Rain Patterns Research Articles

Wetland vegetation zoning as a response to groundwater complex systems.

This research investigates the interplay between groundwater flow systems and the zoning of wetland species. We aimed to elucidate the relationship between these factors through comprehensive field evaluations encompassing plant composition, piezometric levels, and flow direction; groundwater chemistry, vertical and horizontal, at different depths (0.15m, 2m, 4m, and 6m) during both rainy and dry seasons. Our findings reveal distinct patterns. The plant diversity in the FP site was related to low ion concentration. In contrast, the dominance of T. geniculata in the TH site correlated with species adapted to sulfate and chloride (SO42- and Cl-) tolerance. In the CTC site, the dominance of C. jamaicense was linked to local groundwater flow and significant surface water inflows. At the same time, T. domingensis prevailed in the TC site due to sodium chloride facie inhibiting C. jamaicense seed germination. Our study highlights the role of groundwater chemical composition in shaping wetland ecosystems by influencing species tolerance and how the spatial distribution of groundwater discharge and seasonal changes contribute to distinct vegetation patches within wetlands. The relationship between groundwater systems, the water's chemical composition, and vegetation is strongly dependent. This can explain mortality and changes in vegetation composition in wetlands in different parts of the world, stressed by climate change effects (rain pattern alteration).

Fog and rain augmentation for license plate recognition in tropical country environment

Automatic license plate recognition (ALPR) is a critical component in modern traffic management systems. However, ALPR systems often face challenges in accurately recognizing license plates under adverse weather conditions, such as fog and rain, prevalent in tropical regions. Deep learning ALPR models necessitate huge and diverse datasets for robustness, but data availability remains a concern since unpredictable fog and rain patterns hinder data collection. In this study, we address the issue of enhancing ALPR's robustness by introducing a novel augmentation strategy that combines traditional and weather augmentation techniques. By augmenting the dataset with weather-induced variations, we aim to improve the generalization capability of ALPR models, enabling them to handle a wider range of weather-related challenges. We also investigate the synergy between these weather augmentations and established scene text recognition (STR) methods, such as convolutional recurrent neural network (CRNN), TPS-ResNet BiLSTM-attention (TRBA), autonomous bidirectional iterative scene text recognition (ABINet), vision transformer (ViTSTR), and permutated autoregressive sequence (PARSeq), to determine their impact on recognition accuracy. Experiments using different training data sets show that training data containing a combination of traditional and weather augmentation produces the best accuracy and 1-NED performance compared to training data without augmentation and traditional augmentation only. The average increase accuracy of all STR model is 1.13% with the best increase accuracy of 3.68% using TRBA.

Characterization of dam failure flooding in an urban reservoir under different rainfall conditions

In water conservation projects, reservoirs play a key role in controlling and regulating floods. In recent years, extreme rainfall events have occurred more frequently. Urban reservoirs in danger under extreme rainfall happen occasionally. The breaching of urban reservoirs seriously threatens the lives and property of the population downstream. A lack of clarity regarding the characterization of the flooding response remains owing to urban reservoir overtopping under different rainfall conditions. This study used observed data from the heavy rainfall event in Zhengzhou on July 20, 2021, with Jiangang Reservoir in Zhengzhou as the study site. The study designed a storm-flood process with different rainfall scenarios that combines the long-duration rainstorm formula and Chicago rain pattern method, establishing a response relationship between extreme rainfall and flooding processes. Flood evolution simulation analyses were carried out for different scenarios of rainfall and flooding processes under three working conditions: rainfall only, dam failure only and dam failure coupled with rainfall, to characterize the flood response due to dam failure under different rainfall conditions. The results showed that the peak of the rainfall occurs earlier in the total rainfall calendar time, the faster the dam failure rate of the reservoir. As the rain peak occurs at a time that is later in the total rainfall calendar, and the recurrence period and rainfall duration increase, the flood volume increases and urban inundation becomes more severe. The rainfall type had less of an effect on the difference in the extent of inundation between different working conditions. The extent of flood inundation for the dam failure-coupled rainfall scenario is much greater than that for the dam failure only scenario. The increase in inundation extent in the dam failure coupled rainfall scenario at the higher levels of flood severity is smaller. Study results provide hydraulic element support for flood hazard assessment.

Quantifying seed rain patterns in a remnant and a chronosequence of restored tallgrass prairies in north central Missouri

Abstract Seed rain is an influential process related to plant community diversity, composition and regeneration. However, knowledge of seed rain patterns is limited to those observed in forests and late‐assembling grasslands, which might not reflect early assembling communities such as newly restored grasslands. Resolving this gap in our understanding provides further insight into the role of seed dispersal. Here, we measured seed rain in a remnant tallgrass prairie, which was the site of the foundational grassland seed rain study in 1978, and a nearby chronosequence of tallgrass prairie restorations. We sought to determine how the quantity, seed mass traits, timing, diversity and composition of seed rain changed (1) long‐term and (2) during community assembly. To do so, we deployed artificial turf grass seed traps into 2‐year‐old, 5–6‐year‐old and 15‐year‐old restored prairies and the remnant prairie, replacing traps every 2 weeks from May to December 2019. We captured over twice the density and richness of seed rain in the remnant prairie in 2019 compared to 1978. We also found that seed rain patterns changed as prairies aged, with each prairie possessing a distinct community of dispersing species. Significantly more seeds, seed biomass, and species were captured in the youngest restored prairie. However, seed mass traits were similar in all prairies. Except for composition, all other seed rain metrics in the oldest restoration were eventually comparable to the remnant prairie. Synthesis and applications: Our results revealed that grasslands, notably young prairies, produce larger quantities of seed rain than previously known (124,806 seeds m−2 year−1, 97.24 g m−2 year−1), and seed input in all sampled prairies far exceeded restoration broadcast seeding densities. We further found that decreases in seed rain quantity across the chronosequence did not correspond with increases in seed mass, suggesting a lack of trade‐offs between these metrics. Furthermore, tallgrass prairie restorations have not replicated the composition of seed rain seen in remnant systems. Increasing restoration seeding rates of desirable species may be needed to meet composition goals since current rates may not compete with the propagule pressure of undesirable species found in newly restored prairies.

Enhancing the Rainfall Forecasting Accuracy of Ensemble Numerical Prediction Systems via Convolutional Neural Networks

Abstract Ensemble prediction systems are commonly used to demonstrate the potential uncertainty of weather forecasts. These systems also help provide weather predictions to prepare for disasters. Specifically, a type of prediction called a quantitative precipitation forecast (QPF) is calculated from the average of multiple forecasts. The probability-matched mean (PMM) method is used to improve these QPF predictions when they underestimate rainfall. However, the PMM method can exhibit limitations when dealing with certain types of rain patterns. To address this issue, this study focuses on improving short-term heavy rainfall predictions using an artificial intelligence (AI) algorithm that combines deep neural networks (DNNs), including convolutional neural networks (CNNs), with a space-based attention mechanism [convolutional block attention module (CBAM)]. Four experiments were conducted to evaluate the new method, including those performed to optimize the timing of a 24-h QPF model, adjust the training dataset, and refine the training algorithm. Two specific weather events were assessed: rainfall influenced by southwest winds after typhoons and afternoon thunderstorm rainfall. A comparison of the results obtained via the root-mean-square error (RMSE) and structural similarity index measure (SSIM) reveals significant improvements in both accuracy and distribution of the QPF predictions over those of the PMM method. Compared with the PMM method, our approach can reduce the RMSE from 77.51 to 19.52 in the mei-yu case, an improvement of approximately 74.82%. The SSIM also increased from 0.33 to 0.56, indicating a 70.9% enhancement. Overall, the new approach successfully enhances rainfall prediction accuracy via AI techniques and has the potential to be applied in disaster preparedness operations. Significance Statement An ensemble prediction system is a tool that is often used to showcase the uncertainty inherent in numerical weather forecasts and to provide critical insights for disaster preparedness. This study attempts to enhance the short-term predictions produced for intense rain events by integrating an AI algorithm, which exploits the advantages of deep neural networks (DNNs), such as convolutional neural networks (CNNs) with a space-based attention mechanism. The methodology underwent rigorous testing, including lead time optimization, training data sensitivity assessment, and algorithm refinement. When applied to weather events, such as rainfall patterns influenced by posttyphoon southwestern winds and afternoon thunderstorms, this approach outperforms traditional methods, as evidenced by verification metrics, such as the RMSE and SSIM. This research introduces an innovative framework that leverages artificial intelligence (AI) and DNNs to achieve enhanced rainfall prediction accuracy, which is a critical advancement for effective disaster readiness.

A multi-frame fusion video deraining neural network based on depth and luminance features

Rainy weather is a common natural occurrence, but capturing clear and accurate images in rainy conditions can be challenging. Consequently, addressing the effects of rain on video imagery is essential for enhancing visual quality and the accuracy and reliability of computer vision applications. In this work, we introduce a novel rain model accompanied by a multi-frame fusion video rain removal algorithm that exploits depth and luminance features. We present an innovative three-stage video deraining methodology that synergizes multi-frame fusion with advanced neural network design to achieve highly realistic and clear rain removal. Our Option-Flow-Depth-Luminance (OFDL) Derain algorithm propels the research forward by offering a new perspective on rain removal. Incorporating depth and luminance features into image processing, we devise a novel rain model that allows the algorithm to handle complex environments and a variety of rain patterns effectively. Additionally, our model takes into account degradation factors such as rain streaks, accumulation, runoff, and occlusion, leading to the generation of more authentic rain-affected images that improve the training and performance evaluation of the model. Experimental tests conducted on the RainSynLight25, RainSynComplex25, and NTURain datasets demonstrate that our method outperforms current state-of-the-art techniques, achieving increases of 4.3 dB, 2.2 dB, and 2.2 dB in PSNR, and enhancements of 0.062, 0.109, and 0.008 in Structural Similarity Index Measure (SSIM), respectively. Moreover, our approach exhibits superior processing speed, further underscoring its practical advantages.

GIS-based method for assessing the viability of solar-powered irrigation

To attain food security, sub-Saharan Africa needs to increase its use of irrigation to improve agricultural productivity and resilience to climate change-induced droughts and erratic rain patterns. Stand-alone solar-powered irrigation systems are a promising technology to power irrigation where grid electrification is unavailable. However, these systems’ economic viability and sustainability are affected by spatially varying factors such as climate, water availability, and solar potential, making it challenging to plan their roll-out. This study addresses the planning challenge by developing a novel open-source reproducible geospatial methodology to determine irrigation water demand, peak power demand, life cycle cost, and locations where solar-powered irrigation systems would be cheaper than alternative off-grid irrigation technologies. The method considers spatial–temporal data for meteorological conditions, soil quality, water distance and depth, solar resource potential, technology efficiency and local equipment and fuel costs. Maize cultivation in Kenya is taken as a case study to demonstrate the method. We find the median peak irrigation water demand in Kenya to be 65,000 l/ha/day and the median peak power demand to be 2.6 kW/ha. The levelized cost of solar-powered irrigation systems ranges from US$ 0.09/kWh to US$ 0.25/kWh, making solar irrigation cheaper than diesel-powered irrigation in all Kenyan regions. The results from the method offer valuable insights to governments, farmers and investors on where water use regulations are required for sustainable water use, where subsidies and innovative financial models are needed for the affordability of solar irrigation and where alternative uses of energy from the solar systems beyond irrigation are possible.

High-Resolution Rainfall Estimation Using Ensemble Learning Techniques and Multisensor Data Integration.

In Indonesia, the monitoring of rainfall requires an estimation system with a high resolution and wide spatial coverage because of the complexities of the rainfall patterns. This study built a rainfall estimation model for Indonesia through the integration of data from various instruments, namely, rain gauges, weather radars, and weather satellites. An ensemble learning technique, specifically, extreme gradient boosting (XGBoost), was applied to overcome the sparse data due to the limited number of rain gauge points, limited weather radar coverage, and imbalanced rain data. The model includes bias correction of the satellite data to increase the estimation accuracy. In addition, the data from several weather radars installed in Indonesia were also combined. This research handled rainfall estimates in various rain patterns in Indonesia, such as seasonal, equatorial, and local patterns, with a high temporal resolution, close to real time. The validation was carried out at six points, namely, Bandar Lampung, Banjarmasin, Pontianak, Deli Serdang, Gorontalo, and Biak. The research results show good estimation accuracy, with respective values of 0.89, 0.91, 0.89, 0.9, 0.92, and 0.9, and root mean square error (RMSE) values of 2.75 mm/h, 2.57 mm/h, 3.08 mm/h, 2.64 mm/h, 1.85 mm/h, and 2.48 mm/h. Our research highlights the potential of this model to accurately capture diverse rainfall patterns in Indonesia at high spatial and temporal scales.

Simulation and Risk Assessment of Flood Disaster at the Entrance to a Rail Transit Station under Extreme Weather Conditions—A Case Study of Wanqingsha Station of Guangzhou Line 18

With the rapid development of urbanization and underground transportation, as well as the frequent occurrence of extreme weather conditions such as extreme rainfall, flooding disasters for rail transit are becoming severe, and need to be urgently clarified in terms of the mechanism causing them. In this study, a comprehensive model for water damage at the entrance to a rail transit station is proposed, emphasizing the entire process of extreme weather–surface ponding–underground intrusion. The model is validated by the inundation process of Line 5 of the Zhengzhou Metro during the “7.20” event and further applied to Wanqingsha Station of Guangzhou Metro Line 18 in China to determine the surrounding water depth, distribution, total water inflow volume, and water damage time under different rainfall intensities, rain patterns and protection scenarios. It was found that when rainfall reaches the level of a 1-in-2000-years event, the surface water begins to invade the internal rail transit system through the rail transit entrances. When facing extreme rainfall akin to the “7.20” event in Zhengzhou, the rail transit system in Wanqingsha Station meets a heightened risk of water damage, resulting in significantly deeper water levels compared to 1-in-5000-year rainfall event in Guangzhou and exceeds the height of the subway entrances. Analysis of the water intrusion process reveals that, as rainfall intensity escalates, the total inflow water volume into the rail transit system increases while escape time diminishes. Moreover, under identical rainfall intensity, pre-type rainfall yields the highest total water inflow, whereas mid-type rainfall exhibits the shortest escape time. Enhancing the protection conditions can markedly attenuate surface water intrusion into the subterranean rail transit system, thereby enhancing the evacuation time for individuals within the system.

Image Deraining Algorithm Based on Multi-Scale Features

In target detection, tracking, and recognition tasks, high-quality images can achieve better results. However, in actual scenarios, the visual effects and data quality of images are greatly reduced due to the influence of environmental factors, which affect subsequent detection, recognition, and other tasks. Therefore, this paper proposes an image rain removal algorithm based on multi-scale features, which can effectively remove rain streaks. First of all, this paper proposes a deraining algorithm that combines spatial information to improve the network’s generalization ability on real images, aiming at the problem of synthetic datasets used by previous deraining algorithms. Then, by proposing a multi-scale rain removal algorithm, it improves the feature extraction capabilities of existing deraining algorithms. Before extracting deep rain features, a preliminary fusion of multi-scale shallow features can be performed, which can show better performance in images of different sizes. In addition, a spatial attention module and channel are introduced. The attention module increases the ability to extract rain information at each scale; the resulting multi-scale feature image rain removal algorithm is called MFD. Finally, the rain removal algorithm is validated on the rain removal dataset, and the proposed method can effectively remove rain patterns, provide strong performance improvement on several datasets in the image rain removal task, and provide high-quality images for subsequent detection and recognition tasks.

Rainfall Estimation Model in Seasonal Zone and Non-Seasonal Zone Regions Using Weather Radar Imagery Based on a Gradient Boosting Algorithm

Indonesia, a country located in the equatorial region with hilly and valley lands surrounded by vast oceans, has complex rainfall patterns that can generally be classified into three types: equatorial, monsoon, and local. Rainfall estimates have only been derived based on local data and characteristics so far, and have not yet been developed based on universal data for all of Indonesia. This study aimed to develop a rainfall estimation model based on weather radar data throughout Indonesia using ensemble machine learning with the gradient boosting algorithm. The proposed rainfall estimation model is universal, can be applied to different rainfall pattern areas, and has a temporal resolution of 10 min. It is based on determining the root mean square error (RMSE) and R-squared (R2) values. Research was conducted in six areas with different rainfall patterns: Bandar Lampung and Banjarmasin with monsoon rain patterns, Pontianak and Deli Serdang with equatorial rain patterns, and the Gorontalo and Biak areas with local rain patterns. The analysis of the proposed model reveals that the best hyperparameters for the learning rate, maximum depth, and number of trees are 0.7, 3, and 50, respectively. The results demonstrate that the estimated rainfall in the six areas was very accurate, with RMSE < 2 mm/h and R2 > 0.7.

Spatiotemporal Variation and Causes of Typical Extreme Precipitation Events in Shandong Province over the Last 50 Years

In this study, based on hourly ERA5 reanalysis data from July to September, from 1971 to 2020, for Shandong Province, we used mathematical statistical analysis, the Mann–Kendall nonparametric statistical test, cluster analysis, and other methods to extract and analyze the spatiotemporal variation characteristics and causes of typical extreme precipitation events. The results indicated the following: (1) The total number and duration of precipitation events show a nonsignificant upward trend, while the average and extreme rainfall intensities show a nonsignificant downward trend. (2) Extreme precipitation events are primarily concentrated in Qingdao, Jinan, Heze, and Binzhou, with fewer events occurring in central Shandong Province. (3) Extreme precipitation events are classified into four types (namely, patterns I, II, III, and IV). Pattern I exhibits two rain peaks, with the primary rain peak occurring after the secondary rain peak. Similarly, pattern II also displays two rain peaks, with equivalent rainfall amounts for both peaks. In contrast, pattern III has multiple, evenly distributed rain peaks. Finally, pattern IV shows a rain peak during the first half of the precipitation event. Pattern I has the highest occurrence probability (46%), while pattern IV has the lowest (7%). (4) The spatial distributions of the different rain patterns are similar, with most being found in the eastern coastal and western regions. (5) Extreme precipitation events result from interactions between large-scale circulation configurations and mesoscale convective systems. The strong blocking situation and significant circulation transport at middle and low latitudes in East Asia, along with strong convergent uplift, abnormally high specific humidity, and high-water-vapor convergence centers, play crucial roles in supporting large-scale circulation systems and triggering mesoscale convective systems.

Local social-ecological context explains seasonal rural-rural migration of the poorest in south-west Bangladesh

Bangladesh is one of the countries most affected by climate change. Internal migration is often presented as a response to environmental degradation. Here, using a people-centred perspective, we explore the complexity of the links between climate-induced change, environmental degradation caused by waterlogging and seasonal rural migration. We used an inductive qualitative approach in social sciences, conducting fourteen semi-directed interviews and six focus group discussions in March-April 2022. We related those results to a rainfall analysis on CHIRPS data for 1981-2021and we represented interactions and feedback between changes and livelihoods in a model. A complex picture of the situation is emerging, showing the interweaving effects of non-climatic and climatic changes, their interplay at different scales, their cumulative effects, the interactions between livelihood types and feedback between social and natural systems. Most of the climate-induced changes gradually become noticeable over the past 25 years. Climate data confirm these changes in recent decades, with July being wetter and January being dryer. Villagers reported waterlogging as the most significant change in their community, pointing to its multiple causes, originating in non-local and local, non-climatic anthropic changes, exacerbated by shrimp farm enclosures and worsened by climate-induced changes such as heavier rains, wetter monsoons and cyclones. Tiger prawn farms, reported as a lucrative and local adaptation to waterlogging and salinisation for the ones who can afford it, worsen the situation for the less wealthy, causing waterlogging and salinisation of the adjacent agricultural lands and buildings, the disappearance of traditional fishing and a reduction of the local job market. In addition, erratic rain patterns, droughts and cyclones affect local production and labour markets. COVID-19 lockdowns, by impacting markets and mobilities, further aggravated the situation. Inequality has increased as the range of adaptations of the less wealthy appears limited in this context of multiple crises.

Urban inundation rapid prediction method based on multi-machine learning algorithm and rain pattern analysis

Urban inundation disasters caused by extreme rainfall events are becoming increasingly severe. However, the numerical inundation model based on physical process has relatively slow computational speed and struggle to meet the demands of current forecasting and early warning systems. Addressing how to integrate machine learning technology into rapid urban inundation prediction, especially when dealing with limited training samples, and selecting the most appropriate machine learning algorithms for different rain patterns, is a crucial and pressing issue. To tackle this problem, this work combines a coupled hydrological-hydrodynamic model known for its high-computational accuracy with efficient machine learning algorithms. Taking different rain patterns into account, a rapid urban inundation prediction method is proposed. This method utilizes data-driven results from the coupled hydrological-hydrodynamic model and constructs multiple machine learning algorithms based on the conclusions drawn from the analysis regarding the most suitable machine learning algorithms for different rain patterns. The results show that for single-peak, double-peak, and uniform rain patterns, the Ridge algorithm, KNN algorithm, and RF algorithm should be respectively employed. The method achieves a mean absolute percentage error of 5.32 %, 7.73 %, and 2.49 % for single-peak, double-peak, and uniform rainfall scenarios and can predict inundation within a 3.68 km2 area in 14.07 s. This method offers high prediction accuracy and fast computational speed, not only meeting the requirements of daily forecasting and early warning but also providing a new direction for the application of machine learning technology in urban inundation prevention.

Mitigation and adaptation of local organization to achieve food security: Case of traditional irrigation system in Bali, Indonesia

Climate change has affected the agricultural sector, particularly the food crop cultivated on rice fields. Subak as a local farmers organization is always expected to have important role to to improve rice farming practices and increase its productivity. The changes of rainfall and rain patterns happened in Bali should be anticipated by the subak through the mitigation and adaptation efforts to prevent the failure of harvest, and ensure better production. The aims of this study were to describe subaks activities and its problem, and the adaptation and mitigation efforts for achieving food security. The study was purposively conducted in 14 subaks which obtained irrigation water from the Buleleng River, Bali Province. A total of 140 farmers were selected as samples by using non-proportional sampling technique. Primary data and secondary data were collected using survey, direct observation, and FGD. Data collected were analyzed by a qualitative descriptive method. The results indicate that subaks have some problems, such as water availability, competition of water among the users, and rice field conversion. The efforts of adaptation and mitigation comprise the changes in cropping patterns and planting schedules, borrowing water between subak members, and forming a subaks federation to increase rice productivity and achieve food security programs. The government should provide useful weather information for farmers regarding water availability, make a digital map of forest, regulations, and law enforcement.

Aedes aegypti oviposition dynamics in towns with low human population density in yungas and dry chaco, Salta, Argentina

We assessed the presence of Aedes aegypti in five ecorregions of Salta province and compared the oviposition activity of Ae. aegypti using ovitraps in towns of two contrasting ecoregions (yungas and Chaco dry forests) in the province of Salta, Argentina, a major contrast in these ecoregions are rain patterns and altitude. Our aim was to estimate how oviposition activities were associated with the ecoregion and site scale local environmental variables. Mosquito oviposition activity was monitored weekly during the summer using ovitraps. Predictor variables were ecoregion, town, and meteorological variables. The effect of the predictor variables was measured on the response variables using multi-model inference. Besides yungas, the presence of Aedes aegypti was confirmed in towns of dry Chaco and High Monte. The only factor that had a significant effect on the presence of eggs in the ovitraps was the ecoregion, with the frequency of positives being higher in yungas. For the number of eggs, the ecoregion, the night temperature of the first week and the NDVI would explain said variable. Overall, results indicate that the variations between towns would be more related with their ecological and climatic characteristics than with the more immediate meteorological variations.

Lasting effects of avian‐frugivore interactions on seed dispersal and seedling establishment

Abstract The consequences of plant–animal interactions often transcend the mere encounter stage, as those encounters are followed by a chain of subsequent stages on the plant's reproductive cycle that ultimately determine fitness. Yet, the dissemination and recruitment stages of animal‐mediated seed dispersal are seldom analysed jointly, hindering a full understanding of the ecology of seed dispersal. We analyse the dispersal and recruitment of a fleshy‐fruited plant (Pistacia lentiscus), from fruit production to seedling survival up to their second year. We link early reproductive investment of individual plants to seedling recruitment and explore the role played by seed viability, the coterie of frugivores and microhabitat seed deposition. The proportion of viable seeds was generally low (mean = 34%) but highly variable among individual plants (range: 0%–95%). Seed viability did not seem to have a direct effect on individual plant's recruitment. We recorded 28 bird species feeding on P. lentiscus fruits or seeds. Their contribution to plant recruitment was mainly determined by their intensity of fruit consumption and probability to disperse viable seeds. Most frugivores presented non‐random microhabitat preferences, delivering uneven seed contributions to different sites. Post‐dispersal seed predation by rodents was the most limiting phase in P. lentiscus recruitment. Yet, microhabitats showing the lowest predation rates received the lowest seed rain. Hence, we found a decoupling of the dissemination and recruitment stages: most seeds do not arrive at the most suitable microhabitats. We estimate P. lentiscus plants need to produce c. 5 × 105 fruits to recruit a single seedling that survives to its second summer in our study site. Its success as a prevalent species in Mediterranean lowland landscapes relies on its high fecundity and thorough fruit removal and dispersal by a diversified frugivore assemblage, which compensates for the high seed unviability characteristic of this genus. Synthesis: Measuring the delayed, post‐dispersal outcomes of animal frugivory interactions may overturn inferences based on consumption observations only. Seed rain patterns are often decoupled from microhabitats' suitability for seedling recruitment. Hence, more integrative studies that encompass the entire plant reproductive cycle (from fruit production to seedling recruitment) are needed to fully understand frugivores' lasting contribution to plant regeneration in natural populations.

Classification of tropical cyclone rain patterns using convolutional autoencoder

Heavy rainfall produced by tropical cyclones (TCs) frequently causes wide-spread damage. TCs have different patterns of rain depending on their development stage, geographical location, and surrounding environmental conditions. However, an objective system for classifying TC rain patterns has not yet been established. This study objectively classifies rain patterns of North Atlantic TCs using a Convolutional Autoencoder (CAE). The CAE is trained with 11,991 images of TC rain rates obtained from satellite precipitation estimates during 2000−2020. The CAE consists of an encoder which compresses the original TC rain image into low-dimensional features and a decoder which reconstructs an image from the compressed features. Then, TC rain images are classified by applying a k-means method to the compressed features from the CAE. We identified six TC rain patterns over the North Atlantic and confirmed that they exhibited unique characteristics in their spatial patterns (e.g., area, asymmetry, dispersion) and geographical locations. Furthermore, the characteristics of rain patterns in each cluster were closely related to storm intensity and surrounding environmental conditions of moisture supply, vertical wind shear, and land interaction. This classification of TC rain patterns and further investigations into their evolution and spatial variability can improve forecasts and help mitigate damage from these systems.

Spatial patterns of torrential rain in the Haihe River Basin and the corresponding large‐scale circulation

AbstractThe classification of torrential rain is crucial to the flood mitigation and water resources, but the relevant research in the Haihe River Basin (HRB) is lacking. In this study, 156 regional torrential rain events from 1972 to 2022 were classified into two types using cluster analysis: eastern type and southern type. The eastern type is the typical torrential rain in the HRB, accounting for 61.5% of the total days, whereas the southern type mainly occurs in the piedmont plain with larger intensity and scope, which would pose more severe hazards. Corresponding large‐scale circulation patterns of the two types reveal a clear relationship between the HRB torrential rain and ascending motion, associated with the anomalies of low‐level moisture convergence and upper‐level wind divergence. The anomalous upper‐level divergence is accompanied by the northward expansion of South Asian High and intensified westerly jets. In the eastern type, strong low‐pressure anomalies north of Lake Baikal at mid‐level produce anomalous northwesterlies. The cold air from mid–high latitudes encounters enhanced southwesterly moist air accompanied by the northward expansion of Western Pacific Subtropical High (WPSH), resulting in the low‐level moisture convergence. In the southern type, the low‐level anomalous convergence is caused by a relatively weak cyclonic anomaly and significantly strengthened southeasterly moisture from the Pacific. The enhancement of southwesterlies results from the cooperative effect of the remarkably northwesterly WPSH and low‐pressure anomalies in South China Sea. As a whole, the eastern type is the result of convergence of cold air and southwesterly moisture driven by a dipole pattern at mid–high latitudes, whereas the southern type is mainly attributed to the accumulation of abundant southeasterly water vapour. This classification can identify high‐risk areas for torrential rain in the HRB and contribute to the forecast skills of extreme precipitation.

A review of natural and managed revegetation responses in two de-watered reservoirs after large dam removals on the Elwha River, Washington, USA

Large dam removals are increasing in frequency and the response of natural and managed revegetation is a critical consideration for managed restoration of dewatered reservoir landscapes post dam removal. The removal of two large dams on the Elwha River in 2011-2014 provides insight into reservoir revegetation. We review literature and datasets from 2012 through 2018, 1-6 years since reservoir dewatering, to compare pre-dam removal predictions on the Elwha to post-dam removal of natural revegetation, managed revegetation effects and invasive non-native vegetation response. Pre-dam removal hypotheses about natural revegetation did not predict species performance on reservoir sediments, seed rain patterns, or seed bank response. Sediment texture and landform affected multiple aspects of revegetation, including vegetation cover, species richness, woody stem densities and species composition. Reservoir drawdown timing influenced species composition and seedling densities. Predictions about managed revegetation effects were mixed. Planting trees and shrubs did not accelerate woody cover but did increase species richness. Seeding reduced non-native vegetation frequency and species richness, had no effect on vegetation cover on fine sediments, but increased vegetation cover on coarse sediments. Planting trees and shrubs during drawdown appeared to result in higher survival rates compared to plantings installed 1+ years post drawdown. Seeding Lupinus rivularis (riverbank lupine) on coarse sediments was successful and increased foliar nitrogen in planted conifers. Invasive non-native vegetation was correctly predicted to be more abundant in the Aldwell reservoir but did not preclude native species establishment in either reservoir, likely due to rapid establishment of native species and robust management that occurred before, during and after dam removal.