Ecological restoration and management of the Linwood paddocks
Thesis DisciplineEnvironmental Sciences
Degree GrantorUniversity of Canterbury
Degree NameMaster of Science
This study investigates salt marsh restoration within the Canterbury region as a way to remedy the current loss of these vital systems. Previous attempts have concentrated largely on aesthetic appeal and lack detailed scientific basis or understanding. The proposed Linwood Paddock restoration site (historically a brackish wetland) provides an excellent opportunity for scientifically-based salt marsh restoration. Restoration should reestablish functions and attributes symptomatic of a 'healthy' wetland system, from the presently imposed pastoral system. The historic and current salt marsh composition and relation to environmental conditions are described for the Canterbury region and the long-term impacts of implementation and management decisions in restoration are evaluated. The likely success of plant growth on heavy metal and nutrient-enriched Linwood soils and the potential for phytoremediation to improve wastewater quality are also determined. Quantitative survey's of the fragmented Avon, Heathcote, Brooklands and Saltwater Creek salt marshes, provided a complete revegetation template, which details the optimal range for each native species with respect to elevation and salinity. The marked zonation appeared to be the result of both competitive displacement and interspecific variation in physiological tolerance. Lower marsh elevations were dominated by Leptocarpus similis and upper marsh elevations by Sarcocornia quinqueflora. All other species' requirements overlap, but they would be expected to exist in a mid tidal mosaic "patchwork" fashion in revegetation, since the clonal nature of these species means that patches will be mono specific, rather than mixed-species patches. A mesocosm experiment was performed in an area of the Heathcote marsh by planting Leptocarpus similis, Juncus maritimus and Schoenoplectus pungens into transferred Linwood Paddock soil. Plant growth was viable in Linwood Paddock topsoil and plants sourced from natural stocks had significantly higher survival and biomass than nursery stock (p < 0.05). In addition, there was a significant difference (p < 0.05) in survival and biomass between the two species remaining at the end of the experiment, with Leptocarpus similis having increased survival and greater biomass than Juncus maritimus. Assessments of previous local revegetation attempts at the margins of Oxidation Pond Nos. 5 and 6, the Charlesworth Street Reserve and the Devil's Elbow bankworks showed that salt marsh herbs (Sarcocornia quinqueflora in particular) were more likely to self-colonise a restoration site than rushes or sedges. The most obvious impediments to revegetation success were the use of a too coarse substrate and insufficient tidal flushing through lack of maintenance. It was concluded that adequate monitoring and record-keeping are necessary to determine strategies for wetland implementation and to understand the long-term impacts of wetland management decisions from such attempts. The scientific research was integrated in the design of an ecological salt marsh restoration and management plan for the Linwood Paddocks, which is not only representative of Canterbury salt marshes, but also self-sustaining and accessible to the general public. Creation of a freshwater wetland within the existing Bromley Oxidation Ponds is presented as a unique opportunity to remediate wastewater and to supply high quality freshwater to the salt marsh restoration, thereby enhancing estuary wildlife and water quality values simultaneously. The freshwater species recommended for phytoremediation are Typha and Scirpus. This design is scientifically based and its principles could be used (with the application of local data) in further restoration projects within New Zealand or elsewhere in the temperate world. As salt marshes decline, this approach is timely to ensure that restoration is successful and not a technique that is relearned each time.