Home Archived October 29, 2018
A.R.M. Loxahatchee National Wildlife Refuge
Enhanced Water Quality Monitoring and
photo collage showing a butterfly, flowers, water, and a black bird

reports > spatially explicit modeling of hydrodynamics and constituent transport within the a.r.m. loxahatchee national wildlife refuge report

Water Quality Monitoring Network
Water Quality Modeling
Canal Sediment and Water Quality Characterization
.: Reports :.
Contact Us
Photo of A.R.M. Loxahatchee National Wildlife Refuge swamp
Report No. LOXA10-003

Spatially Explicit Modeling of Hydrodynamics and Constituent Transport within the A.R.M. Loxahatchee National Wildlife Refuge, Northern Everglades, Florida

Chunfang Chen
Ehab Meselhe
Michael Waldon
Alonso Griborio
Hongqing Wang
Matthew C. Harwell

Prepared for the US Fish and Wildlife Service, Department of Interior

Institute of Coastal Ecology and Engineering
University of Louisiana-Lafayette

September 2010


The Arthur R. Marshall Loxahatchee National Wildlife Refuge (Refuge) is a 58,725 ha remnant of the Northern Everglades. Changes in water quantity, timing and quality have negatively impacted the Refuge. Therefore, a priority for the Refuge is to develop water quantity and quality models to identify appropriate water management strategies that will minimize negative impacts and protect fish and wildlife, while meeting flood control and water supply uses. Modeling identifies data gaps, improves understanding of impacts, and quantifies comparisons of management alternatives.

This report focuses on the development and application of a spatially explicit hydrodynamic and constituent transport surface water model for the Refuge. The spatially explicit MIKE FLOOD and ECO Lab (DHI) modeling frameworks were used to simulate the hydrodynamics and chloride (CL) transport within the Refuge. This MIKE FLOOD model dynamically links a one-dimensional model of the 100km perimeter canal with a 400m uniform grid of over 3600 two-dimensional marsh model cells, and allows for exchange of water and constituents between the two systems. Constituent transport is driven by modeled water flows and dispersion, as constituent concentrations are transformed through reactive and settling processes modeled within the ECO Lab framework. The model was calibrated for a 5-year period (2000-2004), and validated for a 2-year period (2005-2006). The graphical and statistical comparisons of stage, water depth, discharge and concentration demonstrate the applicability of this model for temporal and spatial prediction of water levels, discharge and water quality concentrations, and also demonstrate that MIKE FLOOD is a feasible alternative for modeling large wetlands that are flooded by overbank flow. The model quantifies the importance of mechanisms across the Refuge linking wetland concentration to inflow concentration and volume. Two example model applications presented here illustrate the model's ability to provide quantitative information for decision support.

map showing boundaries of the Loxahatchee Refuge
Figure 1. Boundaries of the Loxahatchee Refuge. Adapted from USFWS (2000). [larger image]

1. Introduction

The Arthur R. Marshall Loxahatchee National Wildlife Refuge (Refuge) overlays Water Conservation Area 1 (WCA-1), and is managed by the United States Fish and Wildlife Service (USFWS). WCA-1 is a 58,725 ha remnant of the Northern Everglades in Palm Beach County, Florida (USFWS, 2000). Wetland loss and degradation has taken place in the Everglades, with much of this deleterious impact associated with hydrological changes (Thompson et al., 2004). The U.S. Fish and Wildlife Service (USFWS) recognized that there have been changes to the Refuge's water quantity, timing, and quality which have caused negative impacts to the Refuge's ecosystem. The Refuge is impacted by changes in water flow and stage (Brandt et al., 2000; USFWS, 2000; Brandt, 2006), excessive nutrient loading (Newman et al., 1997; USFWS, 2000), and altered dissolved mineral concentrations including chloride (Swift, 1981; Swift, 1984; Swift and Nicholas, 1987; Browder et al., 1991; Browder et al., 1994; McCormick and Crawford, 2006). According to the USFWS (2000), changes in hydroperiod and water depth patterns affect wading bird feeding patterns, apple snail reproductive output, bird and alligator nesting, and also alter the distribution of aquatic vegetation and tree islands. In addition, high concentration of nutrients in runoff causes proliferation of cattails, and other undesirable species that negatively affect the ecosystem's balance (USFWS, 2000). It is important to manage water for the benefit of fish and wildlife in the Refuge. Refuge objectives are to minimize nutrient impacts, while meeting ecosystem, flood protection, and water supply needs.

The ability to predict the effects of manipulation of water operations upon wetlands is central to the success of wetland management and restoration (Gilvear and Bradley, 2000; Hollis and Thompson, 1998). Hydrodynamic and water quality models provide the predictive tool needed for management and scientific support. A calibrated hydrodynamic and water quality model provides such information as movement of water, fate and transport of constituents, and water quality management (Kadlec and Hammer, 1988; Tsanis et al., 1998; Koskiaho, 2003). Models form the basis of information for questions regarding the hydrologic, hydrodynamic, and water quality conditions occurring under present conditions and management rules, and project how these processes would be altered by different structural changes and management scenarios.

The complexity and spatial resolution required of a model are dependent on the specific hydrological and ecological system under study, and the nature of the questions being addressed. Two types of models have been used in hydrological modeling, namely the compartment-based model (or box model), and the distributed model. The compartment-based model, which conceptualizes the system as spatially averaged compartments, has been widely used in assessing the environmental fate of chemicals (e.g., Mackay et al., 1992). It has also been applied to the Refuge modeling (Arceneaux et al., 2007; Wang et al., 2008; Wang et al., 2009; Roth et al., 2009). The most attractive feature of these models is their simplicity. The models are capable of quickly examining a broad range of alternative scenarios, which is a distinctive advantage for decision makers. However, such models are often inappropriate for site-specific applications or where detailed spatial visualization is needed. Therefore, efforts reported here have been directed towards development of distributed physically based models (Martin and Reddy, 1991; Alvord and Kadlec, 1996).

In modeling the hydrodynamics and water quality of the Refuge, a spatially explicit model was developed based on MIKE FLOOD (DHI, 2008) to provide a detailed quantitative framework to address the management tasks. This report focuses on the model development, calibration, validation, and its application to two management scenarios.

Download PDF of entire report (1.3 MB)
(You will need the free Adobe Acrobat Reader in order to view the file.)

Table of Contents:

  • Acknowledgements
  • Abstract
  • 1. Introduction
  • 2. Study Area
    • 2.1 Site description
    • 2.2 Regulation Schedule
    • 2.3 Field data
  • 3. Methods
    • 3.1 MIKE FLOOD
    • 3.2 Hydrodynamic Model
    • 3.3 Advection-Dispersion (AD) Model
  • 4. Model Setup
    • 4.1 MIKE 21
    • 4.2 MIKE 11
    • 4.3 Coupled Model
  • 5. Calibration and Validation
  • 6. Results
    • 6.1 Stage and Depth
    • 6.2 Discharge
    • 6.3 Chloride (Cl) Concentration
    • 6.4 Management Scenarios
  • 7. Discussion
  • 8. Conclusions
  • References


  • Figure 1. Boundaries of the Loxahatchee Refuge. Adapted from USFWS (2000).
  • Figure 2. Topography of the Refuge (in feet NGVD 1929) based on USGS published elevations. The site of the S-5A pump station is shown in this figure. Desmond (2003).
  • Figure 3. Thalweg profiles for the sediment surface elevation and channel bottom elevation for the eastern canal (L-40) (left) and the western canals (L-7 and L-39) (right) (Meselhe et al., 2005).
  • Figure 4. Location of hydraulic structures located in the Refuge.
  • Figure 5. Water Regulation Schedule for WCA-1. Adapted from USFWS (2000).
  • Figure 6. Location of rain gages, ET site, and water level stations in the Refuge.
  • Figure 7. Water quality monitoring sites located in the Refuge. Enhanced sites, labeled LOXA elsewhere, are labeled A here for brevity).
  • Figure 8. Discharge and stage difference relation based on historic data of S10 outflows (1/1/1995-8/31/2007).
  • Figure 9. Comparisons of modeled and observed water level at marsh (USGS) and canal stations.
  • Figure 10. Comparisons of simulated and observed water depth at DCS stations.
  • Figure 11. Comparisons of simulated and measured annual discharge at the outflow structures individual and combined.
  • Figure 12. Comparison of Chloride concentration with measured data (including time series, scattered plot, and percentage exceedance plot) at stations of EVPA (a-c), XYZ (d-e), enhanced (f-g), and the canal station (h-j).
  • Figure 13. Erosion Chloride concentration of measured and extracted profiles along the X and Z transects with a two-week window before and after the measurement (a) event of 9/20/2000 (window 9/6/2000-10/4/2000) (b) event of 10/15/2002 (window 10/2/2002-10/29/2002).
  • Figure 14. Comparison of Chloride concentration with original boundary inflow concentration (Base) and reduced concentration (2 mg/L, same as rainfall concentration) at marsh stations (a-b) and canal station (c).
  • Figure 15. Comparison of stage (a-b) and Chloride concentration (c-g) without berm (Base) and with berm.


  • Table 1. Manning's n for different vegetation classes
  • Table 2. Calibrated hydrodynamic and chloride model parameters
  • Table 3. Calibration statistics of water level (2000-2004)
  • Table 4. Validation statistics of water level (2005-2006)
  • Table 5. Statistics of annual discharge for the calibration and validation period (2000-2006)
  • Table 6. Statistics of chloride concentration at selected stations for the calibration period (2000-2004)
  • Table 7. Statistics of chloride concentration at selected stations for the validation period (2005-2006)

< Back to Reports

[ Disclaimer | Privacy Statement | Accessibility ]

Technical support for this Web site is provided by the U.S. Geological Survey
This page is: http://sofia.usgs.gov/lox_monitor_model/reports/LOXA10_003_Waldon2011MIKEFLOOD.html
Page last updated: 09 April, 2015 @ 09:52 PM
Comments about website: Heather Henkel (hhenkel@usgs.gov)