Simulated Precipitation Research Articles

Interpretable Multiscale Machine Learning‐Based Parameterizations of Convection for ICON

AbstractMachine learning (ML)‐based parameterizations have been developed for Earth System Models (ESMs) with the goal to better represent subgrid‐scale processes or to accelerate computations. ML‐based parameterizations within hybrid ESMs have successfully learned subgrid‐scale processes from short high‐resolution simulations. However, most studies used a particular ML method to parameterize the subgrid tendencies or fluxes originating from the compound effect of various small‐scale processes (e.g., radiation, convection, gravity waves) in mostly idealized settings or from superparameterizations. Here, we use a filtering technique to explicitly separate convection from these processes in simulations with the Icosahedral Non‐hydrostatic modeling framework (ICON) in a realistic setting and benchmark various ML algorithms against each other offline. We discover that an unablated U‐Net, while showing the best offline performance, learns reverse causal relations between convective precipitation and subgrid fluxes. While we were able to connect the learned relations of the U‐Net to physical processes this was not possible for the non‐deep learning‐based Gradient Boosted Trees. The ML algorithms are then coupled online to the host ICON model. Our best online performing model, an ablated U‐Net excluding precipitating tracer species, indicates higher agreement for simulated precipitation extremes and mean with the high‐resolution simulation compared to the traditional scheme. However, a smoothing bias is introduced both in water vapor path and mean precipitation. Online, the ablated U‐Net significantly improves stability compared to the non‐ablated U‐Net and runs stable for the full simulation period of 180 days. Our results hint to the potential to significantly reduce systematic errors with hybrid ESMs.

Relative contribution of dynamic and thermodynamic components on Southeast Asia future precipitation changes from different multi-GCM ensemble members

To address the gap in understanding precipitation changes in Southeast Asia and to enhance the reliability of climate projections for the region through moisture budget analysis, this study examines the differences among six multi-model ensembles of CMIP6 simulated precipitation in term of moisture budget analysis. It investigates the relative contributions of thermodynamic and dynamic components to seasonal precipitation changes over Southeast Asia under the highest emission scenario, SSP5-8.5. The comparison between ensembles indicates that Good performance model ensembles slightly outperform the combination of all resolution and all category ensembles in reducing the biases. There is no strong evidence showing that good category ensembles outperform the combination of all model ensemble groups in simulating the spatial pattern of historical seasonal precipitation. From the perspective of moisture budget, regions receiving seasonal high rainfall intensity are mainly influenced by the moisture convergence during the monsoon seasons: northeast monsoon (December‒January‒February) and southwest monsoon (June–July–August). By the late 21st century (2081–2100), all model ensemble projections show an increase in December‒January‒February precipitation over the northern Southeast Asia and decreased June‒July‒August rainfall in the southern regions. The moisture budget analysis explained that the seasonal mean rainfall change in Southeast Asia is largely influenced by evaporation and followed by moisture flux convergence. The changes in moisture flux convergence are contributed by both the dynamic and thermodynamic components. Greater inter-model uncertainty was found in the precipitation dynamic component compared to the thermodynamic component suggesting the existence of large discrepancy between the various approaches used by GCMs in describing atmospheric dynamics. The study highlights that the Good model ensemble with middle to low resolution is able to narrow the inter-model uncertainties in terms of the moisture budget analysis compared to the combination of all Good model ensembles.

Novel optimized coupled rainfall model simulation based on stepwise decomposition technique.

Precipitation forecasting plays a pivotal role in guiding the effective management of regional water resources and providing crucial warnings for regional droughts and floods. Finding a monthly precipitation simulation model with robust fitting performance is a significant research endeavor in practical precipitation prediction. This paper introduces two modified African vulture optimization algorithms (MAVOA1 and MAVOA2). It provides hyperparameter optimization techniques for the least squares support vector machine (LSSVM), long short-term memory neural network (LSTM), and random forest (RF) models. These techniques are used to construct a monthly precipitation simulation model based on algorithmic optimization coupled with variational mode decomposition for full decomposition. The test results at five typical stations in the North China Plain reveal the following: (1) the LSSVM model demonstrates significantly better performance than the LSTM and RF models. (2) the MAVOA2-LSSVM model has the best-integrated effect: the average test fitting error is RMSE = 17.50 mm/month, MRE = 117.25%, NSE = 0.90, which shows its superiority in practical application and can significantly improve the accuracy of precipitation prediction; MAVOA2 is more suitable for machine learning models with more hyperparameters of its own, which provides a reference for hyperparameter optimization algorithms in the other fields.

WRF numerical simulation of summer precipitation and its application over the mountainous southern Tibetan Plateau based on different cumulus parameterization schemes

The extreme scarcity of meteorological observation data caused by harsh natural environments greatly limits our understanding of precipitation and related atmospheric processes over the mountainous regions of the southern Tibetan Plateau and its surrounding areas (hereafter mountainous southern TP). Precipitation simulation has been challenging in this region due to the complex topography and large elevation differences. Considering the high sensitivity of precipitation to the choice of cumulus parameterization scheme (CPS) over the TP, this study systematically evaluated the precipitation simulation performance of 11 CPSs in July 2018 using the WRF model. The characteristics of atmospheric circulation and gradient variation of precipitation were also revealed based on the optimal CPS over this region. The results indicated that the scale-aware schemes (KIAPS SAS [KSAS] and multi-scale Kain–Fritsch [MSKF]) demonstrated excellent potential for precipitation simulations. Due to the reasonable expression of atmospheric stability and the resulting accurate simulation of the timing and amount of regional heavy precipitation events, the KSAS scheme was superior to the MSKF scheme. The output of the KSAS-configured experiment revealed two main water vapor transport channels for the mountainous southern TP in summer, namely the southern monsoonal channel and the western westerly channel. Based on the precipitation profiles along the typical paths of these two channels (longitudinal path of 95.5°E and latitudinal path of 29.5°N), two precipitation belts and one precipitation belt existed on the southern and western slopes of the TP, respectively.

Precipitation patterns impact soil aggregates and organic carbon of an alpine wetland on the Qinghai-Tibetan Plateau

Global climate change is predicted to influence both precipitation amount and frequency in many regions. However, little is known about how precipitation changes affect soil aggregates and soil organic carbon (SOC) in alpine wetlands. Plots of an alpine wetland on the Qinghai-Tibetan Plateau were subjected to two precipitation frequencies (high vs. low, i.e., 1 and 3 times of nature rainfall intervals) crossed with two precipitation amounts (high vs. low, i.e., 100 % and 70 % of the amount of natural rainfall). Two years later, we investigated changes of soil aggregates (silt + clay, < 53 μm; microaggregates, 53∼250 μm; and macroaggregates, >250 μm) and SOC in three soil depths, i.e., 0–15 cm, 15–30 cm and 30–50 cm. Averaged across the three soil depths, the content of SOC and dissolved organic carbon (DOC) in the simulated natural precipitation (high amount, high frequency treatment) were 127.22 g kg−1 and 0.19 g kg−1, respectively. Reducing precipitation amount or frequency led to a 19.6 %-26.2 % increase in SOC, and 28.7 %-44.4 % increase in DOC. Microbial biomass carbon in natural precipitation was 1.33 g kg−1 across the three soil depths, and its response to precipitation varied with soil depths. The fraction of macroaggregates in the 0–15 cm soil depth was 63.5 % in the simulated natural precipitation treatment, and decreased by 18.6 %–21.9 % in response to reduced precipitation amount or frequency. Aggregate fractions in the 15–30 cm soil depth did not respond significantly to the changes of precipitation patterns. Reducing precipitation frequency or amount decreased the contribution of macroaggregate-associated organic carbon to SOC compared to the simulated natural precipitation, but increased the contribution of microaggregate-associated organic carbon, especially in the 0–15 cm soil depth. Structural equation models revealed that SOC was regulated by soil aggregate composition and microbial biomass carbon. The buffering of the surface soil weakened the effects of precipitation changes on aggregates and aggregate-associated organic carbon in the deep soil. These results suggest that reducing precipitation frequency or amount can decrease the fraction of macroaggregates in the surface soil, and weaken the physical protection of macroaggregates for SOC. Thus, changes in precipitation patterns may have a significant impact on SOC storage and stability in alpine wetlands through their influence on soil aggregate composition.

Global variable-resolution simulations of extreme precipitation over Henan, China, in 2021 with MPAS-Atmosphere v7.3

Abstract. A historic rainstorm occurred over Henan, China, in July 2021 (“7.20” extreme precipitation event), resulting in significant human casualties and socioeconomic losses. A global variable-resolution model (MPAS-Atmosphere v7.3) was employed to simulate this extreme precipitation event. A series of simulations have been done at both quasi-uniform (60 and 15 km) and variable-resolution (60–15 and 60–3 km) meshes from hydrostatic to nonhydrostatic scale with two parameterization scheme suites. For the 48 h peak precipitation duration (20–22 July), the 60–3 km variable-resolution simulation coupled with the scale-aware convection-permitting parameterization scheme suite stands out predominantly among other simulation experiments as it reproduces this extreme precipitation event most accurately. At 15 km resolution, the 60–15 km variable-resolution simulation achieves comparable forecasting skills to the 15 km quasi-uniform simulation but at a much reduced computing cost. In addition, we found that the default mesoscale suite generally outperforms the convection-permitting suite at 15 km resolution as simulations coupled with the convection-permitting suite missed the third peak of this extreme precipitation event, while the mesoscale suite did not. Furthermore, it is found that the large-scale circulation plays a critical role in the peak precipitation simulations at 15 km resolution, via influencing the simulated low-level wind. During the second peak precipitation period, simulations with the convection-permitting parameterization scheme suite at 15 km resolution generate a prominent low-level easterly wind component bias, which is largely attributed to the excessively evaporative cooling in the lower troposphere. This study further reveals that at 15 km resolution the diabatic heating from the grid-scale precipitation accounts more for the low-level wind bias than the convective-scale precipitation. Given that two different cloud microphysics schemes, namely Thompson and WSM6 schemes, are used in the convection-permitting and default mesoscale parameterization scheme suites, respectively, these microphysics schemes are found to be the primary contributor to the low-level wind simulation bias.

Deep learning-based bias correction of ISMR simulated by GCM

Simulated data of atmospheric variables like precipitation is very important for climate science research, especially for understanding future scenarios. Such data is generated by global or regional climate models (GCM/RCM), for both past and future periods under different initial conditions. Unfortunately, these models frequently display biases in simulated precipitation data compared to ground observations, due to their inability to incorporate the entire physics of the process accurately. It is imperative to mitigate such biases using contemporary techniques to make the simulated data a valuable end product. Our aim in this research is to correct seasonal (from June to September) Indian Summer Monsoon Rainfall (ISMR) data as simulated by the Climate Forecast System (CFS) over the Indian subcontinent, with the Global Precipitation Climatology Project (GPCP) data serving as the ground truth reference. This period is important as more than one-third of India's annual rainfall occurs during these months. This study presents a novel Deep Learning (DL) based architecture known as the Convolutional Neural Network for Bias Correction (CNNBC) that aims to calibrate the model-simulated data with past observations, to address these biases while preserving the statistical properties and spatial correlations relationships among grid points and enhancing the spatial mean of precipitation estimates. To evaluate the effectiveness of our proposed method, we compare its performance to that of three other statistical and DL techniques: quantile mapping (QM), quantile delta mapping (QDM), and the SRDRN model. The comparative analysis demonstrates that the CNNBC model outperforms the other in terms of our task.

Assessing Multi-Scale Atmospheric Circulation Patterns for Improvements in Sub-Seasonal Precipitation Predictability in the Northern Great Plains

This study leverages the relationships between the Great Plains low-level jet (GP-LLJ) and the circumglobal teleconnection (CGT) to assess the enhancement of 30-day rainfall forecast in the Northern Great Plains (NGP). The assessment of 30-day simulated precipitation using the Climate Forecast System (CFS) is contrasted with the North American Regional Reanalysis, searching for sources of precipitation predictability associated with extended wet and drought events. We analyze the 30-day sources of precipitation predictability using (1) the characterization of dominant statistical modes of variability of 900 mb winds associated with the GP-LLJ, (2) the large-scale atmospheric patterns based on 200 mb geopotential height (HGT), and (3) the use of GP-LLJ and CGT conditional probability distributions using a continuous correlation threshold approach to identify when and where the forecast of NGP precipitation occurs. Two factors contributing to the predictability of precipitation in the NGP are documented. We found that the association between GP-LLJ and CGT occurs at two different scales—the interdiurnal and the sub-seasonal, respectively. The CFS reforecast suggests that the ability to forecast sub-seasonal precipitation improves in response to the enhanced simulation of the GP-LLJ and CGT. Using these modes of climate variability could improve predictive frameworks for water resources management, governance, and water supply for agriculture.

Evaluation of daily precipitation modeling performance from different CMIP6 datasets: A case study in the Hanjiang River basin

To effectively compare and analyze daily precipitation modeling capabilities across different CMIP6 datasets, our study introduces a novel method to compare daily precipitation models across CMIP6 datasets in the Hanjiang River Basin (HRB). We quantify indicators such as precipitation distribution, temporal correlation, wet-dry detection, extreme value error, and spatio-temporal variability, enabling a comprehensive rating of precipitation accuracy. It has been found that while both CMIP6(CP6) and NEX-GDDP-CMIP6 (GCP6) models show similar simulation accuracy, GCP6 excels in several aspects like distribution, temporal correlation, extreme value simulation, and spatial variability, yet lags in wet-dry detection and temporal change. Notably, using the comprehensive rating score (CRS) analysis, significant differences in precipitation simulation accuracy exist between models, particularly CP6, with variations of up to 0.22 (51.2%) between the highest and lowest scores. Among the top ten models, GCP6 occupies four positions such as MRI-ESM2-0, GFDL-CM4, MPI-ESM1-2-HR, and TaiESM1, while CP6 holds the remaining six like CanESM5, EC-Earth3-Veg, MPI-ESM1-2-HR, GFDL-CM4, MRI-ESM2-0, and IPSL-CM6A-LR. These findings not only offer a clear understanding of the simulation performance of CMIP6 datasets across various precipitation characteristics, but also quantitatively compare the modeling capabilities of different models for watershed precipitation through CRS. This aids climate adaptation research, hydrological forecasting, and flood management in the basin.

Comparison of CMIP5 and CMIP6 models for temperature and precipitation simulation over the Yarlung Tsangpo-Brahmaputra River Basin

ABSTRACT This study compares simulations from 13 CMIP5 and CMIP6 homologous models and their multi-model ensemble (MME) based on the generalized three-cornered hat method (TCH) for temperature and precipitation over the Yarlung Tsangpo-Brahmaputra River Basin (YBRB). The results showed that: (1) the MME estimations outperform the most single model, demonstrating the effectiveness of TCH in reducing model uncertainty. (2) CMIP5 and CMIP6 are more applicable for spring and autumn maximum temperatures (Tasmax), minimum temperatures (Tasmin) and precipitation in YBRB. Moreover, they showed better performance for precipitation on floodplain, for Tasmax on Tibetan Plateau, and for Tasmin in whole YBRB. (3) Relatively, CMIP5 and CMIP6 can better simulate the spatial distribution of temperature rather than precipitation (the Temporal Correlation Coefficient of Tasmax and Tasmin: 0.72 – 0.89; that of precipitation: 0.43–0.6). (4) The bias of CMIP6 for temperature and precipitation in YBRB is mostly lower than CMIP5, but still has a cold bias over YBRB (Tasmax: −7.98 °C to −14.88 °C, Tasmin: −6.24 °C to −21.45 °C) and wet bias for precipitation on Tibetan Plateau (0.56 mm − 2.00 mm), dry bias on Himalayan belt (−0.69 mm to −7.56 mm) and floodplain (−0.46 mm to −6.98 mm).

How Well Does the DOE Global Storm Resolving Model Simulate Clouds and Precipitation Over the Amazon?

AbstractThis study assesses a 40‐day 3.25‐km global simulation of the Simple Cloud‐Resolving E3SM Model (SCREAMv0) using high‐resolution ground‐based observations from the Atmospheric Radiation Measurement (ARM) Green Ocean Amazon (GoAmazon) field campaign. SCREAMv0 reasonably captures the diurnal timing of boundary layer clouds yet underestimates the boundary layer cloud fraction and mid‐level congestus. SCREAMv0 well replicates the precipitation diurnal cycle, however it exhibits biases in the precipitation cluster size distribution compared to scanning radar observations. Specifically, SCREAMv0 overproduces clusters smaller than 128 km, and does not form enough large clusters. Such biases suggest an inhibition of convective upscale growth, preventing isolated deep convective clusters from evolving into larger mesoscale systems. This model bias is partially attributed to the misrepresentation of land‐atmosphere coupling. This study highlights the potential use of high‐resolution ground‐based observations to diagnose convective processes in global storm resolving model simulations, identify key model deficiencies, and guide future process‐oriented model sensitivity tests and detailed analyses.

Phase-Field Simulation of Precipitation and Grain Boundary Segregation in Fe-Cr-Al Alloys under Irradiation.

A phase-field model for the precipitation of Fe-Cr-Al alloy is established incorporating grain boundary (GB) effects and irradiation-accelerated diffusion. The radiation source and grain boundary effect are incorporated to broaden the applicability of the Fe-Cr-Al precipitated phase-field model. The model is firstly employed to simulate the precipitation of the Cr-rich α' phase in a single-crystal alloy. The precipitation rate and the size distribution of the precipitated phase were analyzed. Subsequently, the model is utilized to simulate segregation at GBs in a double-crystal system, analyzing the enrichment of Cr and depletion of Al near these boundaries. The simulation results are consistent with experimental observations reported in the references. Finally, the model is applied to simulate the precipitation in a polycrystalline Fe-Cr-Al system. The simulated results revealed that the presence of GBs induces the formation of localized regions with enhanced Cr and Al content as well as depleted zones adjacent to these boundaries. GBs also diminish both the quantity and precipitation rate of the formed phase within the grains.

Green Infrastructure Microbial Community Response to Simulated Pulse Precipitation Events in the Semi-Arid Western United States

Processes driving nutrient retention in stormwater green infrastructure (SGI) are not well quantified in water-limited biomes. We examined the role of plant diversity and physiochemistry as drivers of microbial community physiology and soil N dynamics post precipitation pulses in a semi-arid region experiencing drought. We conducted our study in bioswales receiving experimental water additions and a montane meadow intercepting natural rainfall. Pulses of water generally elevated soil moisture and pH, stimulated ecoenzyme activity (EEA), and increased the concentration of organic matter, proteins, and N pools in both bioswale and meadow soils. Microbial community growth was static, and N assimilation into biomass was limited across pulse events. Unvegetated plots had greater soil moisture than vegetated plots at the bioswale site, yet we detected no clear effect of plant diversity on microbial C:N ratios, EEAs, organic matter content, and N pools. Differences in soil N concentrations in bioswales and the meadow were most directly correlated to changes in organic matter content mediated by ecoenzyme expression and the balance of C, N, and P resources available to microbial communities. Our results add to growing evidence that SGI ecological function is largely comparable to neighboring natural vegetated systems, particularly when soil media and water availability are similar.

Assessment of CMIP6 models performance in simulation precipitation and temperature over Iran and surrounding regions.

This study investigates the performance of CMIP6 models in reproducing historical temperature and precipitation data for Iran and neighboring countries (Afghanistan, Pakistan, Turkmenistan, Azerbaijan, Armenia, Turkey, and Iraq) from 1980 to 2014. Reanalysis data from the ECMWF database (ERA5) for temperature and precipitation were utilized as a reference for the period 1980-2014. Additionally, ten Atmosphere-Ocean General Circulation Models (AOGCMs) from CMIP6 were employed to simulate temperature and precipitation data for the study region based on the IPCC Sixth Assessment Report databases. The Kling-Gupta Efficiency (KGE) index was used to evaluate the accuracy of CMIP6 models in replicating daily temperature and precipitation. The results indicate that different CMIP6 models exhibit varying degrees of accuracy in simulating historical temperatures and precipitation, depending on the month and the country. For instance, the IPSL-CM6A-LR model demonstrated the best annual performance in estimating temperature in Azerbaijan (KGE = 0.5), while the HadGEM3-GC31-LL model showed the lowest annual performance in Pakistan (KGE = -1.4). Interestingly, the models were found to be more accurate in simulating temperatures during warm months compared to cold ones. Furthermore, the accuracy of different models in estimating annual precipitation varied significantly, ranging from -0.64 (MRI-EMS2-0 model in Afghanistan) to 0.05 (CMCC-ESM2 model in Armenia). Similar to temperature, the study found that models were generally more accurate in simulating precipitation during cold months compared to warm ones.

Effects of aging of radioactive fallout on timely decontamination of concrete using low and mild pressure washing

Timely decontamination will reduce the consequences of a radiological contamination event. For this purpose, pressure washing can be rapidly deployed, but its effectiveness will change if the interactions between the surface and radionuclides changes as the contamination “ages” under the influence of time and precipitation. While effects of this aging have been reported for dissolved cesium, they have not been studied for radionuclides present as particulate, e.g., fallout. This work studied the effects of aging on decontamination with low (<280 kPa/40 psi) and mild (14,000 kPa/2000 psi) pressure washing, on concrete contaminated with surrogate fallout consisting of soluble Cs-137, 0.5 μm silica particles, and 2 μm silica particles. The samples were aged up to 59 days (time between contamination and decontamination) with and without simulated precipitation. The percent removal following decontamination of the soluble cesium decreased over the first ten days of aging until the removals were less than 10 % for both low and mild pressure washing. The particle decontamination was independent of aging time but decontaminating via mild pressure washing (>80 % particle removal) significantly outperformed decontaminating by low pressure washing by flowing solution across (parallel to) the contaminated surface (<25 % particle removal). The observed changes in decontamination efficacy are explained via measurements of the penetration depth of contaminants. For soluble cesium, the results compared favorably with prior studies and theoretical treatment of cesium penetration, and they yielded additional insight into the effect of washing pressures on decontamination. There are no comparable studies for particulate contamination, so this study resulted in several novel observations which are operationally important for timely decontamination of surfaces following a radiological incident. It also suggests an evidence-based pressure washing procedure for timely decontamination of soluble and insoluble radionuclides which can be used throughout the emergency phase and into the early recovery phase.

WITHDRAWN: Long-term warming and decreased precipitation regulated microbial community structure and potential function through enhanced predator-prey relationships in the Tibetan Plateau

Global climate change can shape the interactions among soil microbes and, in turn, mediate ecosystem functions. However, how these interactions were regulated remains to be investigated. This study utilized 16S rRNA, ITS, and 18S rRNA high-throughput sequencing to investigate the effects of simulated warming and precipitation changes on the major components of soil micro-food webs in the Qinghai-Tibetan Plateau through a field experiment. Adonis and non-metric multidimensional scaling (NMDS) analyses showed that compositions of bacteria, fungi and protists were all affected by changes in temperature and precipitation. Correlation and cross-trophic network analyses revealed that warming and decreased precipitation, both separately and together, enhanced the relationships between bacteria and protists, especially protistan predators. We found that the modified stochasticity ratio of the bacterial community assembly was best predicted by protists. The potential functional structures of bacteria were positively correlated with protistan predators under warming and decreased precipitation condition, suggesting enhanced predator-prey relationships between protists and bacteria might further influence the potential functional characteristics of bacterial communities. Our study indicated that climate change altered bacterial compositional and functional structure via enhanced predator-prey interactions. Therefore, in the context of global climate change, broader and more comprehensive studies on protist-regulated soil bacterial and fungal communities are imperative.

Impacts of Moisture Advection Scheme on Precipitation in the Steep Topography Region between the Tibetan Plateau and the Sichuan Basin

Abstract Water vapor transport is a crucial process in modeling and can contribute to errors in precipitation forecasts. To investigate the sensitivity of precipitation to the moisture advection scheme, this study introduced the two-step shape-preserving advection scheme (TSPAS), which has been proven to improve precipitation simulation over steep topography at lower resolutions, into the Southwest Center Weather Research and Forecast (WRF)-based Intelligent Numerical Grid Forecast System (SWC-WINGS) at a convection-permitting resolution. According to experiments conducted throughout the summer of 2021, the precipitation over the eastern slope of the Tibetan Plateau (TP) is highly sensitive to the moisture advection scheme. TSPAS successfully improved precipitation over the eastern slope of the TP, especially for torrential rainfall. The fractions skill score (FSS) is improved by 0.075 (27.78%) for daily precipitation with a threshold of 100 mm. Compared with the experiment with the original WRF advection scheme, the TSPAS reduced the overestimation of precipitation in the topographic region and excessive water vapor transport in a low-level atmosphere. To understand the precipitation improvement contributed by the advection scheme, additional experiments were conducted for a particular precipitation process from two approaches: switching advection schemes during the rainfall evolution and updating the variables related to moisture advection individually. Results demonstrate that the precipitation improvement is mainly contributed by the moisture advection scheme before the precipitation. Among the different variables, the combination of wind and water vapor was the most influential factor causing the precipitation improvement under the TSPAS.

Impacts of Atlantic and Pacific Multidecadal Variability on South American Precipitation and Temperature in the CESM Simulations and Observations

AbstractThe Community Earth System Model version 1 (CESM1) and version 2 (CESM2)'s abilities to simulate the impacts of Atlantic multidecadal variability (AMV) and Pacific multidecadal variability (PMV) on South American precipitation and temperature have not been assessed, and how the AMV and PMV modulate each other's influences on South American climate is not well understood. Here we use observations, reanalyses, and CESM1 and CESM2 simulations from 1920 to 2015 to study those problems. The models can reproduce the observed precipitation and temperature responses to AMV well, but can only roughly reproduce such responses to PMV. The precipitation response over the South Atlantic convergence zone (SACZ) is better simulated by CESM2 compared to CESM1, which is associated with an improved horizontal moisture flux over this region. However, the models cannot accurately simulate the observed differences between the influences of Pacific interannual and multidecadal variability on South American precipitation and temperature. The impacts of AMV and PMV on South American precipitation are modulated by the other mode via changes in horizontal moisture flux over the SACZ and River Plate basin in summer, as well as changes in vertical motion over the equatorial regions in winter. Similarly, the impacts of AMV and PMV on South American temperature are also modulated by the other mode. Over water‐limited regions, such as northeastern Brazil and southern Argentina, the precipitation and temperature responses are anti‐correlated, possibly via surface evaporation.

Bimodality in simulated precipitation frequency distributions and its relationship with convective parameterizations

Bimodality in precipitation frequency distributions is often evident in atmospheric models, but rarely in observations. This study i) proposes a metric to objectively quantify the bimodality in precipitation distributions, ii) evaluates model simulations contributed to the Coupled Model Intercomparison Project (CMIP) phase 5 (CMIP5), phase 6 (CMIP6), and the DYnamics of the Atmospheric general circulation Modeled On Non-hydrostatic Domains (DYAMOND) project by comparing them to satellite-based and reanalysis precipitation products, and iii) investigates possible origins of bimodal precipitation distributions. Our results reveal that about 83% (20 out of 24) of CMIP5 and 70% (21 out of 30) of CMIP6 models used in this study exhibit bimodal distributions. The few DYAMOND models that use a deep convective parameterization also show bimodal distributions, while most DYAMOND models do not. Predictably, the bimodality originates from the separation of precipitation process between resolved grid-scale and parameterized subgrid-scale. However, in a larger number of models bimodality arises from the parameterized subgrid-scale convective precipitation alone.

Deep‐learning‐based downscaling of precipitation in the middle reaches of the Yellow River using residual‐based CNNs

AbstractThe middle reaches of the Yellow River (MRYR), located in northern China, are the transition zone between semi‐arid and semi‐humid climates. As one of the climate‐sensitive regions in China, MRYR has a fragile ecological environment and serious soil loss, which leads to geological disasters such as landslides, collapses, and mudslides caused by extreme precipitation. However, scarceness of high‐resolution precipitation data over MRYR limits assessment of the environmental impacts caused by climate change, especially for extreme precipitation. In this article, we design a Residual‐in‐Residual Dense Block based Network (RRDBNet) model for the statistical downscaling of precipitation in MRYR, and compare the proposed RRDBNet with a generalized linear regression model (GLM) and two popular deep‐learning‐based models. The multi‐level residuals and dense connectivity strategies introduced in RRDBNet help it to learn more abstract features and complex nonlinear relationships among climate variables to improve downscaling performance. The results show that the proposed RRDBNet has good performance in precipitation simulations, which can reproduce the spatial–temporal characteristics of high‐resolution precipitation well. RRDBNet reduces the root‐mean‐squared error (RMSE) by 19% and improves the Pearson correlation coefficient (CC) by 6% relative to GLM for climatology mean precipitation. Especially, RRDBNet has substantial improvements in extreme precipitation compared with other models. It reduces RMSE by 58% (79%) and improves CC by 38% (145%) relative to GLM for R95P (R99P), where R95P and R99P represent extreme precipitation and very extreme precipitation, respectively. For the probability density function of daily precipitation, it is further demonstrated that RRDBNet performs better as regards extreme precipitation frequency. Our results suggest that statistical downscaling based on RRDBNet may be an effective tool for historical and future climate simulations from global climate models.