Numeric hydrologic models can aid in water resource management by providing predictive simulations of water behavior. As computers become more advanced, the models developed also become more complex using more data to represent larger areas for forecasting hydrologic behavior. Unfortunately, as the simulations use more data, the output often becomes difficult to manage and share without investing time and effort into setting up server environments or decreasing the quality of the output to compensate for an efficient and effective user experience. A proposed solution to facilitate the accessibility of massive hydrologic model output is through the web-based visualization tool developed at Carnegie Mellon University called Time Machine. For a more efficient and automated workflow, a Python tool named TMAPS was developed from this research for rendering hydrologic model results, geoprocessing the rendered output, and generating Time Machines seamlessly. The tool can be installed from the CI-WATER GitHub repository and allows the user to 1) select the output parameters and visualization settings desired to be rendered, 2) run the code on a local or HPC setup, and 3) use a web browser interface to view the tiled transient results seamlessly while maintaining high quality. Currently, the only hydrologic model supported is ADHydro - a large-scale high-resolution multi-physics watershed simulation. In an effort to facilitate organizing the library of Time Machine products, an app was created through Tethys - a server-based Django application designed to aid in the development and sharing of water resource engineering apps.

Advances in water resources modeling are improving the information that can be supplied to support decisions that affect the safety and sustainability of society, but these advances result in models being more computationally demanding. To facilitate the use of cost- effective computing resources to meet the increased demand through high-throughput computing (HTC) and cloud computing in modeling workflows and web applications, I developed a comprehensive Python toolkit that provides the following features: (1) programmatic access to diverse, dynamically scalable computing resources; (2) a batch scheduling system to queue and dispatch the jobs to the computing resources; (3) data management for job inputs and outputs; and (4) the ability for jobs to be dynamically created, submitted, and monitored from the scripting environment. To compose this comprehensive computing toolkit, I created two Python libraries (TethysCluster and CondorPy) that leverage two existing software tools (StarCluster and HTCondor). I further facilitated access to HTC in web applications by using these libraries to create powerful and flexible computing tools for Tethys Platform, a development and hosting platform for web-based water resources applications. I tested this toolkit while collaborating with other researchers to perform several modeling applications that required scalable computing. These applications included a parameter sweep with 57,600 realizations of a distributed, hydrologic model; a set of web applications for retrieving and formatting data; a web application for evaluating the hydrologic impact of land-use change; and an operational, national-scale, high- resolution, ensemble streamflow forecasting tool. In each of these applications the toolkit was successful in automating the process of running the large-scale modeling computations in an HTC environment.

Floods are a significant threat to communities around the world and require substantial resources and infrastructure to predict. Limited local resources in developing nations make it difficult to build and maintain dense sensor networks like those present in the United States, creating a large disparity in flood prediction across borders. To address this disparity, I operated the Iowa Flood Center Top Layer model to predict floods in Puerto Rico without relying on in-situ data measurements. Instead, all model forcing was provided by satellite remote sensing datasets that offer near-global coverage. I used three datasets gathered via satellite remote sensing to build and operate watershed streamflow models: elevation data obtained by the Space Shuttle Endeavour through the Shuttle Radar Topography Mission (SRTM), rainfall estimates gathered by a constellation of satellites through the Global Precipitation Measurement Mission (GPM), and evapotranspiration rate estimates collected by Moderate Resolution Imaging Spectroradiometer (MODIS) sensors aboard the Aqua and Terra satellites. While these satellite remote sensing datasets make observations of nearly the entire world, their spatiotemporal resolution is coarse compared to conventional on-the-ground measurements. Hydrologic models were assembled for 75 basins upstream of streamflow gages monitored by the United States Geologic Survey (USGS). Model simulations were compared to real-time measurements at these gages. Continuous simulations spanning 58 months achieve poor Nash Sutcliffe Efficiency and Klinge Gupta Efficiency of -112.0 and -0.5, respectively. The sources of error that influence model performance were investigated, underlining some limitations of relying solely on satellite data for operational flood prediction efforts.

At the large watershed scale, we emphasize the effects of flow through a river network on streamflow under dry conditions. An immediate consequence of assuming dry conditions is that evapotranspiration causes flow in the river network to exhibit oscillations. When all links in the river network combine their flow patterns, the oscillations interact in ways that change the timing and amplitude of the streamflow waves at the watershed outlet. The geometric shape of the river network is particularly important, so we develop an analytic solution for streamflow which emphasizes that importance. Doing hydrology backward is a strategy recently developed by several researchers to deal with uncertainty in measurements of forcing terms applied to hydrologic models. The strategy has also been applied to resolve the assumption of homogeneity on realistic catchments that exhibit many heterogeneous properties. In this work, we demonstrate hydrology in the backward direction applied to two examples: using streamflow at the catchment scale to determine runoff at the hillslope scale and using the hillslope runoff to infer the applied evapotranspiration forcing under the assumption of dry conditions. In order to work across scales, we utilize the analytic solution for streamflow at the outlet of a river network. At the hillslope scale, we develop a soil model to create fluxes consistent with observed soil processes.

This work explores the use of hydroinformatics tools to provide a user friendly and accessible interface for executing and visualizing the output of distributed hydrological models for Iowa. It uses an IFIS-based web environment for graphical displays and it communicates with the ASYNCH ODE solver to provide input parameters and to gather modeling outputs. The distributed hydrologic models used here are based on the segmentation of the terrain into hillslope-link hydrologic units, for which water flow processes are represented by sets of nonlinear ordinary differential equations. This modeling strategy has shown promising results in in modeling extreme flood events in the state of Iowa – USA. The usage and evaluation of outputs from hillslope-link models (HLM) has been limited to a restrict group of academics due to the demand of high processing capability and the number of customized tools needed to visualize model outputs. HLM-based models provide abundant output information on rainfall-runoff processes of the hydrological cycle, including estimates of discharge for all streams in the state of Iowa, and for all conceptual vertical layers of water storage in soils. The interfaces and methodologies developed in this thesis respond to the constant demand for communicating effectively water-related information from academic communities to the public using hydroinformatics tools to provide an accessible portal to the information generated by complex hydrological models. It also facilitates model development and evaluation by allowing rapid development of what-if scenarios. This work represents a significant advance in this direction, and the results have been made publicly available online under the URL http://ifis.iowafloodcenter.org/ifis/sc/modelplus/.

Floodplains, and the sediment accumulating naturally on them, are important to maintain stream water quality and serve as sinks for organic and inorganic carbon. Newer theories contend that land use and hydrologic connectivity (water-mediated transport of matter, energy, and/or organisms within or between elements of the hydrologic cycle) play important roles in determining sediment accumulation on floodplains. This study hypothesizes that changes in hydrologic connectivity have a greater impact on floodplain sediment accumulation than changes in land use. Nine sediment cores from seven sub-basins were collected from the Savannah River Site (SRS), South Carolina, and processed for grain-size, radionuclide dating (7Be, 137Cs, 210Pb), particulate organic carbon (POC), and microscopy. Historical records, including aerial and satellite imagery, were used to identify anthropogenic disturbances in the sub-basins, as well as to calculate the percentages of natural vegetation land cover at the SRS in 1951, and 2014. LiDAR and field survey data identified 251 flow impediments, measured elevation, and recorded standard stream characteristics (e.g., bank height) that can affect hydrologic connectivity. Radionuclide dating was used to calculate sediment mass accumulation rates (MARs) and linear accumulation rates (LARs) for each core. Results indicate that sedimentation rates have increased across all SRS sub-basins over the past 40-50 years, shortly after site restoration and recovery efforts began. Findings show that hydrologic connectivity proxies (i.e., stream characteristics and impediments) have stronger relationships to MARs and LARs than the land use proxy (i.e., vegetation cover), confirming the hypothesis. As stream channel depth and the number of impediments increase, floodplain sedimentation rates also increase. This knowledge can help future stream restoration efforts by focusing resources to more efficiently attain stated goals, particularly in terms of floodplain sediment retention.

Green Stormwater Infrastructure (GSI) has become a popular method for flood mitigation as it can prevent runoff from entering streams during heavy precipitation. In this study, a recently developed neighborhood in Gresham, Oregon hosts a comparison of various GSI projects on runoff dynamics. The study site includes dispersed GSI (rain gardens, retention chambers, green streets) and centralized GSI (bioswales, detention ponds, detention pipes). For the 2017-2018 water year, hourly rainfall and observed discharge data is used to calibrate the EPA's Stormwater Management Model to simulate rainfall-runoff dynamics, achieving a Nash-Sutcliffe efficiency of 0.75 and Probability Bias statistic of 3.3%. A synthetic scenario analysis quantifies the impact of the study site GSI and compares dispersed and centralized arrangements. Each test was performed under four precipitation scenarios (of differing intensity and duration) for four metrics: runoff ratio, peak discharge, lag time, and flashiness. Design structure has significant impacts, reducing runoff ratio 10 to 20%, reducing peak discharge 26 to 68%, and reducing flashiness index 56 to 70%. There was a reverse impact on lag time, increasing it to 50 to 80%. Distributed GSI outperform centralized structures for all metrics, reducing runoff ratio 22 to 32%, reducing peak discharge 67 to 69%, increasing lag time 133 to 500%, and reducing flashiness index between 32 and 62%. This research serves as a basis for researchers and stormwater managers to understand potential impact of GSI on reducing runoff and downstream flooding in small urban watersheds with frequent rain.

East Iowa experienced large flooding during June of 2008. This study used Cuencas hydrological model to simulate the discharges of June 2008 at Eldorado and Elkader, in the Turkey River Watershed, in North East Iowa. The results of this study were used to test the performance of Cuencas modeling this flood event and to explore the role of land cover change in the floods of 2008 at Elkader, Iowa. Cuencas was found to be a suitable tool to predict this event, that requires relatively low resources. The total time to run each simulation was around two hours which is reasonable for such large watershed (900 mi2), but a computer cluster was needed to run these simulations. The results from this study suggest that the role of land cover change from pre-settlement to current conditions was significant when using the rainfall conditions of 2008. The discharges simulated at Elkader, Iowa were almost twice as large when using the 2001 land cover, than when using the land cover found during 1832-1859, recorded during the General Land Office (GLO) survey. These results need to be taken only as preliminary results, since there is no data to validate the model at the time of the GLO survey, and since it is the first time that Cuencas is used to model the effects of land cover in Iowa's hydrology. However, the potential large reduction on discharge of the pre-settlement land cover is an incentive to investigate this issue further and continue developing Cuencas to capture the effects of less drastic land cover changes.

Intensive human-environment interactions are taking place in Midwestern agricultural systems. An integrated modeling framework is suitable for predicting dynamics of key variables of the socio-economic, biophysical, hydrological processes as well as exploring the potential transitions of system states in response to changes of the driving factors. The purpose of this dissertation is to address issues concerning the interacting processes and consequent changes in land use, water balance, and water quality using an integrated modeling framework. This dissertation is composed of three studies in the same agricultural watershed, the Clear Creek watershed in East-Central Iowa. In the first study, a parsimonious hydrologic model, the Threshold-Exceedance-Lagrangian Model (TELM), is further developed into RS-TELM (Remote Sensing TELM) to integrate remote sensing vegetation data for estimating evapotranspiration. The goodness of fit of RS-TELM is comparable to a well-calibrated SWAT (Soil and Water Assessment Tool) and even slightly superior in capturing intra-seasonal variability of stream flow. The integration of RS LAI (Leaf Area Index) data improves the model's performance especially over the agriculture dominated landscapes. The input of rainfall datasets with spatially explicit information plays a critical role in increasing the model's goodness of fit. In the second study, an agent-based model is developed to simulate farmers' decisions on crop type and fertilizer application in response to commodity and biofuel crop prices. The comparison between simulated crop land percentage and crop rotations with satellite-based land cover data suggest that farmers may be underestimating the effects that continuous corn production has on yields (yield drag). The simulation results given alternative market scenarios based on a survey of agricultural land owners and operators in the Clear Creek Watershed show that, farmers see cellulosic biofuel feedstock production in the form of perennial grasses or corn stover as a more risky enterprise than their current crop production systems, likely because of market and production risks and lock in effects. As a result farmers do not follow a simple farm-profit maximization rule. In the third study, the consequent water quantity and quality change of the potential land use transitions given alternative biofuel crop market scenarios is explored in a case study in the Clear Creek watershed. A computer program is developed to implement the loose-coupling strategy to couple an agent-based land use model with SWAT. The simulation results show that watershed-scale water quantity (water yield and runoff) and quality variables (sediment and nutrient loads) decrease in values as switchgrass price increases. However, negligence of farmers risk aversions towards biofuel crop adoption would cause overestimation of the impacts of switchgrass price on water quantity and quality.

This dissertation represents the research and development of new concepts and techniques for modeling the knowledge about the many concepts we as hydrologists must understand such that we can execute models that operate in terms of conceptual abstractions and have those abstractions translate to the data, tools, and models we use every day. This hydrologic knowledge includes conceptual (i.e. semantic) knowledge, such as the hydrologic cycle concepts and relationships, as well as functional (i.e. procedural) knowledge, such as how to compute the area of a watershed polygon, average basin slope or topographic wetness index. This dissertation is presented as three papers and a reference manual for the software created. Because hydrologic knowledge includes both semantic aspects as well as procedural aspects, we have developed, in the first paper, a new form of reasoning engine and knowledge base that extends the general-purpose analysis and problem-solving capability of reasoning engines by incorporating procedural knowledge, represented as computer source code, into the knowledge base. The reasoning engine is able to compile the code and then, if need be, execute the procedural code as part of a query. The potential advantage to this approach is that it simplifies the description of procedural knowledge in a form that can be readily utilized by the reasoning engine to answer a query. Further, since the form of representation of the procedural knowledge is source code, the procedural knowledge has the full capabilities of the underlying language. We use the term "functional ontology" to refer to the new semantic and procedural knowledge models. The first paper applies the new knowledge model to describing and analyzing polygons. The second and third papers address the application of the new functional ontology reasoning engine and knowledge model to hydrologic applications. The second paper models concepts and procedures, including running external software, related to watershed delineation. The third paper models a project scenario that includes integrating several models. A key advance demonstrated in this paper is the use of functional ontologies to apply metamodeling concepts in a manner that both abstracts and fully utilizes computational models and data sets as part of the project modeling process.