcing (−1.52 W m−2). Adjustment to primary organic matter hygroscopicity reduces direct and indirect forcing to more reasonable values: −0.13 W m−2 and −1.31 W m−2, respectively. On climatology, improved aerosol treatments mitigate overestimation of aerosol optical depth. Plain Language Summary: The Energy Exascale Earth System Model (E3SM) exhibits strong direct and indirect aerosol forcings, after advanced stratiform and convective cloud treatments are implemented. This study identifies the primary cause of these excessive aerosol forcings as the significant overestimation of anthropogenic aerosols due to insufficient removal of aerosols by precipitation. To address this issue, we made aerosol wet removal representation more physical, which effectively reduced the overestimation of aerosols, bringing both direct and indirect forcings to the expected ranges. Furthermore, cloud and aerosol climatology are notably improved as the result of these developments in cloud and aerosol treatments. Key Points: The new cloud microphysics scheme P3 improves simulations of cloud properties in the NH and aerosol forcingInsufficient wet removal leads to an overly strong aerosol forcing after incorporating the enhanced deep convection parameterization suiteThe aerosol wet removal improvements lead to more reasonable aerosol forcing and improved aerosol climatology in Energy Exascale Earth System Model [ABSTRACT FROM AUTHOR]

Current global climate models (GCMs), limited to grid‐scale land‐atmosphere coupling, cannot represent subgrid urban‐rural precipitation contrasts. This study develops an innovative two‐way subgrid land‐atmosphere coupling framework in the National Center for Atmospheric Research (NCAR) Community Earth System Model version 2 (CESM2) to explicitly resolve land‐atmosphere interaction over subgrid individual land units. Results show that urban heat island (UHI) leads to the urban rainfall effect (URE), which in turn alleviates overestimated UHI over China in CESM2. The URE manifests as a shift toward more heavy precipitation and less light precipitation in world urban areas than in surrounding rural counterparts. This feature is consistent with available observations. In heavy precipitation situations, the UHI promotes atmospheric instability and enhances atmospheric water vapor holding capacity, resulting in more heavy precipitation in urban areas. Conversely, in light precipitation situations, the UHI and decreased evaporation from urban impermeable surfaces diminish atmospheric relative humidity, suppressing light precipitation. Plain Language Summary: Given the grid‐scale land‐atmosphere coupling (operating at a scale of ∼100–200 km), all the current global climate models cannot represent precipitation over urban areas. This obstructs the explicit simulation and future projections of precipitation in world urban regions, where over half of the world's population resides and a substantial portion of economic activities occurs. To address this issue, this study develops a novel land‐atmosphere coupling framework that explicitly allows land‐atmosphere interaction over individual land units within the grid cell in global climate models. The modeling results show a greater daily average of precipitation intensities on precipitation days in urban areas than in rural areas due to urban warming, with a shift toward more heavy precipitation and less light precipitation. These findings indicate an escalation in the risk of extreme precipitation and flooding in urban regions around the globe, as urban land continues to expand. Key Points: A novel framework of two‐way subgrid land‐atmosphere coupling for GCMs is developed and implemented in CESM2Overestimated urban heat island over China in CESM2 is significantly alleviated by explicitly representing urban rainfall effectsWorld urban regions feature more heavy precipitation and less light precipitation than the surrounding rural areas [ABSTRACT FROM AUTHOR]

Abstract We propose a parameterization scheme of convective organization effects based on a moisture‐distribution approach, which can reflect aggregation of convective cells within a model grid as well as the interaction between convection and spatial heterogeneity in free‐troposphere moisture. With this concept, convective cells in an aggregated state are surrounded by air that is moister than the grid‐mean condition, which provides a “shielding” effect favorable for deeper convective updrafts. Such effects are represented quantitatively via utilizing a schematic diagram about the geometric interpretation of convective cluster, dry area, and their contact area, in which both the convective cluster size and dry area fraction increase during the development of convection. Our sensitivity analyses indicate that the new scheme performs well in capturing the mean precipitation features. Excluding the convective organization effects in the model leads to a considerable reduction in the simulated precipitation magnitude. Notably, there exists an inverse relationship between background mean moisture condition and sub‐grid moisture variability for a given rain rate, suggesting that with a high spatial heterogeneity in free‐troposphere moisture, convection can still develop and maintain its strength under relatively dry background conditions, consistent with the convection‐permitting model simulation. Overall, our new parameterization of convective organization effects can successfully reproduce the relationship between precipitation and sub‐grid moisture variability, which is a missing element in traditional convection parameterization schemes but important for the simulations of precipitation variability at various scales in climate models.

This research is aimed at predicting the physical stability for amorphous solid dispersion by utilizing deep learning methods. We propose a prediction model that effectively learns from a small dataset that is imbalanced in terms of class. In order to overcome the imbalance problem, our model performs a hybrid sampling which combines synthetic minority oversampling technique (SMOTE) algorithm with edited nearest neighbor (ENN) algorithm and reduces the dimensionality of the dataset using principal component analysis (PCA) algorithm during data preprocessing. After the preprocessing, it performs the learning process using a carefully designed neural network of simple but effective structure. Experimental results show that the proposed model has faster training convergence speed and better test performance compared to the existing DNN model. Furthermore, it significantly reduces the computational complexity of both training and test processes.

Abstract We revise the convective triggering function in Department of Energy's Energy Exascale Earth System Model (E3SM) Atmosphere Model version 1 (EAMv1) by introducing a dynamic constraint on the initiation of convection that emulates the collective dynamical effects to prevent convection from being triggered too frequently and allowing air parcels to launch above the boundary layer to capture nocturnal elevated convection. The former is referred to as the dynamic Convective Available Potential Energy (dCAPE) trigger and the latter as the Unrestricted Launch Level (ULL) trigger. Compared to the original trigger in EAMv1 that initiates convection whenever CAPE is larger than a threshold, the revised trigger substantially improves the simulated diurnal cycle of precipitation over both midlatitude and tropical lands. The nocturnal peak of precipitation and the eastward propagation of convection downstream of the Rockies and over the adjacent Great Plains are much better captured than those in the default model. The overall impact on mean precipitation is minor with some notable improvements over the Indo‐Western Pacific, subtropical Pacific and Atlantic, and South America. In general, the dCAPE trigger helps to better capture late afternoon rainfall peak, while ULL is key to capturing nocturnal elevated convection and the eastward propagation of convection. The dCAPE trigger also primarily contributes to the considerable reduction of convective precipitation over subtropical regions and the frequency of light‐to‐moderate precipitation occurrence. However, no clear improvement is seen in intense convection and the amplitude of diurnal precipitation.

The scale-aware convective parameterization for high resolution global climate models must satisfy the requirement that the parameterized subgrid convective transport diminishes as the model resolution increases to convection-resolving resolutions. A major assumption in current scale-aware convection schemes is that the differences between convective cloud properties and their environmental counterparts are independent of cloud fraction. This study examines convective cloud vertical velocity, moist static energy (MSE), moisture, and the vertical eddy transport of MSE and moisture for different averaging subdomain sizes and fractional convective cloudiness using a cloud resolving model simulation of a midlatitude mesoscale convective system. Results show that convective cloud fraction, mass flux, and vertical transport of MSE and moisture increase with decreasing subdomain size. The differences between convective cloud properties in both updrafts and downdrafts and their environment depend on both cloud fraction and the averaging subdomain size. For a given subdomain size, the differences increase with cloud fraction, in contrast to the assumption used in current scale-aware convection parameterization schemes. A consequence of this is that the parameterized convective eddy transport reaches maximum at a higher cloud fraction than believed in previous studies. This has implications on how fast the subgrid convective transport should diminish as GCM resolution increases.

Abstract We modify the Zhang‐McFarlane (ZM) deep convection scheme in the Community Atmosphere Model version 5 to couple it with a unified parameterization for boundary‐layer turbulence and shallow convection, that is, Cloud Layers Unified by Binormals (CLUBB). By assuming a lognormal distribution of entrainment rate across the entire moist convective regimes, we link mass fluxes between shallow and deep convection, which are partitioned by the entrainment rate of the shallowest deep convective plume. Hence, a new deep convective closure is established which is coupled to the sub‐grid vertical motion variability in CLUBB. The convection feedback (or memory) effects are also considered to decrease the entrainment spectrum width and enhance the vertical velocity variability that further affect deep convection. Results show that the revised scheme improves the precipitation simulations in terms of the mean state and variability at various timescales, such as the alleviated double‐intertropical convergence zone and more realistic simulations of the seasonal variation of monsoon precipitation over East Asia, Madden‐Julian Oscillation, and precipitation diurnal phase propagations downstream of large terrains. The improvements are still seen in many aspects such as the mean‐state precipitation when turning off the convection feedback impacts in the revised scheme, emphasizing the benefits of using the modified mass‐flux closure. However, the convection feedbacks have considerable effects on the precipitation diurnal cycle simulations over regions with late‐afternoon precipitation peaks. Overall, the revised scheme provides a unified treatment for sub‐grid vertical motions across regimes of boundary‐layer turbulence, shallow convection, and deep convection, leading to better‐simulated precipitation at various timescales.

Abstract Current moist physics parameterization schemes in general circulation models (GCMs) are the main source of biases in simulated precipitation and atmospheric circulation. Recent advances in machine learning make it possible to explore data‐driven approaches to developing parameterization for moist physics processes such as convection and clouds. This study aims to develop a new moist physics parameterization scheme based on deep learning. We use a residual convolutional neural network (ResNet) for this purpose. It is trained with 1‐year simulation from a superparameterized GCM, SPCAM. An independent year of SPCAM simulation is used for evaluation. In the design of the neural network, referred to as ResCu, the moist static energy conservation during moist processes is considered. In addition, the past history of the atmospheric states, convection, and clouds is also considered. The predicted variables from the neural network are GCM grid‐scale heating and drying rates by convection and clouds, and cloud liquid and ice water contents. Precipitation is derived from predicted moisture tendency. In the independent data test, ResCu can accurately reproduce the SPCAM simulation in both time mean and temporal variance. Comparison with other neural networks demonstrates the superior performance of ResNet architecture. ResCu is further tested in a single‐column model for both continental midlatitude warm season convection and tropical monsoonal convection. In both cases, it simulates the timing and intensity of convective events well. In the prognostic test of tropical convection case, the simulated temperature and moisture biases with ResCu are smaller than those using conventional convection and cloud parameterizations.

Abstract In order to improve the physical consistency between shallow and deep convection, we modify the deep convective closure in the Community Atmosphere Model version 5 (CAM5) coupled with a third‐order turbulence closure parameterization (i.e., Cloud Layers Unified by Binormals [CLUBB]). The revised closure reserves a portion of the total convective available potential energy for shallow convection via utilizing the heating and moistening profiles from CLUBB to distribute moisture and energy between shallow and deep convection. Simulations at two resolutions (i.e., 2° and 0.5°, respectively) are conducted to investigate the impacts of convective closure on the simulated precipitation diurnal variations. Results from low‐resolution simulations show that the revised closure suppresses deep convection until the lower troposphere is sufficiently moistened by shallow convection, which improves the precipitation diurnal variation simulations compared with the default closure, with the precipitation diurnal peak over tropical lands delayed from 12LST to 19LST. The revised closure better simulates the diurnal variations for precipitation over the Asian monsoon region, such as the delayed precipitation onset, but still fails to well capture the nighttime peak for precipitation there. This deficiency is alleviated to some extent when applying the revised closure in high‐resolution simulations, but nighttime precipitation is still underestimated probably because key processes responsible for nighttime convection are missing. Overall, our results indicate that establishing the consistency between shallow and deep convection is critical for the precipitation diurnal cycle simulations.

Vonoprazan (VPZ) is the first-in-class potassium-competitive acid blocker (P-CAB), and has many advantages over proton pump inhibitors (PPIs). It is administered as a fumarate salt for the treatment of acid-related diseases, including reflux esophagitis, gastric ulcer, and duodenal ulcer, and for eradication of Helicobacter pylori. To discover novel cocrystals of VPZ, we adopted an artificial neural network (ANN)-based machine learning model as a virtual screening tool that can guide selection of the most promising coformers for VPZ cocrystals. Experimental screening by liquid-assisted grinding (LAG) confirmed that 8 of 19 coformers selected by the ANN model were likely to create new solid forms with VPZ. Structurally similar benzenediols and benzenetriols, i.e., catechol (CAT), resorcinol (RES), hydroquinone (HYQ), and pyrogallol (GAL), were used as coformers to obtain phase pure cocrystals with VPZ by reaction crystallization. We successfully prepared and characterized three novel cocrystals: VPZ–RES, VPZ–CAT, and VPZ–GAL. VPZ–RES had the highest solubility among the novel cocrystals studied here, and was even more soluble than the commercially available fumarate salt of VPZ in solution at pH 6.8. In addition, novel VPZ cocrystals had superior stability in aqueous media than VPZ fumarates, demonstrating their potential for improved pharmaceutical performance.

In recent research on the formulation prediction of oral dissolving drugs, deep learning models with significantly improved performance compared to machine learning models were proposed. However, the performance degradation due to limitations of an imbalanced dataset with a small size and inefficient neural network structure has still not been resolved. Therefore, we propose new deep learning-based prediction models that maximize the prediction performance for disintegration time of oral fast disintegrating films (OFDF) and cumulative dissolution profiles of sustained-release matrix tablets (SRMT). In the case of OFDF, we use principal component analysis (PCA) to reduce the dimensionality of the dataset, thereby improving the prediction performance and reducing the training time. In the case of SRMT, the Wasserstein generative adversarial network (WGAN), a neural network-based generative model, is used to overcome the limitation of performance improvement due to the lack of experimental data. To the best of our knowledge, this is the first work that utilizes WGAN for pharmaceutical formulation prediction. Experimental results show that the proposed methods have superior performance than the existing methods for all the performance metrics considered.

Malaria remains by far one of the most threatening and dangerous illnesses caused by the plasmodium falciparum parasite. Chloroquine (CQ) and first-line artemisinin-based combination treatment (ACT) have long been the drug of choice for the treatment and controlling of malaria; however, the emergence of CQ-resistant and artemisinin resistance parasites is now present in most areas where malaria is endemic. In this work, we developed five machine learning models to predict antimalarial bioactivities of a drug against plasmodium falciparum from the features (i.e., molecular descriptors values) obtained from PaDEL software from SMILES of compounds and compare the machine learning models by experiments with our collected data of 4794 instances. As a consequence, we found that three models amongst the five, namely artificial neural network (ANN), extreme gradient boost (XGB), and random forest (RF), outperform the others in terms of accuracy while observing that, using roughly a quarter of the promising descriptors picked by the feature selection algorithm, the five models achieved equivalent and comparable performance. Nevertheless, the contribution of all molecular descriptors in the models was investigated through the comparison of their rank values by the feature selection algorithm and found that the most potent and relevant descriptors which come from the ‘Autocorrelation’ module contributed more while the ‘Atom type electrotopological state’ contributed the least to the model.

Diurnal precipitation is a fundamental mode of variability that climate models have difficulty in accurately simulating. Here the diurnal cycle of precipitation (DCP) in participating climate models from the Global Energy and Water Exchanges' DCP project is evaluated over the tropics and central United States. Common model biases such as excessive precipitation over the tropics, too frequent light‐to‐moderate rain, and the failure to capture propagating convection in the central United States still exist. Over the central United States, the issues of too weak rainfall intensity in climate runs is well improved in their hindcast runs with initial conditions from numerical weather prediction analyses. But the improvement is minimal over the central Amazon. Incorporating the role of the large‐scale environment in convective triggering processes helps resolve the phase‐locking issue in many models where precipitation often incorrectly peaks near noon due to maximum insolation over land. Allowing air parcels to be lifted above the boundary layer improves the simulation of nocturnal precipitation which is often associated with the propagation of mesoscale systems. Including convective memory in cumulus parameterizations acts to suppress light‐to‐moderate rain and promote intense rainfall; however, it also weakens the diurnal variability. Simply increasing model resolution (with cumulus parameterizations still used) cannot fully resolve the biases of low‐resolution climate models in DCP. The hierarchy modeling framework from this study is useful for identifying the missing physics in models and testing new development of model convective processes over different convective regimes. [ABSTRACT FROM AUTHOR]