Skills and Expertise
Research Items (13)
- Aug 2018
The use of existing component-based modeling frameworks for integrated water resources modeling is currently hampered for some important use cases because they lack support for commonly used, topology-aware, spatiotemporal data structures. Additionally, existing frameworks are often accompanied by large software stacks with steep learning curves. Others lack specifications for deploying them on high performance, heterogeneous computing (HPC) infrastructure. This puts their use beyond the reach of many water resources modelers. In this paper, we describe new advances in component-based modeling using a framework called HydroCouple. This framework largely adopts the Open Modeling Interface (OpenMI) 2.0 interface definitions but demonstrates important advances for water resources modeling. HydroCouple explicitly defines standard and widely used geospatial data formats and provides interface definitions to support simulations on HPC infrastructure. In this paper, we illustrate how these advances can be used to develop efficient model components through a coupled urban stormwater modeling exercise.
Abstract In making the decision whether to use component-based modeling, its benefits must be balanced against computational costs. Studies evaluating these costs using the Open Modeling Interface (OpenMI) have largely used models with simplified formulations, small spatial and temporal domains, or a limited number of components. We evaluate these costs by applying OpenMI to a relatively complex Stormwater Management Model (SWMM) for the City of Logan, Utah, USA. Configurations of coupled OpenMI components resulting from decomposing the stormwater model by process (i.e., runoff coupled to routing) and then by space (i.e., groups of catchments coupled together) were compared to a reference model executed in the standard SWMM configuration. Simulation times increased linearly with the number of connections between components, and mass balance error was a function of the degree to which a component resolved time series data received. This study also examines and proposes some strategies to address these computational costs.
Urbanization, climate, and ecosystem change represent major challenges for managing water resources. Although water systems are complex, a need exists for a generalized representation of these systems to identify important components and linkages to guide scientific inquiry and aid water management. We developed an integrated Structure-Actor-Water framework (iSAW) to facilitate the understanding of and transitions to sustainable water systems. Our goal was to produce an interdisciplinary framework for water resources research that could address management challenges across scales (e.g., plot to region) and domains (e.g., water supply and quality, transitioning, and urban landscapes). The framework was designed to be generalizable across all human–environment systems, yet with sufficient detail and flexibility to be customized to specific cases. iSAW includes three major components: structure (natural, built, and social), actors (individual and organizational), and water (quality and quantity). Key linkages among these components include: (1) ecological/hydrologic processes, (2) ecosystem/geomorphic feedbacks, (3) planning, design, and policy, (4) perceptions, information, and experience, (5) resource access and risk, and (6) operational water use and management. We illustrate the flexibility and utility of the iSAW framework by applying it to two research and management problems: understanding urban water supply and demand in a changing climate and expanding use of green storm water infrastructure in a semi-arid environment. The applications demonstrate that a generalized conceptual model can identify important components and linkages in complex and diverse water systems and facilitate communication about those systems among researchers from diverse disciplines.
- Dec 2014
This study compared the sensitivity of water quality in tropical Aguamilpa Reservoir, as represented by normalized algae mass and dissolved oxygen, to selected projected changes from global climate change and development. The sensitivity of reservoir stratification as an indicator of reservoir sensitivity also was analysed. Model simulations indicated the reservoir was more sensitive to changes during the warm‐dry season than at other times. Both indexes (normalized algal mass and dissolved oxygen mass) were more sensitive to changes in air temperature (climate change) and nitrogen loading (development) than to changes in flow. The sensitivity to air temperature was similar to, but generally less than, the sensitivity to nutrient inflow. At the bounding values for change (3 °C for temperature; 50% increase in nitrogen loading), the algae mass sensitivities were 0.15 mg L−1 per 3 °C and 0.2 mg L−1 per 50% nitrogen load increase, and the dissolved oxygen mass sensitivities were 0.7 mg L−1 per 3 °C and 2.0 mg L−1 per 50% load increase. Changes in air temperature and nitrogen loadings affect the reservoir in different ways, air temperature mostly changing the timing of the algal growth with little change in peak values, while nutrient loadings change the peak values with little change in the timing. While the sensitivities are similar, the total algal mass change is significantly larger for nitrogen loading, compared to air temperature changes. These results imply global climate change effects can be partially mitigated by implementing management measures in the surrounding watersheds to minimize nutrient inflows, especially nitrogen in the case of Aguamilpa Reservoir.
- Jan 2017
One approach for performing uncertainty assessment in flood inundation modeling is to use an ensemble of models with different conceptualizations, parameters, and initial and boundary conditions that capture the factors contributing to uncertainty. However, the high computational expense of many hydraulic models renders their use impractical for ensemble forecasting. To address this challenge, we developed a rating curve library method for flood inundation forecasting. This method involves pre-running a hydraulic model using multiple inflows and extracting rating curves, which prescribe a relation between streamflow and stage at various cross sections along a river reach. For a given streamflow, flood stage at each cross section is interpolated from the pre-computed rating curve library to delineate flood inundation depths and extents at a lower computational cost. In this article, we describe the workflow for our rating curve library method and the Rating Curve based Automatic Flood Forecasting (RCAFF) software that automates this workflow. We also investigate the feasibility of using this method to transform ensemble streamflow forecasts into local, probabilistic flood inundation delineations for the Onion and Shoal Creeks in Austin, Texas. While our results show water surface elevations from RCAFF are comparable to those from the hydraulic models, the ensemble streamflow forecasts used as inputs to RCAFF are the largest source of uncertainty in predicting observed floods.
Project - HydroCouple - Component-based modeling framework for integrated earth systems and environmental modeling
The latest version of the HydroCoupleComposer model coupling environment and associated components for Microsoft Windows is now available. Please download it here (https://github.com/HydroCouple/HydroCoupleComposer/releases/tag/v1.1.0). Suggestions and feedback are welcome.
Transitioning from the traditional approach of executing water resources models on single desktop computers to increasingly ubiquitous High Performance Heterogeneous Computing (HPC) infrastructure introduces efficiencies that could help advance the degree of fidelity of models to the underlying physical processes they simulate. For example, model developers may be able to incorporate more physically-based formulations, perform computations over finer spatial and temporal scales, and perform simulations that span long time periods with reasonable execution times. Additionally, computationally expensive simulations including parameter estimation, uncertainty assessment, multi-scenario evaluations, etc. may become more tractable. The use of HPC for executing these types of simulations within component-based modelling frameworks is an approach that is still largely underutilized in the water resources modeling arena. In this abstract, we describe advancements that we have implemented in the HydroCouple component-based modeling framework to allow water model developers to take advantage of heterogeneous, multi-accelerator clusters. HydroCouple largely employs the OpenMI interface definitions but adds new interfaces to better support standardized geo-temporal data structures, customizable coupled model data exchange workflows, and distributed computations on HPC infrastructure. We also describe how some of these advancements have been used to develop coupled models for two applications: 1) coupling of a one-dimensional storm sewer model with a high resolution, two-dimensional, and overland riverine model for an urban stormwater conveyance system, and 2) coupling of a series of model components being developed to simulate heat transport in heterogeneous rivers with significant longitudinal flow variability.
Project - HydroCouple - Component-based modeling framework for integrated earth systems and environmental modeling
Installers for the first release of the HydroCouple component-based modeling composition graphical user and command line interface called HydroCouple Composer has been released. Please download it on GitHub (https://github.com/HydroCouple/HydroCoupleComposer/releases/tag/v1.0.0).
HydroCouple is a cross-platform, component-based modeling interface definition that largely follows the Open Modeling Interface 2.0 (OpenMI) specification. HydroCouple provides advancements to better facilitate those experimental model investigations that fall into the so-called " embarrassingly parallel " class of simulations, including uncertainty assessment, ensemble forecasting, and optimization simulations. Additionally, HydroCouple explicitly incorporates low level interface definitions for multi-dimensional datasets and geospatial data formats including the Open Geospatial Consortium's Simple Feature Access specification, raster datasets, and meshes that are widely used in the earth systems and environmental modeling field. In this paper, we describe these and other advances provided by the HydroCouple interface definitions. We also illustrate how these advances can be used to facilitate parallelized experimental model simulations that have so far been challenging in OpenMI and other component-based modeling frameworks.
Many researchers have highlighted the need to incorporate uncertainty assessment in flood inundation modeling. One approach to accomplish this goal is to use an ensemble of models with different conceptualizations, initial conditions, and parameterizations that capture the factors that contribute to uncertainty in model predictions. However, the computational expense associated with many hydraulic models, prohibit their use for such evaluations, especially when dealing with large spatial scales in real-time forecasting scenarios. To overcome this challenge, we use a rating curve library approach for inundation delineation. This approach involves pre-computing a rating curve at various cross-sections along a river by running a hydraulic model over a range of flows. River stage at each cross-section for a given flow forecast can then be readily interpolated from the rating curve library, thereby providing a computationally efficient and scalable method to delineate local inundation areas. We apply this method using a forecast automation software tool we have developed called the Rating Curve based Automatic Flood Forecasting (RCAFF; https://github.com/calebbuahin/RCAFF) tool, to streamline the workflow associated with the method. This software extracts cross sections and ratings curves from an existing HEC-RAS model; downloads real-time ensemble flow forecasts derived by downscaling runoff European Center for Medium range Weather Forecasting (ECMWF) model routed through the Routing Application for Parallel Computation of Discharge (RAPID) model; and dynamically delineates inundation extents and inundation probability maps. We investigate the feasibility of using this rating curve library method by applying the RCAFF tool to Shoal Creek in Austin, TX.
Component-based environmental modeling, or loose model coupling, has been proposed as an alternative to traditional tight model coupling, which is characterized by inflexible and often large model codes with highly interdependent functions compiled into a single execution unit. Loosely coupling model components developed by decomposing complex systems into smaller or less complex independent units promises earth systems modelers: (1) an approach to better explore feedbacks between domains of different disciplines that are typically modeled independently, and (2) a way to experiment with different process formulations to select those that are most appropriate for a particular application. The additional function calls, data transformations, and discontinuity at the connection points between model components resulting from the use of component-based modeling may, however, give rise to computational penalties, including increased simulation time and mass balance error. In the study presented here, we sought to investigate these computational penalties as the number of coupled model components increases using the Open Modeling Interface (OpenMI), which is component-based modeling interface specification. A Stormwater Management Model (SWMM) application developed for the City of Logan, Utah, USA and run in its standard, tightly coupled configuration served as a reference against which several configurations of coupled OpenMI-compliant SWMM model components were compared. The various configurations of coupled OpenMI SWMM model components were derived by decomposing the reference SWMM model first by process (i.e., runoff coupled to routing) and then by space (i.e., catchments or groups of catchments coupled together). Results showed that simulation times increased linearly as the number of connections between model components increased. The results also showed that changes in total mass balance error introduced through coupling were dependent on how well each model component was able to resolve the time series data it received. This study also demonstrates and proposes some strategies to address these computational penalties in component-based modeling frameworks.
Visualization of water quality associated with rivers is difficult because most rivers have sparse data collection points and visualizing the spatial trends described by these sparse data is challenging. Traditional isopleths maps (contours) are not suited for display of these data because the data are associated only with the stream or river and the mathematics associated with isopleth generation (e.g., kriging, distance weighted interpolation, etc.) assume the shortest distance between two points and isopleths only present one parameter. We have found that "snake diagrams", popularized by Edward Tufte, are an effective visualization technique for river data. Snake diagrams present two or three data dimensions, such as flow and concentration using width and color, each dimension corresponding to the associated data. A third parameter can be associated with the height of the "snake". The ability to present two or three parameters allows analysis of correlations and trends not apparent if the parameters are viewed individually. We have developed software with several spatial interpolation and data classification techniques to create snake diagrams to visualize and assess water quality in river networks. The Snake diagrams present the spatial treads and the correlation among the study variables and the software allows the analyst to explore various presentations. The software also generates more standard charts and graphs along with the data statistics. The snake diagrams can be exported and visualized in Google EarthTM using KML files. As with the software, the KML files can be animated to show temporal trends in addition to the other data dimensions that are presented.
Modern water quality field sensors, probes, and sondes allow the collection of significant quantities of data with high temporal resolution. These instruments can also collect high-resolution vertical profile data for multiple water quality parameters. Until recently, regular collection of these large datasets has been rare; even more uncommon are tools that can be used to analyze them. We developed a set of tools to import, store, and analyze vertical profile data and single point measurements. We use Microsoft Excel to perform the initial data evaluation and processing, Microsoft Access as a personal geodatabase, and ArcMap tools to analyze, visualize, and communicate the collected data. The ArcMap tools provide vertical plots of multiple variables, spatial distributions of variables at a specific elevations or depths, and animations of any of the data presentations over time. The tools also compute statistics and correlations among the variables from quantitative analysis. The collected data are stored and referenced by location, time, depth, and elevation. The elevations are interpolated to 0.25 foot elevations to allow spatial analysis of the data across monitoring points and through time. Vertical profiles can be analyzed in various ways. For example, the profile at a monitoring point can be plotted over time, this can show correlation among parameters, reservoir processes associated with depth such as thermal clines or oxygen deficient areas, and other phenomenon. These data could be analyzed at multiple monitoring points at a single depth over time, for example looking at spatial averages of selected depth measurements to analyze reservoir spatial processes such as the impact of inflows or using simple time plots of parameters at a single monitoring point at selected depths. Data can also be analyzed by elevation rather than depth since, for most reservoirs, significant depth fluctuations occur annually. Elevation, rather than depth, can be important in some cases such as issues associated with a dam outlet elevation.