Effects of landscape heterogeneity and clearfell harvest size on beetle (Coleoptera) biodiversity in plantation forests

摘要

Compared to natural forests, fast-growing plantations of exotic species such as Pinus radiata are often perceived as marginal habitat or unsuitable habitat for most native species. By studying Coleoptera (beetles) in a variety of landscape elements (pasture, native forest and different aged Pinus radiata stands) in a highly modified and fragmented landscape in New Zealand I aimed to determine the value of exotic plantation forests for native biodiversity, and how these species are affected by different sized clearfell harvest areas. Pitfall trap sampling of beetles showed that plantation forest stands can provide suitable complimentary habitat to native forest for many species. Rarefied species richness of Carabidae, Scarabaeidae and Scolytinae was not significantly different between habitats, however, habitat types differed significantly in their beetle community composition. Comparing different production habitats, Pinus radiata stands had a beetle community composition most similar to native forest. However, a small minority of species, e.g., Dichrochile maura, were restricted to native forest habitat highlighting the importance of retaining indigenous ecosystems within plantations. Unlike human modified habitats, native forests did not provide suitable habitat for exotic species. Clearfell harvesting is controversial and its impact on biodiversity is a key constraint for many forest certification programs, such as that administered by the Forest Stewardship Council (FSC). Despite this, no replicated manipulative experimental studies of the impact of different sized clearfell harvest areas on biodiversity have been undertaken at scales relevant to the New Zealand forest industry. One potential model of the impact of different clearfell harvest sizes is the concept of a threshold size. A threshold scenario may occur where clearfell harvest impacts increase at a rate disproportionate to the change in clearfell size over a small range of harvest areas, but impacts remain relatively unchanged either side of the threshold zone. I sampled Coleoptera in experimentally created 0.01, 0.05, 0.5, 5.0, 50 and 500 ha clearfells within Pinus radiata plantations in the central North Island of New Zealand. The wide range of clearfell harvest sizes, including some very small areas, such as 0.01 ha was instigated in an attempt to document potential clearfell harvest size thresholds. Rarefied native beetle species richness was higher in harvest areas compared to adjacent mature plantation stands. The beetle species richness in 5 ha and 500 ha harvest areas was significantly greater species than that in small 0.01 - 0.5 ha harvest areas. Although, the high beetle diversity recorded in 500 ha clearfells should be treated with caution due to confounding spatial autocorrelation. The degree of change in beetle community composition increased with increasing clearfell harvest area. Beetle assemblages in large harvest areas were less similar to their paired adjacent mature forest than smaller harvest areas. Although, constrained multivariate ordination techniques did show a short-term change in beetle species composition between recently clearfelled harvest areas of as little as 0.05 ha and adjacent mature P. radiata stands. The colonisation by open-habitat disturbance-adapted species was a key driver of this change, some species dispersed into clearfelled stands in significant densities within days post-harvest. Overall, there were no distinct short-term trends to the change in species richness as a function of increasing harvest area that would suggest an ecological impact threshold response. If short-term outcomes of clearfell harvesting are ameliorated by successful recolonisation, the long-term spatial arrangement of different aged stands becomes more important for the maintenance of biodiversity at the landscape level than short-term consequences of harvesting. By sampling selected beetle taxa in 1, 2, 4, 8, 16 and 26 year-old stands, I found that the abundance of seven out of eight of the species selected for analysis recovered to levels similar to those in adjacent mature forest within the timeframe of a single harvest rotation. Individual species utilised different aged stands, indicating different life-history strategies. For example, open-habitat, disturbance-adapted species such as Cicindela tuberculata and Sitona discoideus were prominent in young stands, and forest species such as Pycnomerus sophorae and Paracatops phyllobius were highly abundant in older stands. These alternative life-history strategies highlight the benefits of maintaining a mixture of different aged stands to increase biodiversity at the landscape level. This thesis fills an important gap in our knowledge of biodiversity in production landscapes. I show that plantation forests have value as complimentary habitat to native forest and they make an important contribution to the maintenance of biodiversity at the landscape level. Although clearfell harvesting is a severe disturbance to the forest ecosystem, the long-term recovery of beetle populations suggests that harvesting is not the key limiting factor to the enhancement of biodiversity in the plantation forests studied. This unusual situation is possibly the result of prior land-use history, as many plantations were established on degraded pastoral land, and harvest-sensitive species are unlikely to have survived this initial land-use change. As such, the severity of the long-term impacts of clearfell harvesting on biodiversity are likely to be context specific and will vary accordingly. The importance of spatial heterogeneity of habitat elements, including different aged plantation stands and native forest remnants, needs to be investigated in more detail to determine what limits biodiversity in this plantation landscape. Key points to consider are the proximity to, and proportion of, native forest cover in the landscape and the degree of connectivity among native remnants. It is these landscape-level attributes that may determine biodiversity at a regional scale, and more emphasis should be placed on landscape scale factors and there interaction with stand specific forest management practices. For example, the spatial mosaic of harvesting areas may need to be of a finer-scale when there are fewer native remnants within the landscape.