Coastal saltmarsh distribution in Australia appears to be following a global trend of decline. In the estuaries of NSW, saltmarsh is often squeezed between landward encroachment of mangrove forest and urban/industrial development of foreshore land. Efforts to maintain and rehabilitate saltmarsh are complicated by an incomplete understanding of the hydraulic drivers for estuarine vegetation distribution. Our research is focused on the hydraulic and geomorphologic conditions required to sustain saltmarsh in a rehabilitated wetland, comprised of tidal creeks, mangrove forest, saltmarsh and tidal pools, in the Hunter estuary, NSW. The wetland is an important roost site for migratory shorebirds and is part of the Kooragang Wetlands, which are recognised as a wetland of international importance under the Ramsar Convention. The area is hydraulically complex, with a number of culverts and roads that compartmentalise flow. At a local scale (of the order of m2), vegetation morphology influences the flow field by creating drag, which acts to slow flow through friction losses. Modelling of these fine scale interactions is both numerically and theoretically demanding, requiring solution of the Reynolds-averaged Navier Stokes equations. An alternate approach is to develop a simplified hydrodynamic model of the wetland based primarily on water level. This method, which requires fewer input variables and considerably less computational effort, is appropriate for modelling wetlands where hydraulic controls (e.g. inlet configuration and internal culverts) affect the macro-scale flow field (of the order of ha) to a greater extent than local scale effects such as vegetation roughness. A hydrodynamic model of the study area is required to investigate the effects of various flow control scenarios on habitat distribution. In order to determine the most efficient modelling approach, a statistical review of the sensitivity of the flow field to vegetation type, site location, hydroperiod, elevation, tidal range and suspended particulate matter (SPM) was undertaken. This included comparison of mean velocity and vegetation community, to identify whether vegetation morphology was a significant determinant of mean velocity at the community scale; comparison of mean velocity in each vegetation community at each site, to test whether vegetation morphology was important at the site scale; comparison of mean velocity at each site with distance from the Hunter River, to test the assumption that hydraulic controls drive mean velocity to a greater extent than surface roughness in attenuated wetlands; and multi-variate analysis of hydraulic and SPM variables to identify similarities between sites. Data collection involved measurement of vegetation morphological characteristics; water level monitoring using pressure transducers; flow field measurement by acoustic Doppler velocimeters; and gravimetric analysis of suspended particulate matter. The hydraulic configuration of flow conveyance conduits, such as culverts, in estuarine wetlands was found to be critical to the distribution of the velocity flow field, tidal range, hydroperiod and SPM. Due to the low topographic relief in tidal wetlands, even relatively minor changes in hydraulic control can effect rapid and dramatic changes to vegetation distribution. In areas of tidal attenuation due to constructed flow conduits, vegetation morphology and inlet distance was found not to significantly affect mean velocity. In these areas, a simplified hydrodynamic modelling approach based on hydraulic control configuration, particularly invert level and discharge capacity, may be adopted. In areas of unattenuated flow, a more complex modelling approach is required to simulate the effect of vegetation on the flow field and sediment transport.
MODSIM 2005 Integration Challenges for Urban Systems Analysis, MODSIM05 : International Congress on Modelling and Simulation : advances and applications for management and decision making, Melbourne, 12-15 December (Melbourne, Australia 12-15 December, 2005) p. 332-338