The Role of Post-dispersal Regeneration Processes in Beilschmiedia tawa Forest Fragments, Waikato, Northern New Zealand

Abstract

Areas of high biodiversity and endemism, such as the lowland podocarp-tawa forest type in New Zealand, are at high risk of suffering negative impacts from habitat loss and fragmentation. Shifts in fundamental ecological processes, such as autotoxicity, can exacerbate the effects of human-induced alterations. Restoration activities can help to mitigate the effects of fragmentation and simulation models allow us to improve our understanding of ecosystem dynamics, and are hence useful tools in guiding restoration initiatives. In New Zealand the removal of herbivores by fencing is a common method for restoration, but some recent studies have suggested that fencing alone is insufficient for fragment recovery. Furthermore, the native New Zealand tawa (Beilschmiedia tawa) is faced with the challenge of severe recruitment failure in forest fragments, a threat made more severe by its endemic status. The aim of the present study was to assess the effect of fragmentation and fencing on the regeneration of podocarp-tawa forests of the Waikato region, Northern New Zealand. In Chapter 2 I focused on the consequences of different types of management (fragmentation and fencing) on the current-day composition and structure of podocarp-tawa forest fragments and the associated abiotic conditions (soil, light and climate). In Chapter 3 I explored the role of fragmentation and fencing on the survivorship of B. tawa seedlings and the factors that could explain the recruitment problems that affects it. Later in Chapter 4 I tested the potential allelopathic effect of B. tawa leaf leachate on germination and growth of B. tawa. Finally, in Chapter 5 I gave a detailed description of the development of a spatially explicit individually-based model (SEIBM) built to represent the long-term dynamics and viability of podocarp-tawa forest and subsequently in Chapter 6 I used the aforementioned model to simulate different scenarios to assess the impact of different management and restoration strategies on the long-term persistence of podocarp-tawa forest fragments. Results from Chapter 2 showed that in fragmented sites growth rates of seedlings was reduced. Furthermore, our results show that fencing alone is not enough to ensure the viability of forests fragments over time, but it is an important first step. Soil analyses revealed higher water and nutrient content in unfragmented forests versus forest fragments, regardless of whether they were fenced or not. Furthermore, results from Chapter 3 demonstrated that fencing had no positive effect on seedling growth and survivorship. In addition, in Chapter 4 I found that B. tawa leaf leachate can have a detrimental effect on its own seeds and on the early development stages of its seedlings. The spatial model described in Chapter 5 showed to perform within plausible ranges. The most sensitive parameters were in general the ones that represented anthropogenic activities and life cycle stages. The model described in Chapter 5 allowed us to simulate various future scenarios. These future scenarios simulations are compiled in Chapter 6 were results demonstrated that fenced forest fragments inevitably collapse over time in the absence of other restoration actions. While fencing to remove pressure from large herbivores is a good first step, further restoration actions, such as periodic planting, need to be implemented to secure the persistence of forest fragments.