Cumulus Parameterization Research Articles

WRF prediction of an atmospheric river‐related precipitation event: Sensitivity to cumulus parameterization schemes

AbstractThe present study aimed to evaluate performance sensitivity to the Cumulus Parameterization Scheme (CPS) used for the Weather Research and Forecasting (WRF) model to predict the atmospheric river‐related precipitation (ARP) event with 206 and 57 mm, highest and area‐averaged precipitation (AAP) per 24‐h, respectively, that occurred over the central mountainous basins of Iran on 31 March 2019. For this purpose, experiments were designed using the 12 (almost all) CPSs available in WRF v4. To verify the predicted precipitation (from the inner 4‐km domain), both point‐scale and grid‐scale comparisons were performed against gauge‐ and satellite‐based observational data at three accumulation time‐scales (12‐, 18‐, and 24‐h) and in three distinct sub‐regions. All scores obtained from the different statistical metrics used, are in complete agreement with a strongly dependent performance of WRF on the CPS used. In addition, the use of Kain‐Fritsch, KF‐CuP, and Grell‐3 CPSs could provide a realistic picture of impending heavy precipitation for WRF. Contrary, the New SAS, Tiedtke, and Zhang‐McFarlane CPSs did not perform satisfactorily in predicting the ARP event. As a result, CPSs with the “momentum transport” option in their modification mechanism are unlikely to adequately simulate the conversion of incoming low‐level moisture from atmospheric river to precipitation. However, precipitation predictions are more accurate at the 24‐h accumulation time‐scale than at the 12‐ and 18‐h. Also, a dry bias in the predictions is expected as the terrain elevation and accumulation time‐scale decrease and the distance from the core of the precipitation field increases.

Sensitivity of Different Physical Schemes in WRF Model of a Rainfall Event in Baghdad Station

The Weather Research and Forecasting model (WRF) offers a number of physical options that let users modify it to different scales, regions, and applications. The aim of this study is to test the sensitivity of different physics schemes in the WRF model for rainfall events over Iraq. In this study, six different physics configurations of the climate version of WRF were evaluated for simulation of a rainfall event in Iraq. Possible combinations among two Planetary Boundary Layers (PBL), three Cumulus (CUM) and two Microphysics (MIC) schemes were tested. The study area is the region surrounded by the longitudes 35o E-55o E and latitudes 290o N–38o N, which typically includes the Iraq region. The WRF model is installed on a Linux platform with a 10 km grid size in the zonal and meridional directions. For the six different simulations and the process of choosing the best performing configuration for the Iraq region, the model outputs tested for a single grid point (Baghdad station) of the atmospheric parameters (temperature, pressure and total precipitation) with modeled data and ECMWF. Model outputs using statistical methods: Bias Error (BE), Mean Absolute Error (MAE) and Root Mean Square Error (RMSE). The results show All the simulations predict rainfall with values close to the actual but it was discovered that the cloud microphysics setup had the greatest impact on temperature biases, whereas the cumulus parameterization setup has the greatest impact on precipitation.

Sensitivity of Simulated Conditions to Different Parameterization Choices Over Complex Terrain in Central Chile

This study evaluates the performance of fourteen high-resolution WRF runs with different combinations of parameterizations in simulating the atmospheric conditions over the complex terrain of central Chile during austral winter and spring. We focus on the validation of results for coastal, interior valleys, and mountainous areas independently, and also present an in-depth analysis of two synoptic-scale events that occurred during the study period: a frontal system and a cut-off low. The performance of the simulations decreases from the coast to higher altitudes, even though the differences are not very clear between the coast and interior valleys for 10 m wind speeds and precipitation. The simulated vertical profiles show a warmer and drier boundary layer and a cooler and moister free atmosphere than observed. The choice of the land-surface model has the largest positive impact on near-surface variables with the five-layer thermal diffusion scheme showing the smallest errors. Precipitation is more sensitive to the choice of cumulus parameterizations, with the simplified Arakawa–Schubert scheme generally providing the best performance for absolute errors. When examining the performance of the model simulating rain/no-rain events for different thresholds, also the cumulus parameterizations better represented the false alarm ratio (FAR) and the bias score (BS). However, the Morrison microphysics scheme resulted in the best critical success index (CSI), while the probability of detection (POD) was better in the simulation without analysis nudging. Overall, these results provide guidance to other researchers and help to identify the best WRF configuration for a specific research or operational goal.

Sensitivity of summer precipitation simulation to the physical parameterizations in WRF over the Tibetan Plateau: A case study of 2018

The Tibetan Plateau (TP) holds a pivotal position in regional and global climate, whereas the atmospheric characteristics vary greatly in the simulation with different physical parameterization schemes at the gray-zone scale (around 9 km). In this study, 24 sets of experiments were set up with different cumulus parameterization schemes (CPSs), microphysics parameterization schemes (MPSs) and planetary boundary layer parameterization schemes (PBLSs) using the WRF model driven by the ERA5 reanalysis to explore the sensitivity and try to summarize a combination of parameterization schemes suitable for the simulation over TP. Based on comparisons against in-situ observations, most CPS experiments showed large wet bias and overestimated occurrence of precipitation events, and CTL without CPS and CU6 with Tiedtke CPS show better performance at gray-zone scale. The influence of MPS and PBLS is less than CPS on the simulation, and Thompson MPS and UW PBLS showed a better performance. Most of the CPS experiments reach an earlier and more intensive peak compared to the Integrated Multi-satellite Retrievals for GPM (IMERG) version 6 satellite precipitation product, except CTL, CU11, and CU6, which show closer but weaker peaks. CPS experiments simulate more precipitation associated with water vapor transporting into TP and anomalous cyclonic circulation against CTL. The less water vapor transport from the south edge of TP in the CTL against CU14, CU16, and CU2, leads to more local precipitation. Graupel particles and the warm-rain process play a crucial role in the precipitation simulation with different MPSs. Although with less water vapor in the low atmosphere, MP2 simulates more precipitation due to more graupel particles and warm-rain processes against CTL. PBLSs have a substantial impact on latent heat, then influence evaporation and precipitation. The lower PBL height usually leads to more latent heat and more precipitation.

Assessment of physical schemes for WRF model in convection-permitting mode over southern Iberian Peninsula

Convection-permitting models (CPMs) enable the representation of meteorological variables at horizontal high-resolution spatial scales (≤ 4 km), where convection plays a significant role. In this regard, physical schemes need to be evaluated considering factors in the studied region such as orography and climate variability. This study investigates the sensitivity of the Weather Research and Forecasting (WRF) model as CPM to the use of different physics schemes on Andalusia, a complex orography region in the southern part of the Iberian Peninsula (IP). To do that, a set of 1-year WRF simulations was completed based on two “one-way” nested domains: the parent domain (d01) spanning the entire IP with 5 km spatial resolution and the nested domain (d02) for the region of Andalusia at 1 km of spatial resolution. 12 physic schemes were examined from combinations of microphysics (MP) schemes including THOMPSON, WRF single moment 6-class (WSM6), and WRF single moment 7-class (WSM7), and different options for the convection in d01, the Grell 3D (G3), Grell-Freitas (GF), Kain-Fritsch (KF), and deactivated cumulus parameterization (OFF). The simulated precipitation and 2-m temperature for the year 2018, characterized to be a very wet year, were compared with observational datasets from different sources to determine the optimal WRF configuration, including point-to-point and station-point comparisons at different time aggregations (from annual to hourly). In general, greater differences were shown when comparing the results of convection schemes in d01. Simulations completed with GF or OFF presented better performance compared to the reference datasets. Concerning the MP, although THOMPSON showed a better fit in high mountain areas, it generally presents a worse agreement with the reference datasets. In terms of temperature, the results were very similar and, therefore, the selection of the “best” configuration was based mainly on the precipitation results with the WSM7-GF scheme being suitable for Andalusia region.

The Role of Passive Cloud Volumes in the Transition From Shallow to Deep Atmospheric Convection

AbstractProperly capturing the transition from shallow to deep convection remains a major shortcoming of numerical weather and climate models due to poor understanding of the physical processes controlling this transition. Although recent studies suggest shallow preconditioning and cold pool feedbacks to be important, these studies are unable to explain the initial phase of the transition. We identify an additional mechanism, namely the interaction between passive cloud volumes (PCV)—old cumulus cloud volumes in the decaying stage—and updrafts, and discuss the potential role of this mechanism in the transition from shallow to deep convection. We show that the number of updrafts interacting with PCV is very large during the transition and the updrafts better preserve their buoyancy due to entrainment of much moister air. We argue that PCV might play an important role in the transition and cumulus parameterizations could therefore benefit from including them.

Simulating Heavy Rainfall Associated with Tropical Cyclones and Atmospheric Disturbances in Thailand Using the Coupled WRF-ROMS Model—Sensitivity Analysis of Microphysics and Cumulus Parameterization Schemes

Predicting heavy rainfall events associated with Tropical Cyclones (TCs) and atmospheric disturbances in Thailand remains challenging. This study introduces a novel approach to enhance forecasting precision by utilizing the coupled Weather Research and Forecasting (WRF) and Regional Oceanic Model (ROMS), known as WRF-ROMS. We aim to identify the optimal combination of microphysics (MP) and cumulus (CU) parameterization schemes. Three CU schemes, namely, Betts-Miller-Janjic (BMJ), Grell 3D Ensemble (G3), and Kain-Fritsch (KF), along with three MP schemes, namely, Eta (ETA), Purdue Lin (LIN), and WRF Single-moment 3-class (WSM3), are selected for the sensitivity analysis. Seven instances of heavy (35.1–90.0 mm) to violent (>90.1 mm) rainfall in Thailand, occurring in 2020 and associated with tropical storms and atmospheric disturbances, are simulated using all possible combinations of the chosen physics schemes. The simulated rain intensities are compared against observations from the National Hydroinformatics Data Center. Performance was assessed using the probability of detection (POD), false alarm ratio (FAR), and critical success index (CSI) metrics. While the models performed well for light (0.1–10.0 mm) to moderate (10.1–35.0 mm) rainfall, forecasting heavy rainfall remained challenging. Certain parameter combinations showed promise, like BMJ and KF with LIN microphysics, but challenges persisted. Analyzing density distribution of daily rainfall, we found effective parameterizations for different sub-regions. Our findings emphasize the importance of tailored parameterizations for accurate rainfall prediction in Thailand. This customization can benefit water resource management, flood control, and disaster preparedness. Further research should expand datasets, focusing on significant heavy rainfall events and considering climate factors, for example, the Madden-Julian Oscillation (MJO) for extended-range forecasts, potentially contributing to sub-seasonal and seasonal (S2S) predictions.

High-resolution WRF simulations of a monsoon event (2019) over the Badulu Oya Catchment, Sri Lanka: Role of cumulus parameterization condition and microphysics schemes

High-resolution WRF simulations of a monsoon event (2019) over the Badulu Oya Catchment, Sri Lanka: Role of cumulus parameterization condition and microphysics schemes

On the Resolution Sensitivity of Equatorial Precipitation in a GFDL Global Atmospheric Model

AbstractWe performed a series of aquaplanet simulations at the horizontal resolution from 50 to 6 km with identical parameterization settings using the Geophysical Fluid Dynamics Laboratory's Atmosphere Model version 4 implemented with the two‐moment Morrison‐Gettelman cloud microphysics with prognostic precipitation (GFDL AM4‐MG2). At the finer resolution, the global mean resolved‐scale precipitation increases and that from cumulus parameterization decreases. The model also simulates less/thinner clouds over the low latitudes and more/thicker clouds over the high latitudes as the resolution increases. The precipitation over the deep tropics is investigated in detail. We find little resolution sensitivity in the daily mean precipitation extremes. Changes of the equatorial resolved precipitation with resolution cannot be fully explained by the resolution dependence in the vertical velocity amplitude. We report a robust sensitivity in the convective organization over the deep tropics to the model resolution. In simulations of finer resolution, the localized convection is suppressed, and the organized convective system associated with large‐scale circulations becomes more prominent.

Assessment of the high-resolution estimations of global and diffuse solar radiation using WRF-Solar

Compared with physical models, WRF-Solar, as an excellent numerical forecasting model, includes abundant novel cloud physical and dynamical processes, which enablesenable the high-frequency output of radiation components which are urgently needed by the solar energy industry. However, the popularisation of WRF-Solar in a wide range of applications, such as the estimation of diffuse radiation, suffers from unpredictable influences of cloud and aerosol optical property parameters. This study assessed the accuracy of the improved numerical weather prediction (WRF-Solar) model in simulating global and diffuse radiation. Aerosol optical properties at 550 nm, which were provided by a moderate resolution imaging spectroradiometer, were used as input to analyse the differences in accuracies obtained by the model with/without aerosol input. The sensitivity of WRF-Solar to aerosol and cloud optical properties and solar zenith angle (SZA) was analysed. The results show the superiority of WRF-Solar to WRF-Dudhia in terms of their root mean square error (RMSE) and mean absolute error (MAE). The coefficients of determination between WRF-Solar and WRF-Dudhia revealed no statistically significant difference, with values greater than 0.9 for the parent and nested domains. In addition, the relative RMSE (RRMSE%) reached 46.60%. The experiment on WRF-Solar and WRF-Dudhia revealed a negative bias for global radiation, but WRF-Solar attained a slightly lower RMSE and higher correlation coefficient than WRF-Dudhia. The WRF-Solar-simulated results on diffuse radiation under clear sky conditions were slightly poorer, with RMSE, RRMSE, mean percentage error and MAE of 181.93 W m−2, 170.52%, 93.04% and 138 W m−2, respectively. Based on Himawari-8 cloud data, statistical results on cloud optical thickness (COT) for cloudy days revealed that WRF-Solar overestimated diffuse radiation at COTs greater than 20. Moreover, when the aerosol optical depth was greater than or equal to 0.8, WRF-Solar also overestimated the diffuse radiation, with a mean difference of 58.57 W m−2. The errors of WRF-Solar simulations in global and diffuse radiation exhibited a significant dependence on the SZA. The dispersion degree of deviation increased gradually with the decrease in the SZA. Thus, WRF-Solar serves as an improved numerical tool that can provide high temporal and high-spatial-resolution solar radiation data for the prediction of photovoltaic power. Studies should explore the improvement of cumulus parameterisation schemes to enhance the accuracy of solar radiation component estimation and prediction under cloudy conditions.

Air quality simulation with WRF-Chem over southeastern Brazil, part I: Model description and evaluation using ground-based and satellite data

A comprehensive Weather Research and Forecasting with Chemistry (WRF-Chem) model evaluation is conducted using ground-based and total column observational data from air quality stations and satellite retrievals. Fine particles (PM2.5; ≤ 2.5 μm in aerodynamic diameter), nitrogen oxides (NOx, NO + NO2), carbon monoxide (CO), tropospheric ozone (O3) concentrations and AOD values over southeastern Brazil were analyzed to assess the model's capability in reproducing atmospheric observation. The model simulations were performed over simple one domain at grid resolution of 10 km over southeastern Brazil. This spatial resolution was chosen due to a previous evaluation between five MODIS AOD products with AERONET estimates, resulting in Dark Target at 10 km of spatial resolution the best product to represent the AOD values over our study domain. Model input emissions comprise vehicular emissions derived from a bottom-up emission model, as well as on-line calculations of biogenic and fire emission rates. Given that the atmospheric state affects air pollution dispersion, a model evaluation on the meteorological conditions was carried out to better evaluate the model performance in reproducing the pollutant concentrations. Good agreement between model simulations and observations for air temperature and relative humidity at 2 m height was found, with correlation coefficients higher than 0.85 in most periods. Expected benchmarks for wind speed and direction at 10 m height were also found in this analysis, though with larger uncertainties. Underestimation occurred for daily accumulated precipitation due to the limitations of the cloud microphysics scheme or cumulus parameterization. Model simulations of PM2.5, NOx, CO and O3 agreed well with ground-based observations in terms of temporal variations and trends, with model-observation discrepancies due to uncertainties in the emission inventories. O3 was the better simulated pollutant in terms of temporal variability, with the characteristic large and small amplitudes observed over urban and rural areas being well represented by the model. High O3 concentrations were observed at the Botucatu station, due to transport of pollutants generated in the Metropolitan Area of São Paulo, and were also represented by the model, indicating the need of more active air quality monitoring stations over inland regions in southeastern Brazil. Moderate and high correlation coefficients (ranging 0.46–0.81) were found for tropospheric NO2 VCD and CO column, and AOD at 550 nm due to uncertainties in the emission inventories and aerosol model simplifications. Both the model and satellite captured higher values in similar regions over our study domain. This work represents a first effort, in southeastern Brazil, that combines numerical modeling, remote sensing and ground-based stations to analyze and understand the impact exerted by the emissions of urban pollution over surrounding areas. A more in-depth analysis of the impact of emissions transport to inland regions from urban areas in southeastern Brazil will be discussed in the second part of this work.

Impacts of the stochastically perturbed parameterization on the precipitation ensemble forecasts of the Betts–Miller–Janjić (BMJ) scheme in Eastern China

This study investigated the impact of stochastic perturbations on precipitation forecasts generated by the Betts-Miller-Janjic (BMJ) cumulus parameterization scheme. Two types of stochastic perturbation approaches were compared. The first approach involved perturbing the temperature and humidity tendencies derived from the BMJ scheme. The second approach focused on perturbing the reference profiles of temperature and humidity estimated within the BMJ scheme. These profile perturbations led to reference profiles becoming either warmer and wetter or colder and drier. Ten precipitation cases occurring in eastern China during the summer of 2019 were selected to evaluate the different perturbation methods. The default BMJ scheme exhibited a significant wet bias at the light rain threshold due to overestimating entropy change and a dry bias at the heavier rain threshold. The tendency perturbation approach, which perturbed the temperature and humidity, did not affect the entropy change and produced precipitation placement similar to its unperturbed counterpart. These attributes resulted in a small ensemble spread and relatively low forecast skill scores. Perturbing the reference profiles, conversely, influenced the entropy change in the BMJ scheme, leading to a larger ensemble spread and higher forecast skill scores regarding the spatial distribution of precipitation. Asymmetrically perturbing the reference profiles considered the wet bias, which increased the grid points with negative entropy change and outperformed the symmetric perturbation. The asymmetric perturbation also increased the chance of heavier rain because more water was retained in the air, alleviating the corresponding dry bias.

A Multicloud Model for Coastal Convection

Coastal convection is often organized into multiple mesoscale systems that propagate in either direction across the coastline (i.e., landward and oceanward). These systems interact non-trivially with synoptic and intraseasonal disturbances such as convectively coupled waves and the Madden–Julian oscillation. Despite numerous theoretical and observational efforts to understand coastal convection, global climate models still fail to represent it adequately, mainly because of limitations in spatial resolution and shortcomings in the underlying cumulus parameterization schemes. Here, we use a simplified climate model of intermediate complexity to simulate coastal convection under the influence of the diurnal cycle of solar heating. Convection is parameterized via a stochastic multicloud model (SMCM), which mimics the subgrid dynamics of organized convection due to interactions (through the environment) between the cloud types that characterize organized tropical convection. Numerical results demonstrate that the model is able to capture the key modes of coastal convection variability, such as the diurnal cycle of convection and the accompanying sea and land breeze reversals, the slowly propagating mesoscale convective systems that move from land to ocean and vice-versa, and numerous moisture-coupled gravity wave modes. The physical features of the simulated modes, such as their propagation speeds, the timing of rainfall peaks, the penetration of the sea and land breezes, and how they are affected by the latitudinal variation in the Coriolis force, are generally consistent with existing theoretical and observational studies.

WRF-MODEL PARAMETERIZATION TEST FOR PREDICTING EXTREME HEAVY RAINFALL EVENT OVER KETAPANG REGENCY

Heavy rains that cause floods and landslides in the Ketapang Regency can be predicted by utilizing the weather research and forecast (WRF) model. The WRF model used, of course, needs to be configured to represent the conditions that exist in Ketapang Regency. This study evaluates the combination of cumulus and microphysics parameterization, producing the best prediction of 24-hour accumulated rainfall. The combination of cumulus and microphysics parameterization tested as many as 24 schemes which later will be obtained which combination can produce the best prediction of rainfall accumulation with the comparison of rainfall measured at the Observation Station of the Meteorology, Climatology, and Geophysics Agency (BMKG) in Ketapang Regency. The results show that combining the KF-Scheme cumulus parameterization scheme and the Kessler-Scheme microphysics can better predict 24-hour accumulated rainfall than other tested parameterization schemes. This result is based on the root mean square error (RMSE), which shows that this combination scheme produces the smallest value and large correlation coefficient (CORR). From this research, it can also be seen that cumulus parameterization has a more dominant role than microphysics parameterization.

Comparative study on the characteristics of rainfall simulation over South Korea by summertime weather patterns according to the use of cumulus parameterization

Comparative study on the characteristics of rainfall simulation over South Korea by summertime weather patterns according to the use of cumulus parameterization

On the Collective Importance of Model Physics and Data Assimilation on Mesoscale Convective System and Precipitation Forecasts over Complex Terrain

Abstract Forecasting mesoscale convective systems (MCSs) and precipitation over complex terrain is an ongoing challenge even for convective-permitting numerical models. Here, we show the value of combining mesoscale constraints to improve short-term MCS forecasts for two events during the North American monsoon season in 2013, including the following: 1) the initial specification of moisture, via GPS-precipitable water vapor (PWV) data assimilation (DA); 2) kinematics via modification of cumulus parameterization; and 3) microphysics via modification of cloud microphysics parameterization. A total of five convective-permitting Weather Research and Forecasting (WRF) Model experiments is conducted for each event to elucidate the impact of these constraints. Results show that combining GPS-PWV DA with a modified Kain–Fritsch scheme and double-moment microphysics provides relatively the best forecast of both North American monsoon MCSs and convective precipitation in terms of timing, location, and intensity relative to available precipitation and cloud-top temperature observations. Additional examination on the associated reflectivity, vertical wind field, equivalent potential temperature, and hydrometeor distribution of MCS events show the added value of each individual constraint to forecast performance. Significance Statement Forecasting thunderstorm clouds and rain over mountainous regions is challenging because of limitations in having radar and rain gauges and in resolving physical drivers in forecast models. We examine the value of considering all possible constraints by incorporating moisture into these models, and correcting physics in the model treatment of cumulus and cloud microphysics parameterizations. This study demonstrates that assimilating moisture and using modified Kain–Fritsch and double-moment microphysics schemes provides the best thunderstorm cloud and rain forecasts in terms of timing, location, and intensity. Each correction improves key properties of these storms such as vertical wind, along with distribution of water in various phases. We highlight the need to improve our efforts on effectively integrating these constraints into current and future forecasts.

Impact of convection scheme on ENSO prediction of SINTEX-F2

A spectral cumulus parameterization (spectral scheme) is implemented in Scale Interaction Experiment Frontier version 2 (SINTEX-F2) seasonal prediction system, and the impact on the El Niño Southern Oscillation (ENSO) prediction is examined. By conducting hindcast experiments using the original convection scheme (Tiedtke scheme) and the spectral scheme, and comparing the ENSO prediction skill, the impact of the spectral scheme is analyzed in detail. It was found that prediction skill in terms of ENSO phase and the sea surface temperature (SST) persistence were improved by using the spectral scheme, but the root-mean-square error (RMSE) increased. The ENSO feedback was also changed by changing the convection scheme. The original scheme failed to predict the zonal wind stress anomaly toward the Niño 3.4 region, whereas the spectral scheme simulated it over the equatorial eastern Pacific with narrowing the meridional width, indicating that the spectral scheme strengthened the ENSO feedback. The spectral scheme also improved zonal-vertical atmospheric response to the Niño 3.4 index due to its advantageous features. Analysis of the ENSO feedback terms revealed that strengthened forcing in the eastern Pacific improved the thermocline feedback of ENSO, as its reversed timing of positive and negative tendencies for the mixed layer temperature matched that estimated from the reanalysis data. In conclusion, the spectral scheme can improve ENSO prediction through the atmospheric forcing and mean state in the eastern Pacific which impacted the ocean properties. It improved the phase error by improving thermocline feedback, but did not improve the RMSE. Tuning of the original scheme to obtain additional improvements to ENSO prediction would be difficult, since it requires modification of detailed convective cloud properties to correct the phase error. The spectral scheme tends to overestimate the ENSO amplitude, i.e., large RMSE, but this drawback can be mitigated by tuning the convection scheme so that it suppresses the warm SST climate drift, and this is considered the more promising method to further improve ENSO prediction.

All-sky infrared radiance data assimilation of FY-4A AGRI with different physical parameterizations for the prediction of an extremely heavy rainfall event

The Fengyun-4A (FY-4A) Advanced Geostationary Radiation Imager (AGRI) allows for high-spatiotemporal-resolution observations of the local severe weather systems. The capabilities for assimilating the all-sky AGRI infrared radiances are explored with the WRF three-dimensional variational data assimilation system (WRF-3DVAR) to improve the forecast accuracy of an extremely heavy rainfall event over the Henan Province of China in this study. In particular, data assimilation experiments are conducted with different physical parameterization schemes, including microphysical schemes and cumulus parameterizations. The results show that the all-sky AGRI radiance data assimilation experiments based on the Purdue Lin microphysics scheme and the Kain–Fritsch cumulus scheme can lead to better forecasts of the rainfall intensity and location, respectively. Meanwhile, it is found that the analyzed cloud top temperature fits the observations better when cloudy radiances from the AGRI are assimilated using the Purdue Lin scheme. This finding suggests that proper physical processes could facilitate the effective use of AGRI cloud-affected observations. On the other hand, positive forecast impacts from assimilating the all-sky AGRI radiance data have been confirmed when compared to the experiments that assimilate only the clear-sky AGRI radiances or only conventional observations. The assimilation of the all-sky AGRI water vapor channel is conducive to humidifying the initial conditions in the middle and low troposphere with more realistic analysis increments of water vapor, cloud water, and cloud ice, resulting in improved intensity forecasts of the heavy rainfall system and better forecast skill scores.

Understanding and Reducing Warm and Dry Summer Biases in the Central United States: Analytical Modeling to Identify the Mechanisms for CMIP Ensemble Error Spread

Abstract Most climate models in phase 6 of the Coupled Model Intercomparison Project (CMIP6) still suffer pronounced warm and dry summer biases in the central United States (CUS), even in high-resolution simulations. We found that the cloud base definition in the cumulus parameterization was the dominant factor determining the spread of the biases among models and those defining cloud base at the lifting condensation level (LCL) performed the best. To identify the underlying mechanisms, we developed a physically based analytical bias model (ABM) to capture the key feedback processes of land–atmosphere coupling. The ABM has significant explanatory power, capturing 80% variance of temperature and precipitation biases among all models. Our ABM analysis via counterfactual experiments indicated that the biases are attributed mostly by surface downwelling longwave radiation errors and second by surface net shortwave radiation errors, with the former 2–5 times larger. The effective radiative forcing from these two errors as weighted by their relative contributions induces runaway temperature and precipitation feedbacks, which collaborate to cause CUS summer warm and dry biases. The LCL cumulus reduces the biases through two key mechanisms: it produces more clouds and less precipitable water, which reduce radiative energy input for both surface heating and evapotranspiration to cause a cooler and wetter soil; it produces more rainfall and wetter soil conditions, which suppress the positive evapotranspiration–precipitation feedback to damp the warm and dry bias coupling. Most models using non-LCL schemes underestimate both precipitation and cloud amounts, which amplify the positive feedback to cause significant biases.

Understanding and Reducing Warm and Dry Summer Biases in the Central United States: Improving Cumulus Parameterization

Abstract Most climate models still suffer large warm and dry summer biases in the central United States (CUS). As a solution, we improved cumulus parameterization to represent 1) the lifting effect of small-scale rising motions associated with Great Plains low-level jets and midtropospheric perturbations by defining the cloud base at the level of condensation, 2) the constraint of the cumulus entrainment rate depending on the boundary layer depth, and 3) the temperature-dependent cloud-to-rainwater conversion rate. These improvements acted to (i) trigger mesoscale convective systems in unfavorable environmental conditions to enhance total rainfall amount, (ii) lower cloud base and increase cloud depth to increase low-level clouds and reduce surface shortwave radiation, (iii) suppress penetrative cumuli from shallow boundary layers to remedy the overestimation of precipitation frequency, and (iv) increase water detrainment to form sufficient cirrus clouds and balanced outgoing longwave radiation. Much of these effects were nonlocal and nonlinear, where more frequent but weaker convective rainfall led to stronger (and sometimes more frequent) large-scale precipitation remotely. Together, they produced consistently heavier precipitation and colder temperature with a realistic atmospheric energy balance, essentially eliminating the CUS warm and dry biases through robust physical mechanisms.