Developing a methodology for identifying, mapping and potentially monitoring the distribution of general farming system types in Vietnam’s northern mountain region

Abstract

This study tests a method to identify and map the spatial distribution of general farming system types in five districts of the northern mountain region (NMR) of Vietnam. Over the last 50 years the NMR has suffered from a large loss of forest cover, often blamed on the swidden farming systems that are found in the mountains. As a result different programs have been put forward to change local land use practices: to decrease the amount of swidden land, “sedentarize” farmers responsible for practicing swidden agriculture, and introduce new farming systems. However, some researchers have identified one type of swidden farming system, composite swiddening, as ecologically more stable, causing little net deforestation and less environmental degradation than other farming systems in the NMR. To date no study has identified how widespread this or other types of farming systems are in the NMR. This study makes use of geographic data for five districts in the NMR within a GIS, combined with field checking, to characterize, identify and map the spatial distribution of the general farming system types, including composite swiddening, by commune for these districts. The results of the mapping indicate that there are few communes where a single farming system type is found. As these districts are spread over five provinces, the results reinforce the view that in Vietnam’s NMR, farming system types vary down to below the commune level. An 80% agreement between the results shown on the map and field checking resulted from an examination of 19 communes in the districts mapped. A comparison was done of 2000 data to 1992 data for one district. An analysis of the communes’ farming systems for 1992 and for 2000 indicate that permanent agriculture farming systems of both rice paddy agriculture and permanent upland agriculture are appearing and replacing swiddening systems.