Explore the vast range of titles available.

Although decades of effort have been devoted to building Physical‐Conceptual (PC) models for predicting the time‐series evolution of geoscientific systems, recent work shows that Machine Learning (ML) based Gated Recurrent Neural Network technology can be used to develop models that are much more accurate. However, the difficulty of extracting physical understanding from ML‐based models complicates their utility for enhancing scientific knowledge regarding system structure and function. Here, we propose a physically interpretable Mass‐Conserving‐Perceptron (MCP) as a way to bridge the gap between PC‐based and ML‐based modeling approaches. The MCP exploits the inherent isomorphism between the directed graph structures underlying both PC models and GRNNs to explicitly represent the mass‐conserving nature of physical processes while enabling the functional nature of such processes to be directly learned (in an interpretable manner) from available data using off‐the‐shelf ML technology. As a proof of concept, we investigate the functional expressivity (capacity) of the MCP, explore its ability to parsimoniously represent the rainfall‐runoff (RR) dynamics of the Leaf River Basin, and demonstrate its utility for scientific hypothesis testing. To conclude, we discuss extensions of the concept to enable ML‐based physical‐conceptual representation of the coupled nature of mass‐energy‐information flows through geoscientific systems. Plain Language Summary We develop a physically interpretable computational unit, referred to as the Mass‐Conserving‐Perceptron (MCP). Networks of such units can be used to model the conservative nature of the input‐state‐output dynamics of mass flows in geoscientific systems, while Machine Learning (ML) technology can be used to learn the functional nature of the physical processes governing such system behaviors. Testing using data from the Leaf River Basin demonstrates the considerable functional expressivity (capacity) and interpretability of even a single‐MCP‐node‐based model, while providing excellent predictive performance and the ability to conduct scientific hypothesis testing. The concept can easily be extended to facilitate ML‐based physical‐conceptual representation of the coupled nature of mass‐energy‐information flows through geoscientific systems, thereby facilitating the development of synergistic physics‐AI modeling approaches. Key Points We develop a physically interpretable unit (Mass‐Conserving‐Perceptron) that can be used as a basic component of geoscientific models Off‐the‐shelf Machine Learning technology can be used to learn the functional nature of the physical processes governing system behaviors The concept can be extended to facilitate ML‐based representation of coupled mass‐energy‐information flows in geoscientific systems

We investigate the applicability of machine learning technologies to the development of parsimonious, interpretable, catchment‐scale hydrologic models using directed‐graph architectures based on the mass‐conserving perceptron (MCP) as the fundamental computational unit. Here, we focus on architectural complexity (depth) at a single location, rather than universal applicability (breadth) across large samples of catchments. The goal is to discover a minimal representation (numbers of cell‐states and flow paths) that represents the dominant processes that can explain the input‐state‐output behaviors of a given catchment, with particular emphasis given to simulating the full range (high, medium, and low) of flow dynamics. We find that a “HyMod Like” architecture with three cell‐states and two major flow pathways achieves such a representation at our study location, but that the additional incorporation of an input‐bypass mechanism significantly improves the timing and shape of the hydrograph, while the inclusion of bi‐directional groundwater mass exchanges significantly enhances the simulation of baseflow. Overall, our results demonstrate the importance of using multiple diagnostic metrics for model evaluation, while highlighting the need for properly selecting and designing the training metrics based on information‐theoretic foundations that are better suited to extracting information across the full range of flow dynamics. This study sets the stage for interpretable regional‐scale MCP‐based hydrological modeling (using large sample data) by using neural architecture search to determine appropriate minimal representations for catchments in different hydroclimatic regimes. Plain Language Summary We show that conventional machine learning technologies can be used to develop parsimonious, interpretable, catchment‐scale hydrologic models using the mass‐conserving perceptron (MCP) as a fundamental computational unit. Using data from the Leaf River Basin, we test a variety of minimal, dominant process, representations that can explain the input‐state‐output dynamics of the catchment. Our results demonstrate the importance of using multiple diagnostic metrics for evaluation and comparison of different model architectures, and highlight the importance of choosing (or designing) objective functions for model training that are properly suited to the task of extracting information across the full range of flow dynamics. This depth‐focus study sets the stage for interpretable regional‐scale MCP‐based hydrological modeling (using large sample data) by using neural architecture search to determine appropriate minimal representations for catchments in different hydroclimatic regimes. Key Points We utilize mass‐conserving perceptron (MCP) directed‐graph architectures to develop concise, interpretable catchment‐scale hydrologic models We focus on model complexity (depth) at a single location, rather than universal applicability (breadth) across large samples of catchments This study set the stage for interpretable MCP‐based modeling to find minimal representations in different hydroclimatic regimes

Flood hazard is a dynamic nonstationary phenomenon, which can be categorized based on the origin of the inundation. Inland flood hazard arises primarily from pluvial and fluvial inundations, typically modeled separately with respect to the pertaining spatial domains of the assessment, namely the urban areas and the riverine floodplains. When modeling is based on the catchment‐scale hydrological‐hydrodynamic approach, the inundations such as those resulting from pluvial and fluvial processes are usually not discerned, even though disparities in normative flood risk management exist in different countries. This paper establishes a tracer‐aided criterion to discretize between pluvial and fluvial flooding at a catchment scale, relying on the advection process of a conservative tracer. Applied to a small urban catchment for multiple probabilistic rainfall scenarios, our physically based methodology shows that the incorporation of a transport equation within a shallow water model can be used to define the inundation sources. We highlight the advantages of the proposed approach compared to commonly employed modeling techniques for mapping fluvial inundations, while emphasizing the significance of mapping and regulating pluvial hazards in urban areas. The study shows the potential role of an ion of the tracers' transport toward identifying the hazard sources in a catchment‐scale 2D numerical framework.

Extreme rainfall events often trigger landslides, debris flows, and sediment-laden floods that cause severe damage in built-up areas, yet sediment deposition is rarely quantified in hazard assessments. This study evaluates the capability of the physically based catchment model LISEMHazard to reconstruct sediment generation, transport, and deposition during Hurricane Maria (2017) in two catchments in Dominica (Coulibistrie and Grand Bay). Simulations were performed at 10 m resolution using rainfall, topography, soil, and land-use data. Model calibration and validation used mapped landslides and debris flows, field measurements of deposition height, and DEMs of Difference (DoDs). LISEMHazard reproduced the general magnitude of sediment volumes and the frequency–area distribution of medium and large landslides but showed poor ability to predict their exact locations and overestimated landslide depth and deposition height. Agreement between modeled and observed debris-flow patterns was good in major channels but weak in minor ones. Sensitivity analysis indicated that soil depth and cohesion dominate uncertainties, whereas saturated hydraulic conductivity and surface roughness exert minimal influence. Despite substantial data and model limitations, physically based modeling remains a practical approach for spatial estimation of sediment deposition needed for risk assessment, structural damage evaluation, and cleanup cost estimation.

An improved understanding of bioretention cell (BC) design configuration at both the unit scale and catchment scale is necessary for critical insight into dynamical behaviors of design parameters, which resultantly guides and improves the effectiveness and efficiency of a BC. A comprehensive sensitivity analysis (SA) of BC design parameters was conducted in this study by using the Stormwater Management Model (SWMM) which is globally used for BC's modeling. The preliminary screening of various design parameters is conducted by the one-factor-at-a-time (OAT) SA method and the key influential parameters (i.e., conductivity, berm height, vegetation volume, suction head, porosity, wilting point, and soil thickness) are selected for further SA. To this end, 1,000 random uniformly distributed samples of each sensitive design parameter are simulated by a Python wrapper of SWMM (PySWMM) under different design storms at the unit scale and catchment scale, respectively. Unit-scale SA results found unique characteristics of each design parameter under different storm scenarios, and their behaviors toward different model responses dynamically change within their factor spaces. Catchment-scale SA results conclude vegetation and soil layers design parameters have significant impacts on controlling stormwater at the catchment scale, and optimal selection of design parameters of vegetation (type, density, and height) and soil (type, layer thickness, and void ratio) is necessary for significantly improving the effectiveness of the BC at the catchment scale.

Event flow characteristics were evaluated based on temperature and level of stream water in 22 forested catchments (area: 13.2–281.4 ha) to investigate sustainable flood management measures. Temperature and stream water levels were during 346 rainfall events in the summer season (July–September) from 2020 to 2022. Rising stream water levels responded to falling stream water temperature between ≤100 and >100 ha forested catchments in two types of time of concentration. Stream water temperature decreased by 3.0 °C when the stream water level increased by up to 0.9 m during rainfall events. Falling stream water temperature at two types of time of concentration was negatively correlated with total precipitation and rising stream water level. Based on the relatively high value of regression and cumulative frequency distribution, the estimated rising stream water level was appropriate in small catchments (≤100 ha) when the stream water temperature decreased, and the stream water level increased during rainfall events. Rising stream water levels and falling stream water temperatures are responses to catchment-scale effects, which are influenced by the nature and rapidity of the hydrological responses. Therefore, the results of the present study indicate that spatial and temporal differences in thermal responses of stream water temperature to water levels were controlled by catchment-scale effects under rapidly changing rainfall.

A wavelet-based local mesh refinement (wLMR) strategy is designed to generate multiresolution and unstructured triangular meshes from real digital elevation model (DEM) data for efficient hydrological simulations at the catchment scale. The wLMR strategy is studied considering slope- and curvature-based refinement criteria to analyze DEM inputs: the slope-based criterion uses bed elevation data as input to the wLMR strategy, whereas the curvature-based criterion feeds the bed slope data into it. The performance of the wLMR meshes generated by these two criteria is compared for hydrological simulations; first, using three analytical tests with the systematic variation in topography types and then by reproducing laboratory- and real-scale case studies. The bed elevation on the wLMR meshes and their simulation results are compared relative to those achieved on the finest uniform mesh. Analytical tests show that the slope- and curvature-based criteria are equally effective with the wLMR strategy, and that it is easier to decide which criterion to take in relation to the (regular) shape of the topography. For the realistic case studies: (i) slope analysis provides a better metric to assess the correlation of a wLMR mesh to the fine uniform mesh and (ii) both criteria predict outlet hydrographs with a close predictive accuracy to that on the uniform mesh, but the curvature-based criterion is found to slightly better capture the channeling patterns of real DEM data.

Much land has been cleared of indigenous forest for pastoral agriculture worldwide. In New Zealand, the clearance of indigenous forest on hill country has resulted in high food production, but waterways have become turbid, with high nutrient and E. coli concentrations. A range of on-farm mitigations are available, but it is unclear how they should be applied catchment-wide. We have developed a catchment-scale model that integrates economics with ecosystem services to find a better balance between agriculture and nature. In the upper Wairua catchment, Northland, if three actions are prioritised—(1) keeping stock out of streams, (2) constructing flood retention bunds in first-order catchments, and (3) planting trees on highly erodible land—then sediment loads, E. coli levels, and flooding are significantly reduced. Implementing these actions would cost approximately 10% of catchment net revenue, so it is feasible with a combination of regulation and subsidy. Many catchments in New Zealand are primarily pastoral agriculture, as in other countries (in North and South America, Australasia, and the United Kingdom), and would benefit from the analysis presented here to guide development along sustainable pathways. While pastoral agriculture typically stresses waterways, with increased sedimentation and freshwater contaminants, much can be done to mitigate these effects with improved farm and riparian management.

Practical approaches for managing flooding from fluvial sources are moving away from mitigation solely at the point of impact and towards integrated catchment management. This considers the source areas, flow pathways of floodwaters and the locations and exposure to the risk of communities. For a field site in southern Nepal, we analyse catchment response to a range of simulated rainfall events, which when evaluated collectively can help guide potential flood management solutions. This is achieved through the adoption of SCIMAP‐Flood, a decision support framework that works at the catchment‐scale to identify critical source areas for floodwaters. The SCIMAP‐Flood Fitted inverse modelling approach has been applied to the East Rapti catchment, Nepal. For multiple flood impact locations throughout the catchment, SCIMAP‐Flood effectively identifies locations where flood management measures would have the most positive effects on risk reduction. The results show that the spatial targeting of mitigation measures in areas of irrigated and rainfed agriculture and the prevention of deforestation or removal of shrubland would be the most effective approaches. If these actions were in the upper catchment above Hetauda or upstream of Manahari they would have the most effective reduction in the flood peak.

Increasing rainfall intensity and frequency due to extreme climate change and haphazard land development are aggravating soil erosion problems in Korea. A quantitative estimate of the amount of sediment from the catchment is essential for soil and water conservation planning and management. Essential to catchment-scale soil erosion modeling is the ability to represent the fluvial transport system associated with the processes of detachment, transport, and deposition of soil particles due to rainfall and surface flow. This study applied a spatially distributed hydrologic model of rainfall–runoff–sediment yield simulation for flood events due to typhoons and then assessed the impact of topographic and climatic factors on erosion and deposition at a catchment scale. Measured versus predicted values of runoff and sediment discharge were acceptable in terms of applied model performance measures despite underestimation of simulated sediment loads near peak concentrations. Erosion occurred widely throughout the catchment, whereas deposition appeared near the channel network grid cells with a short hillslope flow path distance and gentle slope; the critical values of both topographic factors, providing only deposition, were observed at 3.5 (km) (hillslope flow path distance) and 0.2 (m/m) (local slope), respectively. In addition, spatially heterogeneous rainfall intensity, dependent on Thiessen polygons, led to spatially distinct net-erosion patterns; erosion increased gradually as rainfall amount increased, whereas deposition responded irregularly to variations in rainfall.