High-intensity Precipitation Research Articles

DINÂMICA HÍDRICA DO SOLO DE FITOFISIONOMIA DE CERRADO RALO DO CHAPADÃO DO DIAMANTE - SERRA DA CANASTRA (MG)

The Cerrado is characterized as an important Brazilian biome, forming strategic hydrographic basins within its territory that are significant both nationally and internationally. Despite its importance, the biome has been increasingly impacted by anthropogenic activities. In this context, this study aims to analyze and understand the physical-hydraulic characteristics of the soil in a sparse Cerrado physiognomy located in the Chapadão do Diamante (Serra da Canastra-MG), providing the necessary foundation for the assessment and development of future management strategies for similar areas. For field data prospecting, a rain simulator and a concentric ring infiltrometer were used. The results demonstrated a high infiltration capacity in the study area, with no significant surface runoff even under high-intensity precipitation (57.35 mm), revealing high values of basic infiltration velocity (BIV) at 626.56 mm/h. Overall, these values are associated with the dynamics of landscape elements, emphasizing the importance of biological processes, such as the role of vegetation and soil fauna, in modulating the soil's ability to retain, infiltrate, and store water. Keywords: Water Infiltration; Rain Simulator; Concentric Ring Infiltrometer.

Incorporation crisis lifecycle theory into full-stage flash drought spatio-temporal pattern identification and risk analysis

Compared to conventional slow-developing droughts, the characteristics and drivers of flash droughts are often more complex. This complexity challenges global drought early warning systems, leading to severe losses in crop yields during flash drought events. To address this problem, our study introduces the Full Life Cycle Flash Drought Identification Framework (FLCFDIF), which incorporates crisis lifecycle theories to link the lifecycle progression of flash droughts with corresponding risk management measures. The results indicate significant differences in the risk characteristics of flash drought events are observed between monsoon and non-monsoon regions of China, with flash droughts in monsoon regions erupting more rapidly, lasting longer, and recovering more slowly than flash droughts in non-monsoon areas. Flash droughts typically feature negative precipitation anomalies and positive anomalies in maximum temperature and evapotranspiration during the progression of the droughts. However, the high-intensity precipitation events that drive flash drought recovery primarily occur during the termination phase rather than the previously assumed recovery phase. Therefore, employing a 40% soil moisture percentile as a threshold for determining the end of flash droughts can reduce disturbances from low-intensity precipitation events and establish a link between flash droughts and seasonal droughts. In Southeast China, the probability of flash droughts transitioning to seasonal droughts is approximately 10%. In future flash drought risk management, it is crucial to integrate the four stages of flash drought progression (development, duration, recovery, and termination) and implement timely emergency response actions, including warning, assessment, mitigation, and monitoring, to address the environmental impacts of flash drought events.

Diurnal Cycle of Precipitation and Extremes in Nepal

This study examines the spatio-temporal distribution of hourly precipitation across Nepal using Integrated Multi-Satellite Retrievals for Global Precipitation Measurement (IMERG) data from 2015 to 2021. The findings indicate that hourly precipitation intensity during the monsoon season can reach up to 0.7 mm/hr, while pre-monsoon intensities peak at 0.2 mm/hr. These intensities are predominantly concentrated in mid- and low-elevation areas of central and eastern Nepal. Post-monsoon and winter seasons exhibit high-intensity precipitation patches over high-elevation regions in the western, central, and eastern peripheries of the country. The annual distribution of hourly precipitation shows a pronounced peak during the monsoon season, indicating that a significant portion of the total annual precipitation occurs during this period. Extreme precipitation events follow a similar seasonal distribution, with monsoon extremes exceeding 15 mm/hr. The diurnal cycle of monsoonal precipitation shows unique characteristics, peaking at 0.65 mm/hr around midnight and decreasing to 0.2 mm/hr by late morning, then increasing steadily in the afternoon and evening. Additionally, the study contrasts hourly heavy precipitation extremes (≥10 mm/hr or day) with daily extremes, noting that hourly extremes, despite being accumulated into daily measures, present a higher frequency, and pose greater short-term hazard risks. This analysis over a seven-year period suggests the need for continued research to determine spatial and temporal trends in hourly versus daily extreme precipitation patterns, particularly in the context of climate change and its impacts on regional hydrology and meteorology.

Drainage water management, woodchip bioreactor, and saturated riparian buffer as stacked conservation practices for improving crop yields and water quality

Stacking edge-of-field practices may improve nutrient removal from crops. To examine the effects of stacking edge-of-field conservation practices, a woodchip bioreactor (WBR) and saturated riparian buffer (SRB) were installed in series by intercepting tile flow from a field having a drainage water management system. Nutrient monitoring from 5 years evaluated nutrient export annually and based on the precipitation intensity. Drainage water was monitored for total suspended solids (TSS), nitrate-N, total-P, total-N, and ortho-P at the inlet and outlet of WBR and control structure of SRB. Nutrient export reductions of WBR and SRB were determined for precipitation events that were categorized as low <12.7 mm, mid 12.7–25.4 mm, high 25.4–50.8 mm, and very high >50.8 mm. Over the five seasons, nitrate-N export was reduced 88 % at the WBR outlet and 78 % at SRB outlet when used in a stacked series configuration. The efficacy of edge-of-field practices was affected by the intensity of precipitation events. The low and mid-intensity precipitation events generated 67 % of the total discharge from the subsurface drainage system which accounted for 74 % of the influent nitrate-N. During low and mid-intensity precipitation events, discharge was reduced by 58–65 %, nitrate-N was reduced by 49–88 % and total-P was reduced by 65–73 % by using stacked practice of WBR and SRB. During high and very high-intensity precipitation events only nitrate-N export was reduced by 61–66 %. This indicates that when designing stacked edge-of-the-field practices the cumulative effect of the practices and their performance during different precipitation events should be taken into account when managing conservation practice-based cropping systems.

Socio-Economic Vulnerability to Potential Flood in Tamakoshi River in Nepal

Tamakoshi, a transboundary river flowing from Tibet, China to Nepal has experienced several disastrous flash floods resulting in the loss of lives, infrastructures, settlements, lands, and other assets. Many households, populations, and public and private properties such as houses, land, roads, bridges, etc. are likely to be affected by the potential flash floods. This study discusses the occurrence of flash floods, losses and damages from such floods in the past, and socio-economic vulnerability to potential flash floods. It further discusses the preparedness and response activities and identifies the measures for flood risk reduction. The methods adopted for data collection included observation, household surveys, and focus group discussions. The villages selected for detailed study are Gongarkhola, Jagat, Jamune, Bhorle, and Sigati bazaar, identified as the most vulnerable settlements along the Tama Koshi River. It is found that flash floods and high-intensity precipitation occur frequently in the study area resulting in loss of life and properties, and people and properties are exposed to the potential flash floods. Local people are still feeling vulnerable to such flash floods. The government authorities alone are not able to achieve greater success in carrying out preparedness and response activities for flood risk reduction without the active involvement and participation of the vulnerable communities. Development and effective implementation of land-use guidelines and building codes, watershed management, early warning systems, and networking among the people residing across the river are some of the activities identified to reduce flood risk. Although people have a good understanding and knowledge about floods and their occurrences, they pay less attention to preparedness and mitigation. It is also concluded that structural measures alone are not sufficient and effective enough for flood risk reduction

On the suitability of a convolutional neural network based RCM-emulator for fine spatio-temporal precipitation

High resolution regional climate models (RCM) are necessary to capture local precipitation but are too expensive to fully explore the uncertainties associated with future projections. To resolve the large cost of RCMs, Doury et al. (2023) proposed a neural network based RCM-emulator for the near-surface temperature, at a daily and 12 km-resolution. It uses existing RCM simulations to learn the relationship between low-resolution predictors and high resolution surface variables. When trained the emulator can be applied to any low resolution simulation to produce ensembles of high resolution emulated simulations. This study assesses the suitability of applying the RCM-emulator for precipitation thanks to a novel asymmetric loss function to reproduce the entire precipitation distribution over any grid point. Under a perfect conditions framework, the resulting emulator shows striking ability to reproduce the RCM original series with an excellent spatio-temporal correlation. In particular, a very good behaviour is obtained for the two tails of the distribution, measured by the number of dry days and the 99th quantile. Moreover, it creates consistent precipitation objects even if the highest frequency details are missed. The emulator quality holds for all simulations of the same RCM, with any driving GCM, ensuring transferability of the tool to GCMs never downscaled by the RCM. A first showcase of downscaling GCM simulations showed that the RCM-emulator brings significant added-value with respect to the GCM as it produces the correct high resolution spatial structure and heavy precipitation intensity. Nevertheless, further work is needed to establish a relevant evaluation framework for GCM applications.

U-HRNet+: a real-time precipitation estimation method based on the fusion of temporal and spatial domain features

ABSTRACT The precipitation estimation with high spatial and temporal resolution over Tibet is of great significance for monitoring the stability of the plateau ecosystem and early warning of natural disasters. In response to the problem of the unsatisfactory accuracy in traditional real-time precipitation inversion methods and quantitative accuracy estimation products in this region. This paper adopts deep-learning methods to explore the corresponding relationship between precipitation and multi-band observation data from geostationary satellite imagery. The U-High Resolution Network (U-HRNet) as backbone is fully exploited to generate multi-scale features and fused using attention gates to extract high-resolution spatial representations representing precipitation clouds, thereby achieving precipitation estimation of target clouds at different scales. In addition, in response to the characteristics of high precipitation intensity and drastic cloud cluster changes in summer convective weather within a short period of time, a multi-head attention mechanism is introduced to extract temporal information from time series inputs and fuse it with spatial features to improve the accuracy of real-time precipitation estimation. Compared with the operational precipitation products and baseline deep-learning models, the classification metrics such as Probability of detection (POD), False Alarm Ratio (FAR) and Critical success index (CSI) show that the proposed model (U-HRNet+) has better performance in test set and has high generalization ability for estimating short-term heavy rainfall over Tibet and its surrounding areas in summer. Therefore, the proposed model has the potential to serve as a more accurate and reliable satellite-based precipitation estimation product.

Global Precipitation Nowcasting of Integrated Multi-satellitE Retrievals for GPM: A U-Net Convolutional LSTM Architecture

Abstract This paper presents a deep supervised learning architecture for 30-min global precipitation nowcasts with a 4-h lead time. The architecture follows a U-Net structure with convolutional long short-term memory (ConvLSTM) cells empowered by ConvLSTM-based skip connections to reduce information loss due to the pooling operation. The training uses data from the Integrated Multi-satellitE Retrievals for GPM (IMERG) and a few key drivers of precipitation from the Global Forecast System (GFS). The impacts of different training loss functions, including the mean-squared error (regression) and the focal loss (classification), on the quality of precipitation nowcasts are studied. The results indicate that the regression network performs well in capturing light precipitation (&lt;1.6 mm h−1), while the classification network can outperform the regression counterpart for nowcasting of high-intensity precipitation (&gt;8 mm h−1), in terms of the critical success index (CSI). It is uncovered that including the forecast variables can improve precipitation nowcasting, especially at longer lead times in both networks. Taking IMERG as a relative reference, a multiscale analysis, in terms of fractions skill score (FSS), shows that the nowcasting machine remains skillful for precipitation rate above 1 mm h−1 at the resolution of 10 km compared to 50 km for GFS. For precipitation rates greater than 4 mm h−1, only the classification network remains FSS skillful on scales greater than 50 km within a 2-h lead time. Significance Statement This study presents a deep neural network architecture for global precipitation nowcasting with a 4-h lead time, using sequences of past satellite precipitation data and simulations from a numerical weather prediction model. The results show that the nowcasting machine can improve short-term predictions of high-intensity global precipitation. The research outcomes will enable us to expand our understanding of how modern artificial intelligence can improve the predictability of extreme weather and benefit flood early warning systems for saving lives and properties.

Microphysical Characteristics of Precipitation Over Eastern China and Its Coastal Regions

AbstractMany studies have investigated the macro‐ and micro‐physical properties of precipitation, but few classified precipitation into warm‐ and cold‐topped, which affects precipitation phase and precipitation type significantly. Furthermore, the impact of moisture and dynamics on microphysical characteristics of precipitation has been less studied. Using precipitation data from the GPM 2A DPR between May and September from 2015 to 2023, this study conducts a systematic study on microphysical characteristics of precipitation over eastern China and its coastal areas. This study divides precipitation cases into 20 categories, including warm‐/cold‐topped (CTP/WTP) convective/stratiform light/moderate/heavy/extreme/the most extreme precipitation. The findings reveal distinct differences along with similarities in various precipitation categories between eastern China and its coastal regions. The convection ratio for CTP increases with higher precipitation intensity, while WTP is dominated by convective precipitation. The more CTP, the higher the solid water content. The heavier precipitation, the higher liquid water content contributes. The mass‐weighted diameter (Dm) for CTP shows three variations from high to low with little or slight decrease with altitude at upper‐layer, followed by a rapid increase at middle‐layer and finally decreasing slightly at low‐layer. The CTP Dm grows rapidly with temperature from −18.58 (−13.93)°C for convective (stratiform) precipitation and their maximum growth rate are almost located in melting layer. Dm (raindrop density) is larger (lower) over land than ocean and larger (lower) for CTP than WTP. Dm shows positive relation to vertical velocity and specific humidity, especially sensitive to the middle‐ and upper‐layers meteorological conditions.

Analysis of heavy precipitation-induced rill erosion

Erosion is an ongoing environmental problem that leads to soil loss and damages ecosystems downstream of agriculture. Increasingly frequent heavy precipitation causes single erosion events with potentially high erosion rates owing to gully erosion. In this study, analyses of croplands affected by heavy precipitation and linear erosion indicate that erosion occurs only on sparsely vegetated fields with land cover ≤ 25% and that slope gradient and length are significant factors for the occurrence of linear erosion tracks. Existing erosion models are not calibrated to the conditions of heavy precipitation and linear erosion, namely high precipitation intensities and long and steep croplands. In this study, natural linear erosion was analyzed using an unmanned aerial vehicle and erosion volumes were determined for 32 rills and gullies of different sizes. Comparisons with the RUSLE2 and EROSION-3D model values showed an underestimation of linear erosion in both models. Therefore, calibration data for erosion models used for heavy precipitation conditions must be adapted. The data obtained in this study meet the required criteria.

How is climate change altering the precipitation pattern over Northwestern India?

AbstractClimate change in India is causing devastating downpour events and shifts in spatial characterization, with apparent regional differences. The Northwest India (NWI), which is in the Aravalli rain shadow zone and was formerly drier, has attracted great attention in recent years due to its changing rainfall patterns. This study throws light on the astonishing behaviour of ISMR over NWI with the advent of intensified rainfall over the region in the recent time frame. The Pettitt test of change point (CP) detection is utilized in this analysis to measure the change in rainfall patterns over a period of 70 years, from 1950 to 2019. This analysis suggests significant variation in the CP's time frame for different months of ISMR. The earliest change was noticed in July, while the latest was for September for the mean as well as for the intense precipitation. Moreover, we found a maximum change in the precipitation during the peak monsoon month (i.e., July and August). The difference in the precipitation of different percentile values before and after CP indicates a decrease (increase) in low (high) intensity precipitation for all the months and seasons as a whole with varying magnitude. The highest reduction of low‐intensity precipitation is noticed for the months of July and August, while the highest increase of high‐intensity precipitation (&gt;95th percentile) is noticed for June and September. The heavier precipitation contributes largely to the mean increase of precipitation, expecting to receive more mean and precipitation extremes in the future. Unlike the increasing precipitation trend, potential evapotranspiration (source of local moisture) shows a declining trend, revealing the negative association among them and the possibility of enhanced advection of remote moisture responsible for enhanced precipitation over NWI. The enhanced vertically integrated moisture transport of the Arabian Sea and Bay of Bengal and its strengthening relationship with precipitation further confirm the contribution of remote moisture to intensified precipitation over NWI, though the in‐depth dynamical cause remains unclear. The increased Convective Available Potential Energy for entire monsoon seasons as a whole and individual months facilitates a favourable condition for enhanced convective activity over the region, resulting in strengthening the precipitation.

Research on Frequency Matching Correction Techniques for South China Precipitation Ensemble Forecast Based on the GRAPES Model

This study focuses on the real-time precipitation forecast products of the GRAPES_MESO regional ensemble forecast model, which is developed by the Numerical Weather Prediction Center of the China Meteorological Administration and is initialized 1–3 days in advance at 12:00 UTC. Using a national-level homogenized precipitation grid dataset from surface meteorological stations as observational data, a frequency matching method (FMM) is employed to correct precipitation forecasts for different precipitation intensity levels, including light rain, moderate rain, heavy rain, and torrential rain. Case studies and statistical tests (TS scores) are conducted to compare the forecast performance before and after correction. The results indicate that the model’s Cumulative Distribution Function (CDF) curves deviate from observations, and the longer the lead time, the more significant the error. The correction coefficients (CCs) show an increasing trend with the growth of precipitation intensity, indicating that for larger precipitation amounts and longer lead times, larger CCs are needed, highlighting the necessity of correction. Analyzing two precipitation events in South China in July 2019, the FMM results in an increase in precipitation intensity and a widening of the range of heavy precipitation. The corrected precipitation magnitudes are closer to the observations. The statistical tests using TS scores reveal that the FMM has a certain correction effect on the overall precipitation forecast in the South China region, especially for longer lead times and higher precipitation intensities, where the correction effect is more significant. The necessity of frequency matching correction becomes more apparent for heavier precipitation, and the correction effect becomes more significant with longer lead times.

The Effect of Convective/Advective Precipitation Partitions on the Precipitation Isotopes in the Monsoon Regions of China: A Case Study of Changsha

Abstract Convective/advective precipitation partitions refer to the divisions of precipitation that are either convective or advective in nature, relative to the total precipitation amount. These distinct partitions can have a significant influence on the stable isotope composition of precipitation. This study analyzed and compared the effect of precipitation partitions on δ18O in precipitation (δ18Op) by using daily precipitation stable isotope data from Changsha station and monthly precipitation stable isotope data from the Global Network of Isotopes in Precipitation (GNIP), under different time scales, time intervals (i.e., annual, warm season, and cold season), and precipitation intensities. The results showed that the correlation between the convective precipitation fraction (CPF) and total precipitation amount was influenced by the intensity of convection in different time intervals. On both the daily and monthly scales, the CPF decreased as the precipitation amount increased in the warm season, while it increased with increasing precipitation amount in the cold season. Regardless of the season, daily δ18Op at Changsha station consistently increased with an increase in daily CPF. On a daily scale, the effect of convective activity on δ18Op was stronger than that of the “precipitation amount effect” in the cold season, as compared to the situation in the warm season. As a result, the regression line slope between δ18Op and CPF increased with increasing precipitation intensity in the warm season, meaning that as the CPF increased, the δ18Op increased at a faster rate under higher precipitation intensity. Similarly, the slope increased with increasing precipitation intensity in the cold season. This suggests that precipitation intensity and convection intensity can affect the relationship between δ18Op and CPF. Our findings shed light on how different precipitation partitions affect stable isotope composition of precipitation, thus enhancing our understanding of the variability of precipitation stable isotopes in the monsoon regions of China. Significance Statement This study aims to better elucidate the influence of different precipitation partitions on precipitation stable isotopes. In the eastern monsoon region of China, stable isotopes in precipitation showed a robust positive relationship with convective precipitation faction. On a daily scale, the convective activity enhanced the influences of the “precipitation amount effect” on precipitation stable isotopes in the warm season and reduced such influences in the cold season. These results improve our understanding of stable isotopic variability of precipitation in the eastern monsoon region, China.

Quantile-Regression-Ensemble: A Deep Learning Algorithm for Downscaling Extreme Precipitation

Global Climate Models (GCMs) simulate low resolution climate projections on a global scale. The native resolution of GCMs is generally too low for societal-level decision-making. To enhance the spatial resolution, downscaling is often applied to GCM output. Statistical downscaling techniques, in particular, are well-established as a cost-effective approach. They require significantly less computational time than physics-based dynamical downscaling. In recent years, deep learning has gained prominence in statistical downscaling, demonstrating significantly lower error rates compared to traditional statistical methods. However, a drawback of regression-based deep learning techniques is their tendency to overfit to the mean sample intensity. Extreme values as a result are often underestimated. Problematically, extreme events have the largest societal impact. We propose Quantile-Regression-Ensemble (QRE), an innovative deep learning algorithm inspired by boosting methods. Its primary objective is to avoid trade-offs between fitting to sample means and extreme values by training independent models on a partitioned dataset. Our QRE is robust to redundant models and not susceptible to explosive ensemble weights, ensuring a reliable training process. QRE achieves lower Mean Squared Error (MSE) compared to various baseline models. In particular, our algorithm has a lower error for high-intensity precipitation events over New Zealand, highlighting the ability to represent extreme events accurately.

Research on the Characteristics of Raindrop Spectrum and Its Water Vapour Transport Sources in the Southwest Vortex: A Case Study of 15–16 July 2021

This study investigated the convective weather features, precipitation microphysical characteristics, and water vapour transport characteristics associated with a southwest vortex precipitation event that occurred on the eastern edge of the Qinghai–Tibet Plateau, coinciding with a southwest vortex event, from 15 to 16 July 2021, using conventional observations of raindrop spectra, ERA5 reanalysis data, CMORPH precipitation data, and the HYSPLIT_v4 backward trajectory model. The findings aim to provide theoretical insights for improving the forecasting and numerical simulations of southwest vortex precipitation events. The findings revealed that the precipitation event induced by the southwestern vortex at Emeishan Station on 15–16 July 2021 was characterised by high rainfall intensity and significant precipitation accumulation. The raindrop spectrum exhibited a broad distribution with a notable bimodal structure. Both the Sichuan Basin and the Tibetan Plateau were dominated by the South Asian high pressure at higher altitudes, while a pronounced low-pressure system developed at mid and low altitudes within the basin, establishing a meteorological context marked by upper-level divergence and lower-level convergence. Throughout the event, notable vertical uplift velocities were recorded across the Sichuan Basin and Tibetan Plateau, along with distinct positive vorticity zones in the lower and middle strata of the Sichuan Basin, indicating that the atmosphere was in a state of thermal instability. The majority of moisture was in the mid and lower troposphere with evident convergence movements, which played a crucial role in the southwest vortex’s development. WRF numerical simulations of the Emeishan precipitation event more accurately modelled the weather conditions for this precipitation but tended to overestimate the level of precipitation. It was observed that the region around Emei Mountain primarily received moisture influx from the southern Bay of Bengal and the South China Sea, with moisture transport chiefly originating from the Sichuan Basin and in a south-westward trajectory.

Flow patterns at river bends and its response to surrounding infrastructures

On January 6, 2023, high-intensity precipitation occurred in the Babon River watershed, causing the river overflow and flood in several areas. Severe impact in the surrounding area was observed, such as in the Dinar Indah Housing Complex. Hydrodynamic complexity around a river bend points out the importance of understanding the flow patterns at a river bend. Therefore, this study aimed to describe the flow pattern of the river bend and analyze its response to the surrounding infrastructure using the HEC-RAS 6.3 model. Furthermore, improvement to structural mitigation was also designed. Field observations were also conducted to gather field data on the flood impact. The model results indicate that the highest water level within the river bend is 1.9 m, and the highest velocity is 5.85 m/s, which is on the outer bend. A dyke was then planned to be about 3.75 meters high, accompanied by a retaining wall to protect the river bank using the interlocking permeable revetment (IPR). Dyke and retaining walls provide stability against shear, overturning, and collapse.

Statistical characteristics of convective and stratiform precipitation during the rainy season over South China based on GPM-DPR observations

South China is one of the wettest regions in the world, and is prone to precipitation-related disasters. The GPM-DPR product is utilized in this study to investigate the vertical structure and drop size distribution (DSD) characteristics of different precipitation types, i.e., stratiform precipitation (SP) and convective precipitation (CP) in South China. The precipitation frequency and accumulation in South China are dominated by SP and deep CP, respectively. Shallow CP has comparable occurrence ratio to deep CP, but the mean intensity is much lower. As such, it contributes <5% to the total rainfall. Vertical features differ among precipitation types and intensities. Higher storm top altitude (STA) is roughly related to more intense precipitation, especially for CPs. As surface PR increases in deep CP, the contribution of increase of PR (ΔPR) in liquid-phase region to total PR also increases, from approximately 10% for 4 mm/h to 55% for 100 mm/h. The intensity of shallow CP is generally low, but ΔPR in the liquid-phase region is significant. For SP and shallow CP, higher precipitation intensity is often caused by higher hydrometer concentration. Differently, for deep CP with PR >15 mm/h, both the increases in hydrometer size and concentration contribute to the increase in PR. The collision-coalescence process is dominant in the low-level layers in SP without bright band and CP, and approximately four fifth of shallow CP samples are influenced by collision-coalescence process. Compared with inland mountainous region, collision-coalescence process plays more important role in intense precipitation in coastal region. The results of this study indicate that it is important to consider the vertical evolution and DSD characteristics of different precipitation types for improving the accuracy of near surface quantitative precipitation estimate.

Root distributions predict shrub–steppe responses to precipitation intensity

Abstract. Precipitation events are becoming more intense around the world, changing the way water moves through soils and plants. Plant rooting strategies that sustain water uptake under these conditions are likely to become more abundant (e.g., shrub encroachment). Yet, it remains difficult to predict species responses to climate change because we typically do not know where active roots are located or how much water they absorb. Here, we applied a water tracer experiment to describe forb, grass, and shrub root distributions. These measurements were made in 8 m by 8 m field shelters with low or high precipitation intensity. We used tracer uptake data in a soil water flow model to estimate how much water respective plant root tissues absorb over time. In low-precipitation-intensity plots, deep shrub roots were estimated to absorb the most water (93 mm yr−1) and shrubs had the greatest aboveground cover (27 %). Grass root distributions were estimated to absorb an intermediate amount of water (80 mm yr−1) and grasses had intermediate aboveground cover (18 %). Forb root distributions were estimated to absorb the least water (79 mm yr−1) and had the least aboveground cover (12 %). In high-precipitation-intensity plots, shrub and forb root distributions changed in ways that increased their water uptake relative to grasses, predicting the increased aboveground growth of shrubs and forbs in these plots. In short, water uptake caused by different rooting distributions predicted plant aboveground cover. Our results suggest that detailed descriptions of active plant root distributions can predict plant growth responses to climate change in arid and semi-arid ecosystems.

Deep learning model based on multi-scale feature fusion for precipitation nowcasting

Abstract. Forecasting heavy precipitation accurately is a challenging task for most deep learning (DL)-based models. To address this, we present a novel DL architecture called “multi-scale feature fusion” (MFF) that can forecast precipitation with a lead time of up to 3 h. The MFF model uses convolution kernels with varying sizes to create multi-scale receptive fields. This helps to capture the movement features of precipitation systems, such as their shape, movement direction, and speed. Additionally, the architecture utilizes the mechanism of discrete probability to reduce uncertainties and forecast errors, enabling it to predict heavy precipitation even at longer lead times. For model training, we use 4 years of radar echo data from 2018 to 2021 and 1 year of data from 2022 for model testing. We compare the MFF model with three existing extrapolative models: time series residual convolution (TSRC), optical flow (OF), and UNet. The results show that MFF achieves superior forecast skills with high probability of detection (POD), low false alarm rate (FAR), small mean absolute error (MAE), and high structural similarity index (SSIM). Notably, MFF can predict high-intensity precipitation fields at 3 h lead time, while the other three models cannot. Furthermore, MFF shows improvement in the smoothing effect of the forecast field, as observed from the results of radially averaged power spectral (RAPS). Our future work will focus on incorporating multi-source meteorological variables, making structural adjustments to the network, and combining them with numerical models to further improve the forecast skills of heavy precipitations at longer lead times.

Simulation of Long-Term Water Erosion in Rainfed Croplands of Eastern Washington

Highlights Water erosion in the inland Pacific Northwest was modeled with WEPP for the past (1940–1982) and present (1983–2020). Water erosion decreased from past to present in the study watersheds. Major factors are increased adoption of conservation practices and decrease in large precipitation events. Abstract. Water erosion is an ongoing problem in eastern Washington due to its hilly terrain, highly erodible silt loam soils, rain on thawing soil, and the prevalence of conventional tillage. The region is characterized by a Mediterranean-type climate with warm, dry summers and cool, wet winters. Three distinct precipitation zones, with annual totals low (&amp;lt;380 mm), intermediate (380–460 mm), and high (&amp;gt;460 mm), dictate the area’s crop rotations. A unique 43-year (1940–1982) dataset of winter erosion measured on multiple agricultural fields in Whitman County, eastern Washington, by Verle Kaiser, a USDA Soil Conservation Service agronomist, showed annual erosion rates averaging 53.8 Mg ha-1, far exceeding the current Natural Resources Conservation Service tolerable limit of 11 Mg ha-1 yr–1 for the soils in the area. Kaiser’s field data allowed us to compare the historical field-measured erosion rates with those simulated by the WEPP (Water Erosion Prediction Project) model. Anthropogenic factors, such as tillage and crop rotation, change with time. Conservation tillage, including reduced- and no-till, has been increasingly adopted in eastern Washington since the mid-1980s. The specific objectives of this study were to (1) apply the WEPP model to simulate soil erosion in eastern Washington and evaluate the interactive effects of climate and management, in addition to topography and soil, on water erosion in the study area, and (2) compare the simulation results with Kaiser’s historical field dataset and elucidate the long-term soil erosion trend. The WEPPcloud interface was used to delineate a watershed within each precipitation zone of the study area. Climate inputs were divided into two periods: the past (1939–1982) and the present (1983–2020). Erosion has noticeably decreased from the past to the present, with WEPP simulated annual erosion averaging 13.5, 34.5, and 52.6 Mg ha-1 for the past, and 9.5, 14.1, and 15.5 Mg ha-1 for the present, in the selected watersheds in the low-, intermediate-, and high-precipitation zones, respectively. The decreasing trend was primarily due to the increased adoption of conservation tillage and crop rotation, as well as a decrease in the number of high-intensity precipitation events in the present climate. Keywords: Inland Pacific Northwest, Soil erosion by water, Temporal trend, WEPP.