Grantee Research Project Results

2000 Progress Report: Development of a Surface Water Object-Oriented Modeling System (SWOOMS) for the Neuse River Estuary, North Carolina

EPA Grant Number: R827957
Title: Development of a Surface Water Object-Oriented Modeling System (SWOOMS) for the Neuse River Estuary, North Carolina
Investigators: Luettich Jr., Richard A. , Stotts, David , Jeffries, Harvey E. , Paerl, Hans , Bowen, J. , Alperin, Marc
Current Investigators: Luettich Jr., Richard A. , Bowen, J. , Alperin, Marc
Institution: University of North Carolina at Chapel Hill , University of North Carolina at Charlotte
EPA Project Officer: Packard, Benjamin H
Project Period: October 1, 1999 through September 30, 2002
Project Period Covered by this Report: October 1, 1999 through September 30, 2000
Project Amount: $897,859
RFA: Computing Technology for Ecosystem Modeling (1999) RFA Text |  Recipients Lists
Research Category: Environmental Statistics

Objective:

The overall objective of this research project is to develop a prototype surface water object-oriented modeling system (SWOOMS) that will allow us to: (1) model and understand the ecological response of the Neuse River estuary (NRE) to varying nutrient (primarily N) loading; and (2) integrate this estuary model within a larger suite of models that can track nutrients from their source in the airshed and watershed to their ultimate arrival and impacts within the estuary. Existing estuarine water quality models have not been designed to be part of such a modeling suite and therefore may be poorly suited for linking with dynamic models of other environmental media. SWOOMS will be developed using object-oriented (OO) design and implementation, together with collaboration with researchers developing models in other media (i.e., the watershed, groundwater, and atmosphere) to yield an integrative environmental multimedia modeling system.

Specific project objectives are to: (1) develop a finite-volume hydrodynamic and transport model component; (2) develop a water column biogeochemical model component; (3) develop a sediment diagenesis model component; (4) design and implement the model components using an OO framework; and (5) design and implement the OO surface water model for seamless input/output with OO models of the atmosphere, watershed, and groundwater media.

Progress Summary:

Development of a finite-volume hydrodynamic and transport model component. The finite volume solution method is attractive for SWOOMS because it is numerically efficient; conserves mass, momentum, and energy locally and globally; is readily extended to three dimensions using irregularly shaped (and therefore shoreline and/or bathymetry conforming) volume elements (containers); and fits a container-based, OO formalism. We are pursuing this aspect of model development along two tracks, in terms of a simple two-dimensional, laterally averaged model (see item 2) and a more complex three-dimensional model. Our initial efforts for the three-dimensional model have focused on selecting an appropriate numerical algorithm. Our literature search has turned up a new methodology, the discontinuous Galerkin finite element technique that appears to be ideally suited to our needs. This finite volume variant is based on formulating the hydrodynamic and transport equations as a system of conservation laws. The equations are integrated over a single element and unknowns approximated by two piecewise, possibly discontinuous, polynomials. Because of its local nature, the method allows for the inclusion of varying order polynomials, is locally conservative, and incorporates upwinded numerical fluxes for problems with high flow gradients. Initial applications of this technique to vertically integrated hydrodynamics and transport problems have been quite successful. We will be testing this methodology further in the coming months prior to implementation in a fully three-dimensional code. We initially will be developing this in an "object-friendly" form of FORTRAN 90.

Development of a water column biogeochemical model component. For SWOOMS model development purposes, we have been working to modify an existing FORTRAN code that is presently being applied by J. Bowen to North Carolina's NRE. The Neuse estuary eutrophication model (NEEM) is based on a customized version of the Army Corps of Engineers CE-QUAL-W2 model. NEEM uses approximately 11,000 lines of FORTRAN-77 code. We have been working to modify NEEM to make the code smaller, simpler, and more "object-friendly." This has consisted of rewriting the code so that it is consistent with FORTRAN 90 programming style and syntax and has included elimination of the many common blocks, includes, and entry points. This effort was completed recently, and the revised model is available on the Internet. Web pages also have been written to document the work and the newly revised version of the model (see http://www.coe.uncc.edu/~jdbowen/neem/w2_v313 ). In the coming months, this model will be converted to OO as described in item 4. In addition, we also are developing a simplified, laterally averaged finite volume, hydrodynamics/water quality model using both procedure-based and OO paradigms. Because the existing CE-QUAL-W2 water quality model contains over 10,000 lines of FORTRAN code, our objective in creating the simple model is to have a model of about 500 lines of code that uses the finite volume numerical solution method and simulates dissolved oxygen (DO), temperature, and biochemical oxygen demand (BOD). This code will serve as a model and benchmark for the early SWOOMS prototype. The simplified model system is a dendritic network of completely mixed reservoirs. Flow, water quality, and temperature time histories are read in from files for each of the "headwater" reservoirs. Meteorological and reservoir parameters also are read in from files. The model solves for the time varying outflow, water surface elevation, DO, BOD, and temperature for each reservoir. We have the procedure-based model working for flow, DO, and BOD, and it should shortly be functional for temperature as well. The simple model consists of approximately 600 lines of FORTRAN code. We also are working on a JAVA implementation of this simplified model.

Development of a sediment diagenesis model component. The NRE and many other southeastern and Gulf coast estuaries in the United States are shallow waterbodies that have a significant portion of their water column in close proximity to the sediment bed. Consequently, exchange processes between the water column and the sediments play critical roles in controlling water quality. This is particularly true in the Neuse, where the sediments comprise a significant source of nutrients and sink of oxygen. Consequently, we have placed considerable emphasis on the development of a model that faithfully represents these processes. Progress to date has included completion of a preliminary diagenesis model that serves as a hypothesis to be tested with detailed laboratory and field experiments. Initial microelectrode-based laboratory studies have been completed linking processes in the sediment bed and the overlying water column via transport through the very thin diffusive boundary layer that separates the two. This work has looked specifically at the behavior of species of nitrogen (e.g., nitrate, nitrite, and ammonia) and oxygen. The resulting datasets are being empirically parameterized to provide required coefficients for the diagenesis model and to provide data for verification of the model.

Design and implementation of model components using an OO framework. At this point in time, it is highly debatable whether development of OO code from the ground up is an efficient and effective means for building an environmental model. Reasons for this include: (1) the generally diminished computational performance of code written in OO languages such as JAVA or C++ as compared to procedure-based languages such as FORTRAN that have been highly optimized for number crunching; (2) the lack of OO expertise among the scientific community experienced in constructing environmental models; and (3) the substantial accumulation of procedure-based legacy code that comprises a valuable resource that should not be discarded simply based on its language choice. Consequently, we are pursuing several avenues for developing the SWOOMS model so that it functions effectively in an OO-based framework, but at the same time retains its numerical efficiencies and pedigree. Specifically, we are using three approaches to attack this problem: (1) apply a JAVA wrapper around existing FORTRAN codes so that they look like OO modules to the outside world, but actually operate in their native code; (2) partially convert existing FORTRAN code into JAVA using a structured approach so that this capability can be built into a cross-language compiler and automated in the future; and (3) fully convert existing FORTRAN code into JAVA. In each case, we end up with an OO module capable of plugging into a larger modeling framework; however, the effort and potential efficiency of each approach will be significantly different. To date, we are partially done with these conversions and should shortly begin testing the relative performance of each.

Design and implementation of an OO surface water model for seamless input/output with OO models of the atmosphere, watershed, and groundwater media. While this is the ultimate goal of all of the MIMS projects, it is clearly quite challenging because it requires close coordination among multiple research groups. To this end, we have actively participated in several MIMS workshops, including the Multi-Media Modeling Workshop, March 1999; the Cross-Discipline Ecosystem Modeling and Analysis Workshop, August 2000; and the MIMS Model Prototype Meeting, January 2001. One outcome of these workshops has been the concept that it may not be optimal for every model to be written in OO from the ground up as outlined under item 4. Rather, considerable interest has developed in using the Argonne National Laboratory's Dynamic Information Architecture System (DIAS) as a model integration system that treats individual media models as objects in a multimedia modeling hierarchy. We will continue to participate in these discussions and, as this or another model integration strategy becomes better defined, we will ensure that SWOOMS fits effectively within this system.

Future Activities:

In the next year, we will continue efforts to develop a prototype surface water object-oriented modeling system. Some specific activities are listed below.

Development of a finite-volume hydrodynamic and transport model component. We will be testing the discontinuous Galerkin finite element technique further in the coming months prior to implementation in a fully three-dimensional code. Initially, we will be developing this in an "object-friendly" form of FORTRAN 90. In the coming months, this modified NEEM model will be converted to OO as described in item 4. In addition, we also will continue to develop a simplified, laterally averaged finite volume, hydrodynamics/water quality model using both procedure-based and OO paradigms.

Design and implementation of an OO surface water model for seamless input/output with OO models of the atmosphere, watershed, and groundwater media. We will continue to participate in discussions and workshops with multiple research groups and, as this or another model integration strategy becomes better defined, we will ensure that SWOOMS fits effectively within this system.

Journal Articles:

No journal articles submitted with this report: View all 44 publications for this project

Supplemental Keywords:

multimedia, interdisciplinary, hypoxia, anoxia, water quality, finite volume., RFA, Scientific Discipline, Water, Waste, Geographic Area, Ecosystem Protection/Environmental Exposure & Risk, Nutrients, Hydrology, Contaminated Sediments, Environmental Chemistry, Chemistry, State, Microbiology, computing technology, ecosystem modeling, fate and transport, aquatic ecosystem, environmental monitoring, nutrient supply, nutrient transport, aquatic modeling, watersheds, contaminated sediment, community decision making, environmental decision making, surface water object-oriented modeling system, surface water, wetland mitigation banking program, water quality, North Carolina (NC), information technology, biogeochemistry, cross-media environmental monitoring, groundwater, stream ecosystem

Relevant Websites:

http://www.coe.uncc.edu/~jdbowen/neem/w2_v313/
http://www.marine.unc.edu/neuse
http://stottspc-cs.cs.unc.edu/

Progress and Final Reports:

Original Abstract

2001

Final Report