|
|
|
Within this section:
 Introduction
Some recent papers
Biodiversity
Conservation
Phylogenetic
inference
Biological
monitoring
The PNG papers
Some older papers
|
|
|
Home Research and CollectionsBiodiversity & Systematics
A Biodiversity Conservation Plan for Papua New Guinea Based on Biodiversity Trade-offs Analysis
Masks
Masks are characteristics which can be used to exclude RMUs from consideration for membership of the BPA set. There were two such characteristics used for these analyses; land use intensity, and RMUs less than 10km2 in area. In addition, any biodiversity attributes with an extent of less than 1km2 in any RMU were not counted as occurring in that RMU. Each of these masks represented an attempt to better ensure the viability/persistence of biodiversity within proposed protected areas (see also Faith et al. 2001b).
Land use intensity
RMUs which have more than half of their area in PNGRIS land use intensity classes 0 - 4 (i.e. very high with tree crops, very high, high, moderate and low) were masked out of the initial analysis. A total of 954 RMUs have more half their area falling into these land use intensity classes. It was found, however, that our biodiversity target level of representation could not be reached when these areas were masked out. That is, some biodiversity attributes only occur in intensively used areas. The strategy in this case was to first run an analysis with these RMUs masked. At the completion of that analysis, no more RMUs could be added to the set, but the biodiversity target was not achieved. The set of RMUs found this way was saved and a new analysis was commenced with this set as a starting point, and with intensively used RMUs now available for selection. When this was done, the target could be achieved with the addition of a small number of extra RMUs, as shown in the results section below.
Figure 1d is a map of RMUs with current high land use intensity.
Small RMUs
It was decided that RMUs less than 10 sq. km. in area were too small to be candidates for biodiversity priority areas because they were unlikely to be large enough to retain many of their species in the long term. They were excluded therefore from consideration. This is an arbitrary cut-off and many RMUs larger than this are probably still too small to form viable priority areas over time. The most practical way of managing smaller priority areas is to group them into larger management units such as Wildlife Management Areas (WMAs). This would seem to be a workable solution, for example, where there are clusters of small priority areas in highland regions. Of course, there would be costs associated with increasing the total area in this way, which have not been included, and therefore not traded-off, in the analysis used to select the current priority area set.
Small coverage by attributes
Some attributes are only represented by very small areas within RMUs. In many cases this may simply be the result of overlaying attributes on map units where boundaries do not properly coincide. Small slivers of attributes may thus occur within RMUs to which they may not properly belong. In order to minimise this problem, any attribute occurrence of less than 1km2 within an RMU was ignored in the analyses.
Top
|
|
|
