Multi-scale analysis of human-environment interactions : a case-study in the northern Vietnamese mountains

The thesis contributes to improve our understanding of drivers of land use/-cover change in an extremely diverse area in the northern Vietnamese highlands by applying a multi-scale and multi-level approach in which regional-scale analyses are linked with household data. The land cover analyses from multi-source datasets showed a net loss of forest area in Lao Cai province during last 60 years (1952-2012). Since early 2000s, there is a significant increase of forest cover and decrease of arable land and shrubs. Reforestation in the area occurred about 10 years later than that was reported at national level. Land use/-cover change trajectories mask substantial heterogeneity that exists at local level. Accessible areas with rapid economic development are characterised by an increase in plantation forests, while poverty-stricken and isolated areas experienced an increase in natural forests. Deforestation is still observed in areas with high land pressure. Forest transition in Lao Cai province was not only the result of strong forest policy but was also enhanced by agricultural intensification, market integration, and economic growth. The analyses at village and household level showed that the creation of off-farm jobs in the tourism sector can be a driver of shifts in coupled human-environmental changes. Tourism increased the average living standard of the ethnic mountain minorities, and led to agricultural intensification with forest regrowth on abandoned fields. Finally, multi-level analysis showed that land use intensification in mountain areas strongly depends on local dynamics at household level. This study suggests that household level analyses are crucial to understand human-environment interactions in mountain areas.(SC - Sciences) -- UCL, 201