Australasian Plant Conservation
Originally published in Australasian Plant Conservation 20(2) September - November 2011, p 9-10
The rehabilitation of coastal wetlands: why small-scale variations in microtopography are critical to success
Paul I Boon
Institute for Sustainability and Innovation, Victoria University. Email: email@example.com
Mounds of dead Common Reed in Dowd Morass at low water level, showing the complex variations they create in sediment microtopography. Photo: Paul Boon.
One of the large agricultural pumps used to drain Dowd Morass. Photo: Paul Boon.
Coastal wetlands are potentially subject to a range of threats and stressors; many suffer from altered water regimes and the loss of native vegetation and are in need of rehabilitation.
Since 2003 we have been running a large, multidisciplinary project to rehabilitate the brackish-water wetlands that fringe the Gippsland Lakes in south-eastern Australia. These wetlands, although recognized as regionally, nationally and internationally important and listed under the Ramsar Convention, are threatened by a wide range of environmental factors, including highly unnatural water regimes, secondary salinization, nutrient enrichment, and the presence of acid sulfate soils (Boon et al. 2007).
The focus of our studies has been Dowd Morass, a large (1,500 ha) brackish water wetland near the confluence of the Latrobe River and Lake Wellington, at the western end of the Gippsland Lakes complex. The dominant vegetation is Swamp Paperbark, Melaleuca ericifolia, which covers about two thirds of the vegetated area of the wetland. Large swards of Common Reed Phragmites australis also occur but their extent has declined over recent decades (Boon et al. 2008).
Dowd Morass suffers from chronic inundation and has been kept artificially inundated since the mid 1970s, in order to provide for duck-hunting opportunities, facilitate breeding by the colonial nesting birds, and limit intrusions of saline water from the Gippsland Lakes (which are connected to the Southern Ocean by a permanent entrance at the township of Lakes Entrance). The result has been water levels that are unnaturally high and fluctuate over only a small range, and so the periodic drying events that would have occurred in the past either no longer occur or are exceptionally rare. Accordingly, the plant communities are floristically depauperate and mostly in poor condition. The application of a drawdown phase, where water levels are reduced to expose sediments and then allowed to fluctuate more naturally, is a management tool often thought to be suitable for rehabilitation and rejuvenation of chronically inundated wetlands.
Rehabilitation of Dowd Morass
The research project had two main components: i) to determine the hydrological requirements of the main plant species present, particularly Swamp Paperbark; and ii) to trial different strategies to rehabilitate the wetland, centred on a whole-of-wetland hydrological intervention to see whether the introduction of a more natural wetting and drying cycle would improve vegetation condition. When the hydrological manipulations commenced in 2003, the wetland’s vegetation was in poor condition: adult plants were dying; there was little sexual recruitment of juveniles into the population; and the understorey was floristically very poor (usually fewer than two or three species).
Dowd Morass was suitable for hydrological manipulations because in the 1970s it had been divided into two regions by a substantial north-south levee. The levee allowed us to drain the westerly part (~500 ha) and to maintain water levels in the easterly part (~1,000 ha) in a modified Before-After/Control-Impact (BACI) experimental design with impact, control and reference sites. Water levels were controlled with two large agricultural pumps, which pumped water from the drawdown side into the Latrobe River and across the levee into the control site, supplemented by opening or closing the gates on a large regulator that linked the wetland with the river.
Importance of microtopographic relief in establishing different water regimes
We assumed that a single water regime could be applied and that it could be modified at a whole-of-wetland scale by pumping and/or regulator control. However, we were unable to maintain a single consistent water regime due to prolonged drought, combined with three severe storms that flooded the site, and vandalism of the regulator used to control the influx and efflux of water. Instead, we used hydrological classifications to determine whether different water regimes operated at smaller spatial scales. To support the hydrological analyses, we had detailed, long term (4 year) data sets on water levels and floristics from forty‑five 50 m transects across the wetland. The hydrological classification, prior to hydrological manipulation, showed that, despite the wetland being nominally permanently inundated, subtle variations in microtopography created four distinct water regimes and this hydrological heterogeneity (variability) subtly controlled vegetation patterning (Raulings et al. 2010). Causes of microtopographic variation included hummocks created by dead clumps of Phragmites australis and other types of grassy wetland vegetation (e.g. Paspalum spp.), and raised areas around the emergent trunks of Swamp Paperbarks.
We re-analysed the hydrological data to see whether the original small-scale water regimes had been maintained after the hydrological interventions (Raulings et al. 2011). The interventions had simplified the range of hydrological regimes that occurred in the wetland, and there were now three distinct water regimes: i) areas that continued to be permanently inundated with deep water; ii) areas that were flooded intermittently but only with shallow water; and iii) areas that had almost completely drained. Floristic richness of the understorey improved in the drained sites with 12–15 understorey species per 50 m transect, compared with <5 (often only 1–2) in permanently flooded sites.
We concluded that the capacity of wetland vegetation in Dowd Morass to respond to a reinstatement of drying following chronic inundation was constrained by abiotic (e.g. salinity and sediment pH) and biotic (e.g. depauperate seed banks) factors. The whole-of-wetland manipulations that we tried were complex, expensive and risky. Moreover, such interventions may not always improve vegetation condition, at least in the short term. Our trials showed that rehabilitation was most successful in areas that had been shallowly flooded prior to drawdown and that remained dry for longest.
Studies across the world show that microtopographic relief plays a major role in wetland ecology and rehabilitation (Larkin et al. 2006; Raulings et al. 2010, 2011). Topographic microheterogeneity is likely to be critical for the health of coastal paperbark-dominated wetlands in south-eastern Australia, given that Melaleuca spp tend to dominate poorly drained ground of low fertility: under these conditions, the drier, less saline and nutrient-enriched hummocks they create are essential precursors to environmental heterogeneity in the wetland. It is likely, therefore, that the success of future attempts to re-instate more natural wetting and drying regimes would be improved by considering the wetlands not as homogeneous (uniform) basins with a single water regime, but instead as a complex mosaic of subtly different water regimes and vegetation patterns. With clever control of water levels, it may be possible to take advantage of such microtopographical heterogeneity to increase floristic richness in the understorey and improve the condition and recruitment success of the overstorey in tree or shrub dominated wetlands.
Field and project staff: Dr Kay Morris, Dr Elisa Raulings, Michael Roache, Dr Matthew Hatton, Dr Randall Robinson and Dr Jacqui Salter. The project was funded by a number of partners, including Land & Water Australia, the Murray-Darling Basin Commission, West Gippsland Catchment Management Authority, Department of Sustainability and Environment, Parks Victoria, Gippsland Coastal Board, Field and Game Victoria, Field Naturalists Club of Victoria, Gippsland Water, Esso, and BHP Billiton.
Boon, P.I., Raulings, E., Morris, K., Roache, M., Robinson, R., Hatton, M. and Salter, J. (2007). Ecology and management of the Lake Wellington wetlands, Gippsland Lakes: a report on the R&D project, 2003-2006. Land & Water Australia, Canberra.
Boon, P.I., Raulings, E., Roache, M. and Morris, K. (2008). Vegetation changes over a four-decade period in Dowd Morass, a brackish-water wetland of the Gippsland Lakes, south-eastern Australia. Proceedings of the Royal Society of Victoria 120: 403-418.
Larkin, D., Vivian-Smith, G. and Zedler, J.B. (2006). Topographic heterogeneity theory and ecological restoration. In Foundations of Restoration Ecology (Edited by Falk, D.A., Palmer, M.A. and Zedler, J.B.). Island Press, Washington.
Raulings, E., Morris, K., Roache, M. and Boon, P.I. (2010). The importance of water regimes operating at small spatial scales for the diversity and structure of wetland vegetation. Freshwater Biology 55: 701-715.
Raulings, E., Morris, K., Roache, M. and Boon, P.I. (2011). Is hydrological manipulation an effective management tool for rehabilitating chronically flooded, brackish-water wetlands? Freshwater Biology (in press)