Combination of spatial logistic regression and geographical information systems in modelling wetland changes in Setiu basin, Terengganu

Wetlands played an important role in protecting our natural ecosystem. However, recently there were some issues has been highlighted that affected the natural function of wetlands led to the decline of the biodiversity of flora and fauna. Therefore, this study was carried out to examine the changes occurred in the wetland area of the Setiu watershed, Terengganu and to identify the factor that influence and cause the wetland changes. Integration between Geographic Information System (GIS) and spatial logistic regression has been used to identify the spatial and non-spatial factors that contribute to the wetland changes and to evaluate the significance of the respective factors. Factors that have been used in this study are mean annual rainfall, distance from slope, soil suitability, distance from roads, distance from streams, population density and distance from agricultural areas. Systeme Pour l'Observation de la Terre (SPOT) 5 images have been used to extract data of the wetland changes from 1990 to 2000 for the development of the model, while data from 2000 to 2010 was used to evaluate the accuracy of the model developed. The result shows that the independent variable that significantly influences the wetland changes is the distance from agricultural areas, while the independent variable that has less significant influence is population density. These generated maps show the spatial extent of the Setiu wetlands was change 35% in 1990-2000 and increased 19% to 54% in 2000 – 2010. The results show that the wetlands area in Setiu district is decreased and should get attention from local authorities for conservation action. Overall, this study has proven that the integration of GIS and statistical analysis become a great tool that has ability to model the wetland change in relation to time and space.