Abstract
Since 2002 Ammophila arenaria has been progressively eradicated from the transgressive dune system at Mason Bay, Stewart Island, New Zealand in a bid to restore dune dynamics. Little is known, however, about the impact of restored dune dynamics on downwind landforms and associated plant communities. With the aim of predicting the response of the Mason Bay stonefield to dynamic restoration, this study examines the patterns of sand transport and accumulation downwind of the dynamic restoration project and considers the implications for plant communities and key species. The Mason Bay stonefield is a deflation surface which is recognised as nationally threatened as it is inhabited by at-risk native plants and is an important habitat for the endangered Dotterel (Caradrius o. obscurus). Sand accumulation in the stonefield may cause a shift from a deflation surface to other dune forms, including nabkha, given the presence of indigenous sand-colonising species.
The methodology of this study reflected the spatial and temporal scales at which A. arenaria invasion and dynamic restoration could impact the stonefield. First the historic development of the Mason Bay landforms was described in relation to A. arenaria invasion and its subsequent removal using a series of historic photographs. Since A. arenaria invasion in 1958, the area of the stonefield has significantly increased by 39% and moved inland. Since A. arenaria removal in 2002, the stonefield has increased in area by 7%. This was attributed to the remnant A. arenaria rhizome and dead plant material creating a lag in geomorphic response to devegetation.
The current sand accumulation in the stonefield was examined over a nine month period, using a series of erosion pins within a 200m x 50m plot in the stonefield. Digital elevation models were derived from regular total station surveys to determine whether sand-drift was accumulating around low Ficinia spiralis nabkha. During the nine month survey period the surface of the study area accreted on average only 3.22mm. Accretion and erosion was not strongly correlated with vegetation cover. The dimensions of the surveyed nabkha did not change significantly during this period. The lack of sand deposition downwind of eroding dunes is attributed to the topography and exposure of the bed.
The importance of the event-scale sedimentation patterns was investigated by measuring the wind speed, direction and sand transport, to determine whether such events deposit or erode sand in the study area. The wind speed and direction measured during two discrete wind events showed that there is no decline in wind speed across the study area, creating little potential for aeolian deposition in the stonefield.
Lastly, the impact of the observed sand burial on the sand binding and non-sand binding stonefield plant communities was assessed. This investigation suggested that at risk non-sand binding plant communities might have a degree of tolerance to sand burial. Increased sand deposition, however, may favor the sand binding plant species and exclude the native non-sand binders like Raoulia hookeri var. hookeri.
The geography of the stonefield has shown to be remarkably dynamic and the ability of plant species to keep pace is evidence of their ability to colonise. Since 2011, approximately five years after A. arenaria spray efforts began the depositional lobes have started to elongate into the stonefield. This suggests that the remnant A. arenaria rhizome is breaking down and sand inputs in the stonefield may be increasing. Sand deposition around F. spiralis plants in the stonefield and the elongation of the depositional lobes may reduce the stonefield area and break up the continuous stonefield feature. To date, the stonefield has been a continuous feature which has most likely aided in the stonefield species’ ability to keep pace with the evolution of the stonefield. However, it is unknown whether the colonisation abilities of these species will be able to adapt to the fragmentation of the stonefield habitat by sand burial from recent destabilisation.