Land suitability evaluation for multiple crop agroforestry planning using GIS and multi-criteria decision analysis: A case study in Fiji

We present novel methods that use geographic information systems (GIS) and multi-criteria decision analysis (MCDA) to evaluate land suitability for a set of seven agroforestry crops, with the aim to differentiate relative land suitability and its potential to achieve different benefit goals for the Sigatoka Valley, Fiji. Our first model (Land Suitability) identifies optimal areas for each crop based on readily available edaphic and environmental cultivation criteria and common GIS datasets. The second model (Weighted Maximum) uses objective approaches for weighting the relative importance of target crops to derive a multi-crop suitability map. We use the analytical hierarchy process (AHP) technique (a form of MCDA) to identify the relative importance of five benefit types (i.e. agroforestry initiative goals/AHP Objectives). AHP criterion weights were used to map the most important crops for different agroforestry goals. Our methods are unique among GIS-MCDA applications of land suitability analysis in that our aim was to investigate and spatially evaluate land use suitability for multiple crops on a per-crop basis, whereas the aim of most GIS-MCDA land suitability analyses is to evaluate relative suitability (e.g. low, medium, high), evaluate potential for different land uses (e.g. production, intensive, or multifunctional) or land suitability for a single crop. We conclude that the methods described can be adapted to agroforestry initiatives and other similar land use suitability applications in the Pacific region and other geographical settings.