A project undertaken at the Centre for Tropical Environmental and Sustainability Science & School of Marine and Tropical Biology, James Cook University and supervised by William Laurance
This project is the subject of a PhD study by Eric Katovai at JCU, supervised by William Laurance.
Work on this project commenced in June 2012. A pilot study was carried out to select appropriate sites for the study and to confirm the feasibility of the proposed study design. This study is focussed entirely on a single island with an extensive logging history. Kolombangara Island in the New Georgia group (see Figure 1) has been selected for the following reasons;
- Kolombangara contains substantial patches of remnant forest reserves adjacent to naturally regenerating coupes of secondary re-growth.
- Logged forests vary in recovery time (i.e. 0 – 50 years).
- Isolation from propagule sources and logging history provide excellent contrast in the parameters thought to be important in forest recovery.
- The climate and geological morphology around the island are relatively homogenous (Hansell & Wall 1975; Whitmore 1969; 1989; 1999) unlike other large islands in the group.
- Flora distribution across remnant forests is relatively homogenous around the island (Katovai et al., 2012).
- The island’s terrain is more hospitable compared to other islands in the group, making accessibility easier.
Fig. 1: Kolombangara Island (delineated) is part of the New Georgia group and one of the highly logged islands in the Solomon Archipelago. (Map adapted from http://westernsolomons.uib.no/)
(Click on image to enlarge)
Fig. 2: Preliminary results. Plant species composition of logged and unlogged forests.
Fig. 3: Preliminary results. Independent t-tests showing comparable means between unlogged and logged forests with 30 years of recovery.
Fig. 4: Bill Laurance and PhD student Eric Katovai at the research site.
Fig. 5: Tree-coring by field assistant.
Fig. 6: Leaf scanning. Field assistants.
Fig. 7: Eric Katovai at the research site.
Kolombangara Island is an excellent choice for one further reason. Early understanding of rainforest dynamics on Oceanic islands has come from work done by Tim Whitmore and David Burslem undertaken on Kolombangara. These studies have been focused mainly on documenting patterns of species richness and composition (Whitmore 1966), vegetation types (Whitmore 1969), comparison of tree growth and forest change over 20-30 year time periods, (Whitmore 1989; Burslem et al., 1998) and forest response to cyclone-induced disturbances (Burslem & Whitmore 1999). Furthermore, all these studies were conducted within natural, intact forest systems. Less attention has been given to escalating human land-use impacts such as commercialised agriculture, logging and urbanisation on these systems (e.g. Denslow, 2003; Denslow et al., 2009; Sax et al., 2002; Sax and Gaines, 2008, Goldman et al., 2008; Katovai et al., 2012), resulting in little information being available to assess these processes against biodiversity and conservation criteria for previously logged forests.
While in the field, we consulted native land owners and stakeholders for permission to access and conduct research on their land. The responses we received were overwhelmingly positive. We acquired topographical and vegetation maps from the GIS division at Kolombangara Forest Product Limited to assist with site selection for the whole study. We also identified potential field assistants and ran preliminary field training for them.
In Honiara, we had meetings with personnel from the government forestry department to inform them about this project and its potential benefits to forestry management in the country. Given that their research arm ceased functioning in the 1980s, the ministry was very pleased with this cause and pledged to support it wherever possible. Meetings were also held with the Solomon Island national herbarium to discuss possible assistance in taxonomical sorting throughout the project. A senior botanist and also director of the herbarium, Mr Myknee Sirikolo has pledged his support and will be the major local collaborator in this project.
The second stage of field work commenced in December 2012, and will run till October 2013 or at an earlier time should the proposed work be completed beforehand. A second session of training will be done in mid-January 2013 before field work resumes after the festive season. We anticipate establishing 144 survey plots [0.35 ha per plot] across three elevation bands in logged forest of different recovery times [i.e. 40-50, 20- 30, 0-10 years] and unlogged forests within this duration.