Search Results (95)

Accurate real-time information about soil moisture (SM) at a large scale is essential for improving hydrological modeling, managing water resources, and monitoring extreme weather events. This study presents a framework using convolutional long short-term memory (ConvLSTM) network to produce short- (1, 3, and 7 days ahead) and mid-term (14 and 30 days ahead) forecasts of SM at surface (0–10 cm) and subsurface (10–40 and 40–100 cm) soil layers across the contiguous U.S. The model was trained with five-year period (2018–2022) datasets including Soil Moisture Active Passive (SMAP) level 3 ancillary covariables, North American Land Data Assimilation System phase 2 (NLDAS-2) SM product, shortwave infrared reflectance from Moderate Resolution Imaging Spectroradiometer (MODIS), and terrain features (e.g., elevation, slope, curvature), as well as soil texture and bulk density maps from the Soil Landscape of the United States (SOLUS100) database. To develop and evaluate the model, the dataset was divided into three subsets: training (January 2018–January 2021), validation (2021), and testing (2022). The outputs were validated with observed in situ data from the Soil Climate Analysis Network (SCAN) and the United States Climate Reference Network (USCRN) soil moisture networks. The results indicated that the accuracy of SM forecasts decreased with increasing lead time, particularly in the surface (0–10 cm) and subsurface (10–40 cm) layers, where strong fluctuations driven by rainfall variability and evapotranspiration fluxes introduced greater uncertainty. Across all soil layers and lead times, the model achieved a median unbiased root mean square error (ubRMSE) of 0.04 cm³ cm⁻³ with a Pearson correlation coefficient of 0.61. Further, the performance of the model was evaluated with respect to both land cover and soil texture databases. Forecast accuracy was highest in coarse-textured soils, followed by medium- and fine-textured soils, likely because the greater penetration depth of microwave observations improves SM retrieval in sandy soils. Among land cover types, performance was strongest in grasslands and savannas and weakest in dense forests and shrublands, where dense vegetation attenuates the microwave signal and reduces SM estimation accuracy. These results demonstrate that the ConvLSTM framework provides skillful short- and mid-term forecasts of surface and subsurface soil moisture, offering valuable support for large-scale drought and flood monitoring. Full article

Probable Maximum Storms (PMS) are synthetic design storms represented by idealized hyetographs. They play a critical role in assessing extreme rainfall events over extended durations and are widely applied in the hydraulic design of infrastructure such as dams, culverts, and bridges. PMS provide essential input for estimating Probable Maximum Floods (PMF), vital for analyzing worst-case flood scenarios with the potential to cause catastrophic loss of life and property. Despite their importance, the estimation of design storms at ungauged locations, particularly across synoptic scales, remains a major scientific and engineering challenge. This study addresses this gap by utilizing the Integrated Multi-satellitE Retrievals for Global Precipitation Measurement (IMERG) dataset, which provides near-global estimated precipitation coverage. IMERG’s 24 h design storm hyetographs (expressed as cumulative percentage of precipitation throughout a 24 h period) were modeled and compared with similar reference data from NOAA Atlas 14 across twenty-eight regions and seven larger zones covering most of the conterminous United States (CONUS). Across the regions, the average root mean square error (RMSE) was 3.7%, with a mean relative bias (RB) of 1.4%. The mean normalized storm loading index (NSLI) from NOAA Atlas 14 was −7.7%, indicating that 57.7% of the total precipitation was received during the first 12 h of the storm, whereas IMERG storms exhibited a mean NSLI of −4.1%, suggesting they are also frontloaded but to a lesser extent. Across the broader zones, the mean RMSE was 4.8% and the mean RB was 1.1%. The mean NSLI values were −9.7% for NOAA Atlas 14 and −5.7% for IMERG, again indicating that IMERG storms are less frontloaded. When design storm families were estimated corresponding with different degrees of frontloading (corresponding to the 10, 20, …, 90% deciles of NSLI), the 40th to 60th percentile range exhibited the strongest agreement between IMERG and NOAA Atlas 14 hyetographs. Full article

Landslides are among the most destructive natural hazards in the Western Ghats region of Kerala, driven by complex interactions between geological, hydrological, and anthropogenic factors. This study aims to generate a high-resolution Landslide Susceptibility Level Map (LSLM) using a machine learning (ML)-based Random Forest (RF) model integrated with Geographic Information Systems (GIS). A total of 231 historical landslide locations obtained from the Bhukosh portal were used as reference data. Eight predictive factors—Stream Order, Drainage Density, Slope, Aspect, Geology, Land Use/Land Cover (LULC), Normalized Difference Vegetation Index (NDVI), and Moisture Stress Index (MSI)—were derived from remote sensing and ancillary datasets, preprocessed, and reclassified for model input. The RF model was trained and validated using a 50:50 split of landslide and non-landslide points, with variable importance values derived to weight each predictive factor of the raster layer in ArcGIS. The resulting Landslide Susceptibility Index (LSI) was reclassified into five susceptibility zones: Very Low, Low, Moderate, High, and Very High. Results indicate that approximately 17.82% of the study area falls under high to very high susceptibility, predominantly in the steep, weathered, and high rainfall zones of the Western Ghats. Validation using Area Under the Curve–Receiver Operating Characteristic (AUC-ROC) analysis yielded an accuracy of 0.890, demonstrating excellent model performance. The output LSM provides valuable spatial insights for planners, disaster managers, and policymakers, enabling targeted mitigation strategies and sustainable land-use planning in landslide-prone regions. Full article

This study aims to explore the methodology for assessing landslide susceptibility by using machine learning techniques based on a geographic information system (GIS) in an effort to develop landslide susceptibility maps and assess landslide risk in the Yiyang region. A landslide dataset in Yiyang was constructed after 16 landslide predisposing factors were identified across four categories, topography, geology, environment, and hydrometeorology, through factor state determination and multicollinearity analysis. A Blending ensemble model was created and achieved higher prediction accuracy by fusing predictions from Random Forest, CatBoost, and XGBoost with logistic regression used as the meta-learner, thus deriving the importance coefficients of the landslide predisposing factors and their contribution rates. The Blending ensemble model achieved high predictive accuracy with an AUC value of 0.8784, demonstrating balanced and stable performance characteristics. With the addition of the rainfall factor, the AUC value of the Blending ensemble model has increased by 0.1199. In combination with the information value method, this model was applied to assess landslide susceptibility and rainfall-induced landslide risks in Yiyang City, demonstrating its validity. In addition, experimental validation confirmed the prediction and evaluation accuracy of the GIS-based Blending ensemble model. Results showed that the frequency ratio (FR) of historical landslide occurrences in high-susceptibility and extremely high-susceptibility zones in Yiyang City exceeded 1, indicating strong consistency between the landslide risk classification and actual distribution of historical landslides. The landslide susceptibility maps created for Anhua County, Heshan District, and Taojiang County in Yiyang City may provide support for the early warning and prevention of landslides and land-use planning in this region. The proposed methodology may be of reference value for improving natural disaster prevention and risk management. Full article

In the recent decades, an ongoing increase in maximum temperature during summer has been observed in Poland, especially in the central-southern and southeastern areas. This raises the vulnerability of these regions not only to heat waves and drought but also to floods. The potential effect of compound heat waves and extreme rainfall events may be more serious than the effects of these events occurring separately. This research is the first attempt in Poland to investigate whether the presence of a heat wave increases the likelihood of extreme rainfall events, if so, by how much, and whether this changes with warming. For this purpose, we used daily maximum temperature values and 6 h precipitation datasets from 44 meteorological stations in Poland for the 1966–2024 period. It was proven that compound heat wave and extreme rainfall events occurred in Poland with spatially differentiated frequency. They occurred the least frequently on the coast and the most frequently in southwestern, southeastern, and northeastern Poland. The extreme rainfall occurred most often between noon and midnight on the last heat wave day. During these hours, the likelihood of extreme rainfall is, on average, 3.5 times higher than that expected according to climatology norms. With warming, the frequency of days with these compound events increases at the rate of 1.22 days per decade, and the frequency of compound events increases at a rate of 3.75 events per decade. Although a detailed analysis of the mechanisms responsible for such events is planned for further research, the preliminary study revealed that in most cases, the approach of a cold front with a mesoscale thundercloud system was responsible for heat wave termination with extreme rainfall. Since we cannot prevent the growing number of heat waves or heavy precipitation events that terminate the heat wave events in Poland, the adaptation strategy needs to be implemented to meet the sustainable development goals regarding climate actions. This refers primarily to urban planning, agriculture (agroecosystems), social health, and well-being. Full article

This study utilizes a low-cost Global Navigation Satellite System (GNSS) buoy platform, combined with multi-system GNSS data, to investigate the impact of GNSS signal quality and multipath effects on the accuracy of atmospheric precipitable water vapor (PWV) retrievals. It also explores the methods for oceanic rainfall event forecasting and precipitation prediction based on GNSS-PWV. By analyzing the data quality from various GNSS systems and using the European Centre for Medium-Range Weather Forecasts (ECMWF) ERA5 dataset as a reference, the study assesses the accuracy of PWV retrievals in dynamic marine environments. The results show that the GNSS-derived PWV from the buoy platform is highly consistent with ERA5 data in both trend and characteristics, with an RMSE of 3.8 mm for the difference between GNSS-derived PWV and ERA5 PWV. To enhance rainfall forecasting accuracy, a balanced threshold selection (BTS) method is proposed, significantly improving the balance between the probability of detection (POD) and false alarm rate (FAR). Furthermore, a Random Forest model based on multiple meteorological parameters optimizes precipitation forecasting, especially in reducing false alarms. Additionally, a particle swarm optimization (PSO)-based BP Neural Network model for rainfall prediction achieves high precision, with an R² of 97.8%, an average absolute error of 0.08 mm, and an RMSE of 0.1 mm. The findings demonstrate the potential of low-cost GNSS buoy for monitoring atmospheric water vapor and short-term rainfall forecasting in dynamic marine environments. Full article

Accurate flood monitoring and forecasting techniques are important and continue to be developed for improved disaster preparedness and mitigation. Flood estimation using satellite observations with deep learning algorithms is effective in detecting flood patterns and environmental relationships that may be overlooked by conventional methods. Soil Moisture Active Passive (SMAP) fractional water (FW) was used as a reference to estimate flood areas in a long short-term memory (LSTM) model using a combination of soil moisture information, rainfall forecasts, and floodplain topography. To perform flood modeling in LSTM, datasets with different spatial resolutions were resampled to 30 m spatial resolution using bicubic interpolation. The model’s efficacy was quantified by validating the LSTM-based flood inundation area with a water mask from Senti-nel-1 SAR images for regions with different topographic characteristics. The average area under the curve (AUC) value of the LSTM model was 0.93, indicating a high accuracy estimation of FW. The confusion matrix-derived metrics were used to validate the flood inundation area and had a high-performance accuracy of ~0.9. SMAP FW showed optimal performance in low-covered vegetation, seasonal water variations and flat regions. The estimates of flood inundation areas show the methodological promise of the proposed framework for improved disaster preparedness and resilience. Full article

Tailings ponds can recycle water resources through the water recirculation system by clarifying and purifying the wastewater discharged from the mining production process. Due to factors such as flooding and heavy rainfall, once a tailings dams burst, the spread of heavy metals in the tailings causes underground and surface water pollution, endangering the lives and properties of people downstream. To effectively assess the potential impact of tailings dams bursting, many problems such as the difficulty of taking values in predicting the volume of silt penetration through empirical formulae, model testing, and numerical simulation need to be solved. In this study, 65 engineering cases were collected to develop a sample dataset containing dam height and storage capacity. The Support Vector Machine (SVM) algorithm was used to develop a nonlinear regression model for tailings discharge volume after tailings dam failure. In addition, the model penalty parameter C and kernel function g were optimized using the powerful global search capability of the Slime Mold Algorithm (SMA) to develop an SMA–SVM prediction model for tailings discharge volume. The results indicate that the volume of tailings discharged increases nonlinearly with increasing dam height and tailings storage capacity. The SMA-SVM model showed higher prediction accuracy compared to the predictions made by the Random Forest (RF), Radial Basis Function (RBF), and Least Squares SVM (LS-SVM) algorithms. The average absolute error in tailings discharge volume compared to actual values was 30,000 m³, with an average relative error of less than 25%. This is very close to practical engineering scenarios. The ability of the SMA-SVM optimization algorithm to produce predictions with minimal error relative to actual values was further confirmed by the combination of numerical simulations. In addition, the numerical simulations revealed the flow characteristics and inundation area of the discharged sediment during tailings dam failure, and the research results can provide reference for water resource protection and downstream safety prevention and control of tailings ponds. Full article

This study evaluates the sensitivity of the Weather Research and Forecasting (WRF-ARW) model in its version 4.3.3 during different experiments on a torrential precipitation event associated with the 2017 El Niño Costero in Peru. The results are compared with two reference datasets: precipitation estimations from CHIRPS satellite data and SENAMHI meteorological station values. The event, which had significant economic and social impacts, is simulated using two nested domains with resolutions of 9 km (d01) and 3 km (d02). A total of 22 experiments are conducted, resulting from the combination of two planetary boundary layer (PBL) schemes: Yonsei University (YSU) and Mellor–Yamada–Janjic (MYJ), with five cumulus parameterization schemes: Betts–Miller–Janjic (BMJ), Grell–Devenyi (GD), Grell–Freitas (GF), Kain–Fritsch (KF), and New Tiedtke (NT). Additionally, the effect of turning off cumulus parameterization in the inner domain (d02) or in both (d01 and d02) is explored. The results show that the YSU scheme generally provides better results than the MYJ scheme in detecting the precipitation patterns observed during the event. Furthermore, it is concluded that turning off cumulus parameterization in both domains produces satisfactory results for certain regions when it is combined with the YSU PBL scheme. However, the KF cumulus parameterization is considered the most effective for intense precipitation events in this region, although it tends to overestimate precipitation in high mountain areas. In contrast, for lighter rains, combinations of the YSU PBL scheme with the GD or NT parameterization show a superior performance. It is worth nothing that for all experiments here used, there is a clear underestimation in terms of precipitation, except in high mountain regions, where the model tends to overestimate rainfall. Full article

This work investigates the impact of rainfall on cellular communication links, leveraging smartphone-collected measurements. While existing studies primarily focus on line-of-sight (LoS) microwave propagation environments, this work explores the impact of rainfall on typical signal metrics over cellular links when the LoS path is not guaranteed. We examine both small-scale and large-scale variations in signal measurements across dry and rainy days, considering diverse locations and time windows. Through statistical and spectral analysis of a large dataset, we uncover novel insights into how rainfall influences cellular communication links. Specifically, we observe a consistent daily fluctuation pattern in key cellular metrics, such as the reference signal received quality. Additionally, spectral features of key mobile metrics show noticeable changes during rainfall events. These findings, consistent across three distinct locations, highlight the significant impact of rainfall on everyday cellular links. They also suggest that the widely available by-product signals from mobile phones could be leveraged for innovative rainfall-sensing applications. Full article

Intense convection is often accompanied by high-frequency lightning and is highly prone to producing heavy rainfall, strong winds, hail, and tornadoes, frequently resulting in significant damage and loss of life. It is necessary to understand the mechanisms and meteorological conditions of intense convection. This study utilizes the Thunderstorm Feature Dataset from 2010–2018 to analyze the characteristics of thunderstorms with extreme lightning activity (TELAs), defined as thunderstorms whose lightning frequency ranks in the top 1%. Four regions with relatively high thunderstorm activity were selected for analysis: Northeast China (NEC), North China (NC), South China (SC), and the Tibetan Plateau (TP). In NEC, TELAs primarily occur just west of upper-level westerly troughs (UWT), including cold vortices. In NC, TELAs are mainly associated with UWT and subtropical highs (STH). In SC, TELAs are related to frontal systems, easterly waves, tropical cyclones, and STH. In TP, TELAs are generated by TP vortices. Before the TELA process, vertically integrated moisture divergence (VIMD) and convective available potential energy (CAPE) show the most notable anomalies. Except for the TP, TELAs are typically located between centers of anomalies with positive and negative geopotential height (500 hPa) and near centers of anomalies with positive CAPE and negative VIMD, accompanied by notable increases in surface temperature and wind speed. These findings offer a valuable reference for the early warning and forecasting of intense convection. Full article

Supercooled liquid water (SLW) refers to droplets in clouds that remain unfrozen at temperatures below 0 °C. SLW is an important intermediate hydrometeor in the processes of snowfall and rainfall that can modulate the radiation budget. This study investigates the distribution of supercooled cloud water over mainland China using the East Asia–Pacific cloud macro- and microphysical properties dataset (2016), derived from Himawari-8 observations. The results show that the highest frequency of SLW in liquid-phase stratus clouds occur at the eastern slope of the Tibetan Plateau, the western side of the Sichuan Basin. Additional SLW is mostly found in liquid-phase clouds over the Sichuan Basin and its adjacent areas in southern China. In the region with the highest frequency of SLW, the mechanical forcing of the Tibetan Plateau causes the convergence of low-level airflow within the basin, which also carries moisture that is forced to ascend stably, creating a favorable condition for the formation of supercooled clouds. As the airflow continues to ascend, it encounters the mid-to-upper-level westerlies and temperature inversion. At the mid-to-upper level, the westerlies exhibit stronger wind speeds, directing flow towards the basin. Concurrently, the temperature inversion stabilizes the atmospheric stratification, limiting the further ascent of airflow. This inversion can also restrain convection and upward motion within the clouds, allowing for SLW to exist and persist for an extended period. Full article

In recent years, extreme climate events have shown to be occurring more frequently. As a highly populated area in central China, the Jialing River Basin (JRB) should be more deeply explored for its patterns and associations with climatic factors. In this study, based on the daily precipitation and atmospheric temperature datasets from 29 meteorological stations in JRB and its vicinity from 1960 to 2020, 10 extreme indices (6 extreme precipitation indices and 4 extreme temperature indices) were calculated. The spatial and temporal variations of extreme precipitation and atmospheric temperature were analyzed using Mann–Kendall analysis, to explore the correlation between the atmospheric circulation patterns and extreme indices from linear and nonlinear perspectives via Pearson correlation analysis and wavelet coherence analysis (WTC), respectively. Results revealed that among the six selected extreme precipitation indices, the Continuous Dry Days (CDD) and Continuous Wetness Days (CWD) showed a decreasing trend, and the extreme precipitation tended to be shorter in calendar time, while the other four extreme precipitation indices showed an increasing trend, and the intensity of precipitation and rainfall in the JRB were frequent. As for the four extreme temperature indices, except for TN10p, which showed a significant decreasing trend, the other three indices showed a significant increasing trend, and the number of low-temperature days in JRB decreased significantly, the duration of high temperature increased, and the basin was warming continuously. Spatially, the spatial variation of extreme precipitation indices is more obvious, with decreasing stations mostly located in the western and northern regions, and increasing stations mostly located in the southern and northeastern regions, which makes the precipitation more regionalized. Linearly, most of the stations in the extreme atmospheric temperature index, except TN10p, show an increasing trend and the significance is more obvious. Except for the Southern Oscillation Index (SOI), other atmospheric circulation patterns have linear correlations with the extreme indices, and the Arctic Oscillation (AO) has the strongest significance with the CDD. Nonlinearly, NINO3.4, Pacific Decadal Oscillation (PDO), and SOI are not the main circulation patterns dominating the changes of TN90p, and average daily precipitation intensity (SDII), maximum daily precipitation amount (RX1day), and maximum precipitation in 5 days (Rx5day) were most clearly associated with atmospheric circulation patterns. This also confirms that atmospheric circulation patterns and climate tend not to have a single linear relationship, but are governed by more complex response mechanisms. This study aims to help the relevant decision-making authorities to cope with the more frequent extreme climate events in JRB, and also provides a reference for predicting flood, drought and waterlogging risks. Full article

Landslide hazard mapping is important to urban construction and landslide risk management. Dynamic landslide hazard mapping considers landslide deformation with changes in the environment. It can show more details of the landslide process state. Landslides in highly vegetated areas are difficult to observe directly, which makes landslide hazard mapping much more challenging. The application of multi-InSAR opens new ideas for dynamic landslide hazard mapping. Specifically, landslide susceptibility mapping reflects the spatial probability of landslides. For rainfall-induced landslides, the scale exceedance probability reflects the temporal probability. Based on the coupling of them, dynamic landslide hazard mapping further considers the landslide deformation intensity at different times. Zigui, a highly vegetation-covered area, was taken as the study area. The landslide displacement monitoring effect of different band SAR datasets (ALOS-2, Sentinel-1A) and different interpretation methods (D-InSAR, PS-InSAR, SBAS-InSAR) were studied to explore a combined application method. The deformation interpreted by SBAS-InSAR was taken as the main part, PS-InSAR data were used in towns and villages, and D-InSAR was used for the rest. Based on the preliminary evaluation and the displacement interpreted by fusion InSAR, the dynamic landslide hazard mappings of the study area from 2019 to 2021 were finished. Compared with the preliminary evaluation, the dynamic mapping approach was more focused and accurate in predicting the deformation of landslides. The false positives in very-high-hazard zones were reduced by 97.8%, 60.4%, and 89.3%. Dynamic landslide hazard mapping can summarize the development of and change in landslides very well, especially in highly vegetated areas. Additionally, it can provide trend prediction for landslide early warning and provide a reference for landslide risk management. Full article

The growing concerns about floods have highlighted the need for accurate and detailed precipitation data as extreme precipitation occurrences can lead to catastrophic floods, resulting in significant economic losses and casualties. Integrated Multi-satellitE Retrievals for the Global Precipitation Measurement (GPM IMERG) is a commonly used high-resolution gridded precipitation dataset and is recognized as trustworthy alternative sources of precipitation data. The aim of this study is to comprehensively evaluate the performance of GPM IMERG Early (IMERG-E), Late (IMERG-L), and Final Run (IMERG-F) in precipitation estimation and their capability in detecting extreme rainfall indices over southwestern Iran during 2001–2020. The Asfezari gridded precipitation data, which are developed using a dense of ground-based observation, were utilized as the reference dataset. The findings indicate that IMERG-F performs reasonably well in capturing many extreme precipitation events (defined by various indices). All three products showed a better performance in capturing fixed and non-threshold precipitation indices across the study region. The findings also revealed that both IMERG-E and IMERG-L have problems in rainfall estimation over elevated areas showing values of overestimations. Examining the effect of land cover type on the accuracy of the precipitation products suggests that both IMERG-E and IMERG-L show large and highly unrealistic overestimations over inland water bodies and permanent wetlands. The results of the current study highlight the potential of IMERG-F as a valuable source of data for precipitation monitoring in the region. Full article