Modelling Evaluation of Dam Removal in the Context of River Ecosystem Restoration

Abstract

Applications of environmental models may provide imperative information to enable informed decision‐making of river management actions, which are often made in the face of high system complexity and uncertainty. We applied Hydrologic Engineering Centers River Analysis System(HEC‐RAS) and Curvilinear Hydrodynamics Three‐Dimensional (CH3D) models to aid in the decision‐making of the proposed removal of the Masten Dam, a small, ‘run‐of‐the‐river’ dam on the Loxahatchee River, a federally designated ‘Wild and Scenic River’ in south‐east coast of Florida (USA). Anthropogenic alteration of the system has led to changing hydroperiods and salinity regimes in the floodplain. Both models are calibrated against measured data taken at varying temporal and spatial scales. The HEC‐RAS modelling results show that removal of the Masten Dam would lower water levels in the upstream riverine reach, leading to reduced soil moisture or inundation in the floodplain. The CH3D modelling results indicate that dam removal would increase river salinity during the dry season in the tidal reach where salinity compliance for environmental flow regulation is measured. These environmental changes would exert additional stress on freshwater vegetation communities in the floodplain. Given the scarcity of water resources in the region, removal of the Masten Dam would not offer an effective restoration strategy. This study demonstrates not only the need for evaluation of dam removal on a case‐by‐case basis but also the usefulness of environmental models in providing the technical basis for such management decisions. Copyright © 2014 John Wiley & Sons, Ltd.

Elucidating multidecadal saltwater intrusion and vegetation dynamics in a coastal floodplain with artificial neural networks and aerial photography

“…The performance of ANNs was comparable or superior to that of physical hydrodynamic models of the Loxahatchee River reported by Wan and Hu () and Wan et al . (). For example, the r 2 and RRE values for model calibration of the Loxahatchee River 3‐D hydrodynamic model, using measured data from January 2003 to April 2004 at the same locations with this study, were 0·72–0·88 and 12·2–15·8%, respectively (Wan et al ., ).…”

“…(). For example, the r 2 and RRE values for model calibration of the Loxahatchee River 3‐D hydrodynamic model, using measured data from January 2003 to April 2004 at the same locations with this study, were 0·72–0·88 and 12·2–15·8%, respectively (Wan et al ., ). The RRE for model calibration of the 2‐D hydrodynamic model using measured data from 1 May to 12 August 2003 was 6·65% for RK 13·1 and 10·76% for RK 9·5 (Wan and Hu, ).…”

“…These parameters are widely used in the calibration and validation of ANNs and estuarine hydrodynamic models (e.g. Dawson and Wilby, 2001;Ji, 2008;Wan et al, 2014). Graphic comparisons between observed and simulated salinities are presented in Figure 3 for visual inspection of the goodness-of-fit.…”

“…There exists a rich literature in estuarine salt transport and associated biophysical changes in floodplains such as the geomorphic evolution of tidal creeks (Knighton et al, 1992), the influence of sea level rise on saltwater intrusion (Hong and Shen, 2012), and the hydrologic and morphologic controls of salt transport in estuaries (Wang, 1988;Savenije, 2012). An important tool aiding in this research is process-based hydrodynamic models that can be used to simulate estuarine salinity under varying physical conditions (Wang, 1988;Ji, 2008;Wan et al, 2014).…”

“…Changes of vegetation community especially canopy species often take a time scale of decades. However, salinity simulations are typically conducted with estuarine hydrodynamic models in a fine time step of minutes with a simulation period of several years (Wang, 1988;Wan et al, 2014). Long-term salinity simulations (e.g.…”