Introduction

This project is directed towards establishing the nature and extent of agricultural land use around the perimeter of the Kakenauwe Forest Reserve. This is of importance as future pressure upon the reserve may well result in damage to the forest ecosystem. The project was therefore focused upon establishing land use patterns, agricultural systems, land quality and future trends in land use in the area bordering the reserve.

Aims

The project was designed in order to provide more detailed information on the extent of agricultural land clearance through the use of hand-held GPS equipment. This would be used alongside remote sensing data to provide an accurate and up to date picture of land use. This information will be supplemented by interviews with farmers which would allow the project to identify current systems of land use and potential areas where future agricultural expansion may be required.

Methodology

In construction of this data two dominant data collection techniques prevailed, this comprised of a semi-structured interview and a mapping exercise. The semi-structured interview was used to extrapolate data such as crop type, yield, length of occupancy and land management practices. The advantage of a semi-structured interview is in its ability to be flexible with questions and answers. This means that the interviewee can expand on any answers and the questioner can also expand on any points raised by the interviewee. To further enhance the understanding of Labundobundo farming, all the interviews were done in compilation of the mapping exercise. This was completed using a hand operated Global Positioning Satellite (GPS) device. This allows each of the four corners in any one plot of land could be mapped according to its seven figure UTM grid reference, however, this technology does carry with it a 50 metre degree of variance, which means the grid reference may be exact or it may be wrong by up to 50 metres. In terms of accuracy of data on a map to the scale of 1:50,000 the data will be reasonably accurate. However, a higher resolution may show a greater number of anomalies. To help increase accuracy two GPS devices were used simultaneously which gave an idea of variance, where-by the mean average of the two would be the recognised point for which to plot geographically. Also, to support data accuracy a variance control was completed over four days. For this control, the same spot was mapped at the same time of the day over a period of twenty minutes during five-minute intervals. This revealed the variance in grid reference for the same location.

Initial results

At this initial stage of the project it is very complicated to interpret the data to any great significance, as the GPS device did not allow hectare area to be mapped, or interview to be complied. To complete this section of the project a Geographical Information Systems (GIS) computer programme will be introduced, which is available at the University of Portsmouth. With this programme both the GPS data and the interview data can be inputted to create a variety of maps interpolating a variety of quantifiable data. This provides a picture view of the raw data together in one geographical region, which can then be further interpreted as a whole. It is hoped that this computer programme will provide a reasonably accurate portrayal of land ownership distribution, duration of occupancy, predominance of subsistence or commercial croppage and what variety of crops are being grown, all in relation to the reserve boundary.

Final report

The results of this project will be collated into a report entitled Agricultural land use adjacent to the Kakenauwe Nature Reserve by Stuart lane, University of Portsmouth. This will be completed by May 2002.