<p>Sand dunes are critically endangered ecosystems, supporting a wide variety of specialist native flora and fauna. They have declined significantly in the past century, due to coastal development, exotic invasions, and stabilization using marram grass (Ammophilia arenaria). An increasing number of restoration groups have carried out small scale rehabilitations of using native sand binding plants spinifex (Spinifex sericeus) and pingao (Desmoschoenus spiralis). However like many other restoration ventures, efforts are not formally monitored, despite the potential for conservation of species in decline. This thesis seeks to investigate the social and ecological aspects of sand dune restoration in New Zealand. Firstly, the status of restoration in New Zealand was examined using web based surveys of dune restoration groups, identifying motivations, methods, and the use of monitoring in the restoration process. Secondly, the ecology of restored and marram dominated sand dunes was assessed. Vegetation surveys were conducted using transects of the width and length of dunes, measuring community composition. Invertebrates were caught using pitfall traps and sweep netting, sorted to order, and spiders, beetles and ants identified down to Recognizable Taxonomic Units (RTUs) or species where possible. Lizards were caught in pitfall traps, and tracking tunnels tracked the presence of small mammals in the dunes. Analysis of each variable involved the comparison of biodiversity data between restored and marram dominated dunes, at six sites across the Wellington region. The survey of dune restoration practitioners confirmed that restoration was generally based on the motivation of erosion protection and foreshore stabilization, however an increasing number of groups were interested in the conservation of flora. Conservation of fauna was a priority for only one of the respondents. Informal monitoring of restoration attempts was carried out by the majority of groups, but specific biodiversity monitoring or monitoring using systematic scientific methods was carried by only a small proportion of groups. Re-vegetation of dunes commonly used a small suite of native sand binding species mostly pingao and spinifex. Species in decline such as sand tussock (Austrofestuca littoralis) and sand daphne (Pimelia arenaria) were only planted at a small proportion of sites. Restoration of dune ecosystems has the potential to not only enhance erosion protection and sand stabilization mechanisms, but to benefit native flora and fauna endemic to sand dunes. Identifying biological change and carrying out biodiversity monitoring may be beneficial in maximizing the ecological effectiveness of restoration attempts. Marram dunes contained higher foliage cover, vegetation height and vegetation species diversity than restored dunes. Abundance and diversity of beetle, spider, and ant families were higher in marram dominated dunes. Estimated population size of common skink (O. nigraplantare polychroma) and mouse population density was also higher in marram dunes. These results were positively correlated with the percentage of vegetation foliage cover and vegetation species diversity, suggesting that the habitat conditions created by marram grass were favored by fauna. These results suggest that for maximum biodiversity gains, future dune restoration attempts should increase vegetation cover, and include a wider range of plant species. Species in decline known to be important for fauna, such as pohuehue (Muehlenbeckia spp.), sand pimelia, and sand coprosma (Coprosma acerosa) should also be included for reciprocal benefits for conservation of flora and fauna. Marram grass could also be incorporated into restoration, as its mass removal may have considerable consequences for fauna using it as a refuge, and it appears to provide desirable habitat for fauna.</p>
Sand Dune Restoration in New Zealand:
Methods, Motives, and Monitoring
