Biotropica 46(3) Editor’s Choice: Projecting the impact of management decisions in Kenya
The May 2014 Editor’s Choice Article is “Linking land-use scenarios, remote sensing and monitoring to project impact of management decisions“, by Nina Farwig, Tobias Lung, Gertrud Schaab and Katrin Böhning-Gaese. Below the study’s lead author — who is also one of our exceptional Subject Editors — describes the motivations behind this project and what went into the work. Congratulations Dr. Farwig and colleagues!
It is well documented that tropical forests are declining across the globe. This is worrying as they provide not only more than 50% of the terrestrial biodiversity but also essential ecosystem goods and services for human well-being. Growing human populations and accelerating demand for forest resources such as fuel wood and timber resulted in increasing areas of degraded forests and forest plantations. It has been shown that even these degraded and modified forests can be crucial for conservation as they often retain many forest species. However, the community structure and composition can show marked differences. Recently, the surrounding landscape has been identified as important factor for the proportion of persisting biodiversity in modern landscapes. Consequently, conservation planning urgently requires a landscape perspective.
Currently most tropical countries feature various types of protection and management. In doing so, conservation prioritization has so far mostly focussed on species richness of an area. However, our previous work in Kenya had not only shown considerable differences in the proportion of forest birds among differently modified forest types but also strong differences in community structure of the avifauna. Thus, we had already shown that not all bird diversity measures are equally susceptible to forest modifications. The aim of our study now was to link these sensitive bird diversity measures to properties of forest composition and configuration. In other words, we intended to use community level modelling to spatially extrapolate bird diversity measures based on forest properties across the complete protected forest area of Kakamega Forest. This forest is special in regard to offering three different types of plantation forest habitats besides near-natural and secondary forest types. Our overall goal was thus to improve communication in conservation planning by visualizing the potential impact of different management decisions on future biodiversity.
The preconditions for such an approach were optimal within the interdisciplinary research project BIOTA (BIOdiversity Transect Analysis in Africa – www.biota-africa.org) as our team comprised both experienced field ecologists and remote sensing & spatial data modelling experts. Furthermore, we were actively involved in contributing to theKakamega Forest Ecosystem Management Plan. The planning process provided the required background for compiling adequate scenarios considering different needs and interests as all important stakeholder groups such as the local community, the managing authorities as well as scientists participated. The start for this publication was made during the international congress ‘Biodiversity of Africa – Observation and Sustainable Management for our Future!’ in 2008, where a set of normative forest use scenarios was developed.
Based on detailed land and forest cover classification data of 2003 (mainly derived from Landsat satellite imagery), we achieved five forest management scenarios as the final outcome of this effort. These included two extreme scenarios, the minimal scenario with a strict focus on forest utilization and the maximal scenario with a strict focus on conservation. Consequently, we assumed for the minimum scenario that the entire forest will be clear-felled and subsequently replanted with exotic monocultureswhereas for the maximal scenario it was assumed that all of the present forest plantations will be removed and that Kakamega will subsequently be strictly protected for decades. In-between these extremes we also selected three scenarios with different levels of forest utilization versus conservation.
In a next step we used GIS (geographical information system) to calculate a broader set of geo-spatial properties to describe the composition and configuration of our plots in the forest, i.e. the area for which we had collected the bird data. These geo-spatial predictors included measures such as forest patch size or the shape of a forest patch. We then identified the most informative forest properties for predicting bird diversity measures across all modified forest types. In a final step we linked the bird survey data with the future forest management scenarios to visualize their spatial distribution.
Our results highlight that management scenarios based on the species richness of forest specialists are very informative. However, projections based on community composition are even more meaningful, as they mirror not only the proportions of near-natural forest types, but also their patch shape. This highlights that it is important to differentiate effects of the areas covered by the various forest habitats and the degree of habitat fragmentation, both for species richness and community composition. Furthermore and most importantly, our study shows that combining land use scenarios as derived from remote sensing and field data on biodiversity can be a powerful tool to visualize potential consequences of forest management decisions on spatial patterns of future forest biodiversity. As such, the map visualizations of alternative scenarios are a valuable input to multi-stakeholder conservation planning processes in species-rich tropical forests where nature conservation and forest resource demands are competing.