Spatial Distribution Of Rainfall Research Articles

Rainfall forecasting in arid regions in response to climate change using ARIMA and remote sensing

Climate changes in increasing at alarming rates which caused outward changes in the hydrological cycle parameters such as temperature and precipitation. The spatial and temporal distribution of rainfall has changed all over the world. Rainfall forecasting is considered an important tool that could help decisionmakers for managing water resources. Syria is facing several drought events that may affect its water resources. Due to the lack of available rainfall data from land stations in Syria after the civil war started in 2010 till now, one of the aims of this study is to depend on rainfall data from satellite images after correcting these data using the available land stations data. This study also aims to assess rainfall trends in Syria due to climate change and forecasting rainfall. The rainfall data were collected from 71 land stations from 1991 to 2009. Satellite images from Climate Data Record (CDR) were downloaded from 1983 to 2020 and ArcMap was used to extract rainfall data at locations of land stations. The extracted rainfall data from satellite images were corrected by land stations data from 1991 to 2009. Then Statistical Package for the Social Sciences (SPSS) program with Auto-Regressive Integrated Moving Average (ARIMA) models are used to fit and forecasting rainfall in Syria. The rainfall trends at the studied stations were analyzed for 40 years (1983–2022) and a forecast of one year is conducted. The results showed decreasing trends at all the stations that lead to a decrease the quantity of water received every year which could affect the water resources in Syria and agriculture and food security will be also affected. The results are consistent with Intergovernmental Panel on Climate Change (IPCC) reports that concluded a decrease in rainfall at southern and eastern Mediterranean regions. The results of the current study could help in the management of water resources in Syria and forecasting using ARIMA is recommended for rainfall predictions.

Rainfall prediction for data scares areas using metrological satellites in the case of the Lake Tana sub-basin, Ethiopia

Abstract In African nations with complex topographies, alternative rainfall estimation methods such as satellites are crucial. This study is aimed at predicting the spatial and temporal distribution of rainfall in the Lake Tana sub-basin from 1990 to 2020. A satellite-based rainfall estimate of Climate Hazards Group Infrared Precipitation with Station data (CHIRPS) was used with the same spanning period (1990–2020). The validation process employs point-to-pixel analysis, comparing CHIRPS estimates with observed data at specific gauge stations. The findings showed that CHIRPS had well-estimated rainfall incidence in the highland areas and significantly overestimated it in the lowland areas. The Mann–Kendall trends for January, June, and August indicate decreasing trends, while the Bega and spring seasons show notable declines. Regression analysis reveals a non-significant decrease in annual rainfall with the highest rainfall in the summer and relatively dry winters. In addition, the coefficient of variation value of 26.37% suggests a moderate level of variability around the mean annual rainfall. In conclusion, the CHIRPS satellite exhibited varied performance across the Tana Sub-basin, with site-specific discrepancies and notable inaccuracies at certain stations. The study underscores the importance of considering local factors and topography in satellite-based rainfall assessments, providing valuable insights for agricultural planning in the region.

Black carbon emissions and its impact on the monsoon rainfall patterns over the Indian subcontinent: Insights into localized warming effects

This study investigates the impact of black carbon (BC) emissions on monsoon rainfall patterns across the Indian subcontinent. The results show that BC exerts localized warming effects, providing valuable insights into the mechanisms influencing the rainfall distribution across the Indian landmass. The study analyzes the vertical profile of mean tropospheric temperature differences between two sets of simulations: (i) with default BC emissions (WBC) and (ii) with BC emissions reduced by 99% (WoBC), conducted for JJAS, 2017, focusing on the region along the eastern coast. The analysis of the tropospheric temperature variations over the eastern coast reveals significant rainfall differences, primarily attributed to the intensification of convective rainfall. The results indicate that WoBC simulation leads to abnormal cooling in the lower troposphere and warming in the mid-upper troposphere, plausibly linked to the release of latent heat from the enhanced convective activity observed over the region. These alterations in the tropospheric temperature profile correspond remarkably well with the changes in the spatial distribution of rainfall over this area, providing valuable insights into the intricate dynamics of the climate system over this region.

Multivariate Analysis of Rainfall Spatial Distribution and Its Effect on Stormwater Magnitudes

Multivariate Analysis of Rainfall Spatial Distribution and Its Effect on Stormwater Magnitudes

Added Value of Three‐Dimensional Horizontal Winds From Geostationary Interferometric Infrared Sounder for Typhoon Forecast in a Regional NWP Model

AbstractHigh temporal resolution geostationary hyperspectral infrared sounders can simultaneously profile atmospheric temperature and moisture, and track moisture features to provide 3D horizontal winds in clear and partially cloudy scenarios. The thermodynamic information obtained has been integral to enhancing tropical cyclone (TC) forecasts. However, the potential benefits derived from the dynamic information provided by geostationary hyperspectral infrared sounders (GeoHIS) are yet to be fully comprehended for numerical weather prediction (NWP). With the re‐estimated observation error and multivariate hydrometeor background error covariance, the 3D horizontal winds from the Geostationary Interferometric Infrared Sounder (GIIRS) are synergistically assimilated with the hydrometeor information. The impact of the GIIRS‐derived 3D horizontal wind assimilation on Typhoon Maria (2018) and Lekima (2019) analysis and forecast are evaluated. The results show that the assimilating GIIRS‐derived 3D horizontal wind notably reduces the RMSEs of analysis and forecast, particularly in the U and V components. The improvement in accuracy of large‐scale fields substantially enhances the forecasts of typhoons' track and maximum wind speed, as well as the central sea‐level pressure. Moreover, the prediction of the spatial distribution and intensity for landfall precipitation is also improved. The detailed diagnoses of the Maria show that a more southerly subtropical high generated by the additional horizontal wind assimilation improves the rainfall spatial distribution, and the reasonable water vapor transportation caused by the improved dynamic conditions corrects the precipitation intensity. This work highlights the importance of dynamic information in TC forecasting and emphasizes the need for efficient and high‐precision 3D wind measurements to improve NWP.

The implication of spatiotemporal rainfall distribution on the estimation of areal rainfall characteristics in Upper Erer Sub-basin, Ethiopia

Abstract The distribution of rainfall is not uniform in various regions of the world. Rainfall study is used to recognize the characteristics, duration, and variability of temporal and spatial rainfall distributions. In Ethiopia, the annual and seasonal rainfall distributions are variable in space and time. This study is focused on the implication of spatiotemporal rainfall distribution on the areal rainfall characteristics evaluation under the Upper Erer Sub-basin, located in Eastern Ethiopia. In the study area, average annual rainfall amount for Dire Dawa, Harar and Haramaya, Girawa, and Gursum stations are found to be 647, 816, 801, 958, and 840 mm with coefficients of variation of 23, 20, 20, 19, and 31%, respectively. However, the rain gauges here are sparsely distributed. The rainfall occurrence and distribution at various gauging stations have been found to vary significantly both temporally and spatially. The rainfall occurrence and distribution at various gauging stations had been found to vary significantly both temporally and spatially. The spatiotemporal rainfall distribution found in the stations was assessed using a joint probability of rain days approach. The result indicated joint probability of rain days estimation approach under monthly time step has better performance than daily, decadal, and seasonal data. The joint probability approach is used along with rainfall amount under monthly rainfall for areal rainfall estimation assessment in rainfall–runoff modeling.

Optimal allocation of soil and water resources in agriculture under the total regional evapotranspiration indicator: A case study in Guangping, China

There is a mismatch between crop water demand, the spatial and temporal distribution of natural rainfall, surface water sources, and arable land resources in the North China Plain. This has been compensated for by massive groundwater exploitation, and as a result, ground water levels have continuously declined and groundwater landing funnels have formed in some areas. Consequently, there is an urgent need for the rational allocation of soil and water resources. In this study, regional soil types and the corresponding crops in Guangping County, Hebei Province, China, alongside the regional annual average total water consumption were used to determine reasonable allocations for crop planting structures, with an aim of maximizing the water utilization efficiency and economic benefits while also achieving a regional groundwater dynamic equilibrium. Daily meteorological data from 1973 to 2021, including precipitation, temperature, sunshine hours, wind speed, and humidity, as well as crop data, including irrigation systems and crop growth status, were collected. Water consumption models for eight soil types and seven typical crops forming 56 calculation partitions were established to obtain annual average water consumption values for the different crops. The maximum economic benefit and water-use efficiency were then used as the objective functions and total water consumption as the control condition for the non-dominated sorted genetic algorithm-Ⅱ (NSGA-II), to optimize the allocation of soil and water resources in the region; the adjusted quantities for the different crops cultivated under different soil types in the study area were obtained. The areas growing sweet potato and vegetable-fruits were found to increase by 0.5255 km2 and 4.4452 km2, respectively, when compared to the pre-optimization period, whereas the areas with winter wheat-summer maize, millet, peanut, cotton, and summer maize decreased by 1.3481 km2, 0.1812 km2, 0.7316 km2, 2.0896 km2, and 0.62 km2, respectively. The economic benefit increased from 810.37 to 853.64 million yuan, while the water-use efficiency increased from 1.5426 kg/m3 to 1.5859 kg/m3. The results provide a basis for soil and water resource management in the region and a technical reference for similar areas to improve agricultural water-use efficiency and economic benefits.

Long-Term Spatiotemporal Investigation of Various Rainfall Intensities over Central India Using EO Datasets

This study involved an investigation of the long-term seasonal rainfall patterns in central India at the district level during the period from 1991 to 2020, including various aspects such as the spatiotemporal seasonal trend of rainfall patterns, rainfall variability, trends of rainy days with different intensities, decadal percentage deviation in long-term rainfall patterns, and decadal percentage deviation in rainfall events along with their respective intensities. The central region of India was meticulously divided into distinct subparts, namely, Gujarat, Daman and Diu, Maharashtra, Goa, Dadra and Nagar Haveli, Madhya Pradesh, Chhattisgarh, and Odisha. The experimental outcomes represented the disparities in rainfall distribution across different districts of central India with the spatial distribution of mean rainfall ranges during winter (2.08 mm over Dadra and Nagar Haveli with an average of 24.19 mm over Odisha), premonsoon (6.65 mm over Gujarat to 132.89 mm over Odisha), monsoon (845.46 mm over Gujarat to 3188.21 mm over Goa), and post-monsoon (30.35 mm over Gujarat to 213.87 mm over Goa), respectively. Almost all the districts of central India displayed an uneven pattern in the percentage deviation of seasonal rainfall in all three decades for all seasons, which indicates the seasonal rainfall variability over the last 30 years. A noticeable variation in the percentage deviation of seasonal rainfall patterns has been observed in the following districts: Rewa, Puri, Anuppur, Ahmadabad, Navsari, Chhindwara, Devbhumi Dwarka, Amreli, Panch Mahals, Kolhapur, Kandhamal, Ratnagiri, Porbandar, Bametara, and Sabar Kantha. In addition, a larger number of rainy days of various categories occurred in the monsoon season in comparison to other seasons. A higher contribution of trace rainfall events was found in the winter season. The highest contributions of very light, light rainfall, moderate, rather high, and high events were found in the monsoon season in central India. The percentage of various categories of rainfall events has decreased over the last two decades (2001–2020) in comparison to the third decade (1991–2000), according to the mean number of rainfall events in the last 30 years. This spatiotemporal analysis provides valuable insights into the rainfall trends in central India, which represent regional disparities and the potential challenges impacted by climate patterns. This study contributes to our understanding of the changing rainfall dynamics and offers crucial information for effective water resource management in the region.

Simulation of Gauged and Ungauged Streamflow of Coastal Catchments across Australia

Australia is a unique continent, surrounded by the ocean, and the majority of its catchments flow to the coast. Some of these catchments are gauged and others are ungauged. There are 405 gauged catchments covering 2,549,000 km2 across the coastal regions of 12 drainage divisions in Australia, whereas there are 771 catchments conceptualised as ungauged covering additional 835,000 km2. The spatial and temporal distribution of mean annual rainfall and potential evaporation (PET) vary significantly from one drainage division to another. We developed a continuous daily streamflow time series of all gauged and ungauged catchments from 1993 onwards. We applied the daily GR4J lumped conceptual model to these catchments. The performance of gauged catchments was analysed through (i) visual inspection of daily hydrographs, flow duration curves, and daily scatter plots; and (ii) performance metrics, including NSE and PBias. Based on the NSE and PBias, performance ratings of 80% and 96% of the models, respectively, were found to be ‘good’. There was no relationship found between the catchment area and the model performance. The ungauged catchments were divided into four categories based on distance from potential donor catchments, where observed data are available for GR4J model calibration, and Köppen climate zone. The total ungauged catchments represent 24.7% of the total drainage division areas. The streamflow from ungauged catchments was estimated using the GR4J model based on the parameters of their donor catchments. Overall, runoff ratios from ungauged catchments were found to be higher compared to their donor-gauged catchments, likely driven by their higher rainfall and less PET. This tendency was particularly evident in two drainage divisions—the Carpentaria Coast (CC) and the Tanami–Timor Sea Coast (TTS)—where ungauged areas comprised 51% and 43%, respectively. The mean gauged annual streamflow varied significantly across drainage divisions—230 gigalitres (GL) from the South Australian Gulf (SAG) to 146,150 GL in TTS. The streamflow from all ungauged catchments was estimated at 232,200 GL per year. Overall, the average streamflow from all drainage divisions, including gauged and ungauged areas, across the coastal regions of Australia was estimated at 419,950 GL per year. This nationwide estimate of streamflow dataset could potentially enhance our understanding of coastal processes and lead to improvements in marine modelling systems and tools.

Performance of Regional Climate Model Precipitation Simulations Over the Terrain‐Complex Andes‐Amazon Transition Region

AbstractRegional climate models (RCMs) are widely used to assess future impacts associated with climate change at regional and local scales. RCMs must represent relevant climate variables in the present‐day climate to be considered fit‐for‐purpose for impact assessment. This condition is particularly difficult to meet over complex regions such as the Andes‐Amazon transition region, where the Andean topography and abundance of tropical rainfall regimes remain a challenge for numerical climate models. In this study, we evaluate the ability of 30 regional climate simulations (6 RCMs driven by 10 global climate models) to reproduce historical (1981–2005) rainfall climatology and temporal variability over the Andes‐Amazon transition region. We assess spatio‐temporal features such as spatial distribution of rainfall, focusing on the orographic effects over the Andes‐Amazon “rainfall hotspots” region, and seasonal and interannual precipitation variability. The Eta RCM exhibits the highest spatial correlation (up to 0.6) and accurately reproduces mean annual precipitation and orographic precipitation patterns across the region, while some other RCMs have good performances at specific locations. Most RCMs simulate a wet bias over the highlands, particularly at the eastern Andean summits, as evidenced by the 100%–2,500% overestimations of precipitation in these regions. Annual cycles are well represented by most RCMs, but peak seasons are exaggerated, especially at equatorial locations. No RCM is particularly skillful in reproducing the interannual variability patterns. Results highlight skills and weaknesses of the different regional climate simulations, and can assist in the selection of regional climate simulations for impact studies in the Andes‐Amazon transition zone.

A Machine Learning Approach to Model Over-Ocean Tropical Cyclone Precipitation

Abstract Extreme rainfall found in tropical cyclones (TCs) is a risk for human life and property in many low- to midlatitude regions. Probabilistic modeling of TC rainfall in risk assessment and forecasting can be computationally expensive, and existing models are largely unable to model key rainfall asymmetries such as rainbands and extratropical transition. Here, a machine learning–based framework is developed to model overwater TC rainfall for the North Atlantic basin. First, a catalog of high-resolution TC precipitation simulations for 26 historical events is assembled for the North Atlantic basin using the Weather Research and Forecasting (WRF) Model. The simulated spatial distribution of rainfall for these historical events are then decomposed via principal component analysis (PCA), and quantile regression forest (QRF) models are trained to predict the conditional distributions of the first five principal component (PC) weights. Conditional distributions of rain-rate levels are estimated separately using historical satellite data and a QRF model. With these models, probabilistic predictions of rainfall maps can be made given a set of storm characteristics and local environmental conditions. The model is able to capture storm total rainfall compared to satellite observations with a correlation coefficient of 0.96 and r2 value of 0.93. Additionally, the model shows good accuracy in modeling hourly total rainfall compared to satellite observations. Rain-rate maps predicted by the model are also compared to historical satellite observations and to the WRF simulations during cross validation, and the spatial distribution of estimates captures rainfall variability consistent with TC rainbands, wavenumber asymmetries, and possibly extratropical transition.

Mapeando a Variabilidade das Chuvas na Bacia do Rio São Francisco: Insights para a Gestão de Recursos Hídricos

Abstract The São Francisco River Basin (SFRB) is extremely important for both the local rural and urban population, running through several states and municipalities in Brazil. Thus, the study of the spatial distribution of rainfall in the SFRB, supported by the application of geostatistics, offers valuable insights for water management and planning. In this study, data from 341 rainfall stations were used, corresponding to the period between 1989 and 2018, and monthly accumulated rainfall analyzed. About the trends, positive estimates for latitude, except for April, highlight that the accumulated precipitation decreases from south to north each month. In December, 85.10% of the spatial variability of rainfall was explained by the trend of rainfall concerning geographical coordinates. Concerning spatial dependence, the analyzed data predominantly exhibited the exponential variogram model. The conducted analyses offered a holistic view of rainfall in the region, facilitating the understanding of spatial patterns and shedding light on challenges in the face of climate change. Our research limited on the spatial modeling of rainfall in the SFRB, and future work should include modeling socio-economic and environmental impacts in the region concerning the spatial irregularity of rainfall.

Mapping Changes in Fractional Vegetation Cover on the Namib Gravel Plains with Satellite-Retrieved Land Surface Emissivity Data

Monitoring changes in vegetation cover over time is crucial for understanding the spatial distribution of rainfall, as well as the dynamics of plants and animals in the Namib desert. Traditional vegetation indices have limitations in capturing changes in vegetation cover within water-limited ecosystems like the Namib gravel plains. Spectral emissivity derived from thermal infrared remote sensing has recently emerged as a promising tool for distinguishing between bare ground and non-green vegetation in arid environments. This study investigates the potential of satellite-derived emissivities for mapping changes in fractional vegetation cover across the Namib gravel plains. Analyzing Moderate Resolution Imaging Spectroradiometer (MODIS) band 29 (λ = 8.55 µm) emissivity time series from 2001 to 2021, our findings demonstrate the ability of both Normalized Difference Vegetation Index (NDVI) and emissivity to detect sudden vegetation growth on the gravel plains. Emissivity additionally allows monitoring the extent of desiccated grass over several years after a rainfall event. Our results support a relationship between the change in fractional vegetation cover, the amount of rainfall and emissivity change magnitude. Information from NDVI and emissivity therefore provide complementary information for assessing vegetation in arid environments.

High-resolution-gridded rainfall dataset derived from surface observation by adjustment of satellite rainfall product

A high-resolution-gridded rainfall dataset is essential for many purposes. Such as analysis of extreme weather conditions, natural-disaster mitigation, or to be used as an input to the hydrological model. Satellite-based rainfall products (e.g., Global Satellite Mapping of Precipitation-GSMaP) can solve the spatial and temporal issues despite their rainfall intensity often being under or overestimated. This research aims to provide a high-resolution rainfall dataset by adjusting the 0.1 deg GSMaP rainfall data to the surface rainfall data from several observation points in the greater Jakarta area (Jabodetabek) during January 2020 when several flooding occurred in Jakarta. The adjustment process includes calculating the bias between the satellite estimation in the nearest observation point and interpolating the error back to the 0.01 deg grid by using radial basis function (RBF) to obtain the correction factor in every grid point, GSMaP data then adjusted by the correction factor. We implemented the method in January 2020 when several floods occurred in Jakarta. The result reveals a more realistic rainfall spatial distribution than regularly interpolating the observation data. The validation of adjusted rainfall estimation at the verification points also shows a reduction in domain-wide RMSE by 30 – 80%.

Análise de agrupamento e tendências de mudanças em séries temporais pluviométricas na bacia hidrográfica do Alto Jacuí, Rio Grande do Sul

Climate studies in watersheds are very important for the water resources planning and management because they indicate climate extremes potentialities and risks. This paper aimed to identify homogeneous rainfall sectors associated with the evaluation of possible changes trends in rainfall series in the Alto Jacuí Watershed (AJW), Rio Grande do Sul State. Were used annual, monthly, and seasonal data from 27 rainfall with a historical series equivalent to 31 years, between the period from 1980 to 2010. The two main methodologies applied in this research corresponded to Cluster Analysis and the Mann-Kendall (MK) test to identify the spatial-temporal rainfall variability. The cluster analysis dendrogram was generated by the Statistica 7.0 software and the MK test and the climate extreme indicators of precipitation were obtained using the R software. Was verified that the orography conditions the rainfall spatial distribution in the AJW. Was identified the most significant trends were identified for the decrease in rainfall during autumn and winter and for the central-southern region of the study area. These results are important to AJW because they can warn to future water scarcity problems for study area.

Impact of Spatial Rainfall Scenarios on River Basin Runoff Simulation a Nan River Basin Study Using the Rainfall-Runoff-Inundation Model

This study aims to investigate the impact of spatial rainfall distribution scenarios from ground observation stations on runoff simulation using hydrological modeling specific to the Rainfall-Runoff-Inundation (RRI) model. The RRI model was applied with six different spatial distribution scenarios of input rainfall, including Inverse Distance Weight (IDW), Thiessen polygon (TSP), Surface Polynomial (SPL), Simple kriging (SKG), and Ordinary kriging (OKG), to simulate the runoff of a 13,000 km2 watershed, namely the Nan River Basin in Thailand. This study utilized data from the 2014 storm event, incorporating temporal information from 28 rainfall stations to estimate rainfall in the spatial distribution scenarios. The six statistics, Volume Bias, Peak Bias, Root Mean Square Error, Correlation, and Mean Bias, were used to determine the accuracy of the estimated rainfall and runoff. Overall, the Simple kriging (SKG) method outperformed the other scenarios based on the statistical values to validate with measured rainfall data. Similarly, SKG demonstrated the closest match between simulated and observed runoff, achieving the highest correlation (0.803), the lowest Root Mean Square Error (164.48 cms), and high Nash-Sutcliffe Efficiency coefficient (0.499) values. This research underscores the practical significance of spatial interpolation methods, such as SKG, in combination with digital elevation models (DEMs) and landuse/soil type datasets, in delivering reliable runoff simulations considering the RRI model on the river basin scale.

Improvement in the skill of CMIP6 decadal hindcasts for extreme rainfall events over the Indian summer monsoon region

Decadal climate predictions have been widely used to predict the near-term climate information relevant for decision-making at multi-year timescales. In the present study, we evaluate the quality of the Coupled Model Intercomparison Project phase-6 (CMIP6) Decadal Climate Prediction Project (DCPP) hindcasts in capturing the extreme rainfall events (EREs) over the monsoon core region during Indian summer monsoon season (June–September) up to lead years 1–10. For the first time, in this study, we have used quantile mapping approach to downscale and bias correct the DCPP CMIP6 simulation/hindcast rainfall for the better representation of EREs. Detailed analysis suggests that the models in general strongly underestimate the rainfall variability over the summer monsoon region. However, after the downscaling and bias correction, the representation of rainfall variability and intensity improved multifold. The bias-corrected decadal hindcasts in fact show ~ 80% improvement in capturing the frequency, intensity, and spatial distribution of rainfall associated with the EREs. Present study brought out a downscaled DCPP product, with potential prediction skill for EREs over India. It is important to highlight that the models predict an increase in the small and medium-area EREs as compared to the large-area EREs over the monsoon core region for the decade 2019–2028.

COMPARISON AND EVALUATION OF MACHINE-LEARNING-BASED SPATIAL DOWNSCALING APPROACHES ON SATELLITE-DERIVED PRECIPITATION DATA

Abstract. Precipitation estimation with high accuracy and resolution is crucial for hydrological and meteorological applications, particularly in ungauged river basins and regions with scarce water resources. Many machine learning (ML) algorithms have been employed in the downscaling of precipitation, however, it remains unclear which algorithm can outperform others. To address this issue, this study evaluates the performance of four ML based downscaling methods to generate high-resolution precipitation estimates at an annual scale. The satellite-derived precipitation data, environmental variables, such as, latitude, longitude, normalized difference vegetation index (NDVI), digital elevation model (DEM), and land surface temperature (LST), as well as the observations from rainfall gauges were used to constructed the regression models. The performance of the four ML algorithms including the Support Vector Regression (SVR), Random Forest (RF), Spatial Random Forest (SRF), and Extreme Gradient Boosting (XGBoost) algorithms was compared with three conventional methods: Multiple Linear Regression (MLR), geographically weighted regression (GWR) and Kriging interpolation model. Results showed that ML-based method generally outperformed traditional interpolation methods in precipitation downscaling, as they had higher accuracy and were better at reproducing the spatial distribution of rainfall. Out of ML approaches, XGBoost received the best performance, followed by SRF, RF and SVR, indicating its robustness of capturing nonlinear relationships. After the XGBoost, better performance of SRF than RF and SVR was found. This might be because the SRF just introduced the spatial autocorrelation into the RF models, which illustrated the importance of capturing spatial variations in ML algorithms. These findings regarding the comparison and assessment provided a novel downscaling method for generating high-resolution precipitation data, which could benefit regional flood forecasting, drought monitoring, and irrigation planning.

The impact of Indian radar and lightning data assimilation on the short‐range forecasts of heavy rainfall events

AbstractThe accurate prediction of high‐impact weather systems using cloud‐resolving models is still a challenge among researchers. This study evaluates the consistency of the combination of the three‐dimensional variational technique within the Gridpoint Statistical Interpolation assimilation system (GSI‐3DVAR) and nudging in the same modelling system on short‐range forecasts of three heavy rainfall events from the southwest monsoon season of 2021 over the Indian subcontinent. Three experiments have been conducted (i) control (CNTL): assimilated conventional and satellite observations; (ii) radar and lightning data assimilation (RLDA): assimilated radar reflectivity and lightning proxy reflectivity data along with all observations used in CNTL; and (iii) lightning data assimilation (LDA): same as RLDA but without the assimilation of radar data; particularly done to test the impact of assimilation of only lightning data. The model‐simulated rainfall is evaluated by using the Integrated Multi‐Satellite Retrievals (IMERG) for Global Precipitation Measurement (GPM IMERG) rainfall data. The intercomparison of LDA and RLDA for event 1 highlighted that both represent the convective regions reasonably better than CNTL, but RLDA outperforms LDA and thus further assimilation experiments are done with RLDA. RLDA provided reasonably accurate forecasts compared to CNTL, which is evident in the spatial distribution of rainfall and area‐averaged three‐hourly accumulated rainfall. Verification metrics for the three selected heavy rainfall events reveal that an optimal forecast performance (especially in the first six hours of free forecast) is obtained by the simulation with assimilation of radar and lightning data during the pre‐forecast period, through correcting the position and timing of convective centres. The probability of detection (POD) values are higher for light rainfall categories than for the heavy rain categories. POD values were higher in RLDA than CNTL throughout simulation for all three events. For all these three selected events, fractions skill scores (FSS) of RLDA are always better than CNTL with different neighbourhood sizes for different threshold values throughout the forecast period.

Assessing the sensitivity of modelled water partitioning to global precipitation datasets in a data‐scarce dryland region

AbstractPrecipitation is the primary driver of hydrological models, and its spatial and temporal variability have a great impact on water partitioning. However, in data‐sparse regions, uncertainty in precipitation estimates is high and the sensitivity of water partitioning to this uncertainty is unknown. This is a particular challenge in drylands (semi‐arid and arid regions) where the water balance is highly sensitive to rainfall, yet there is commonly a lack of in situ rain gauge data. To understand the impact of precipitation uncertainty on the water balance in drylands, here we have performed simulations with a process‐based hydrological model developed to characterize the water balance in arid and semi‐arid regions (DRYP: DRYland water Partitioning model). We performed a series of numerical analyses in the Upper Ewaso Ng'iro basin, Kenya driven by three gridded precipitation datasets with different spatio‐temporal resolutions (IMERG, MSWEP, and ERA5), evaluating simulations against streamflow observations and remotely sensed data products of soil moisture, actual evapotranspiration, and total water storage. We found that despite the great differences in the spatial distribution of rainfall across a climatic gradient within the basin, DRYP shows good performance for representing streamflow (KGE >0.6), soil moisture, actual evapotranspiration, and total water storage (r >0.5). However, the choice of precipitation datasets greatly influences surface (infiltration, runoff, and transmission losses) and subsurface fluxes (groundwater recharge and discharge) across different climatic zones of the Ewaso Ng'iro basin. Within humid areas, evapotranspiration does not show sensitivity to the choice of precipitation dataset, however, in dry lowland areas it becomes more sensitive to precipitation rates as water‐limited conditions develop. The analysis shows that the highest rates of precipitation produce high rates of diffuse recharge in Ewaso uplands and also propagate into runoff, transmission losses and, ultimately focused recharge, with the latter acting as the main mechanism of groundwater recharge in low dry areas. The results from this modelling exercise suggest that care must be taken in selecting forcing precipitation data to drive hydrological modelling efforts, especially in basins that span a climatic gradient. These results also suggest that more effort is required to reduce uncertainty between different precipitation datasets, which will in turn result in more consistent quantification of the water balance.