Modelling multi-criteria land evaluation using GIS-MCDA (Multi-Criteria Decision Analysis) based on fuzzy logic: Case study for identifying suitable lands for human rehabilitation in Southern Bhutan.

Abstract

Systematic and effective land suitability evaluation is an indispensable prerequisite to achieving sustainable and optimum utilization of land resources. However, the land evaluation is a complex multi-criteria and objective decision-making process that is often confronted with conflicting views, ambiguity and consequential risk.

This study attempts to develop a model framework for multi-criteria land evaluation using an integrated Geographic Information System and Multi-Criteria Decision Analysis (GIS-MCDA) technique based on fuzzy logic. The land evaluation model is used in a case study for identifying suitable land for human rehabilitation in two Districts of Southern Bhutan that covers with an area of 2294 km2.

The human rehabilitation program encompasses two objectives viz. evaluating land for residential and agriculture purpose. Based on the requirement of the National Rehabilitation strategy documents, a total of seven main-criteria and twenty sub-criteria considering social, economic, climate, environments and topography are identified, and their relative importance evaluated by a group of relevant experts. The suitability maps are then generated by the GIS-MCDA model.

The model included sensitivity analysis to validate the robustness of the land evaluation model and the stability of criteria against the subjectivity of the expert’s judgements. The suitability class has a minimum change of areal values ranging from 0.2% to 36.0% and the spatial pattern has minimal variation under the changed criterion weight percentage.

It is observed that over 50% and 47% of the study cannot be considered for residential and agriculture land use respectively. This primarily due to the existence of large parks and biological corridors zones which enforce the reservation and protection. The highly suitable (S1) area for future residential and agriculture is only 5.01 km2 (0.2%) and 18 km2 (0.8%) respectively. Under the best scenario, this can be recommended for planning human rehabilitation in future.

There exists conflict between the two objectives. The identified residential land is in full conflict with agriculture land. The conflict is resolved using Multi-Objective Land Allocation (MOLA) tool.

The suitability maps with six different levels of risks and tradeoffs are generated using OWA method which helps decision-maker(s) understand how the suitability values change depending upon the level of risk and tradeoff one wish to assume.

To date, there is no record of any scientific based method like GIS-MCDA used for land suitability analysis in Bhutan. It is positioned that this land evaluation model can significantly enhance the land evaluation for human rehabilitation. In general, this will also serve as a useful tool in achieving the sustainable and optimum use of a limited land resource and ultimately contribute to effective land management and planning in the country.

-