Transition from subsistence grazing to nature-based recreation: A nuanced view of land abandonment in a mountain social-ecological system, southwestern Cape, South Africa

Abstract Over the last 50 years, studies have shown a decline in the use of mountain lands, a phenomenon termed land abandonment. We investigate the causal mechanisms of land use change in a mountain catchment important for regional water supplies in the southwestern Cape of South Africa. Uniquely, we include nature-based recreational land use types typically excluded from land abandonment studies. We repeated a structured questionnaire originally conducted with landowners 38 years previously and analysed orthoimages from 1948, 1972 and 2014. To identify and contextualise causal mechanisms of change we used thematic analysis and generated narratives from in-depth interviews with landowners. The land use system in 1948 and 1972 was dominated by livestock-based, subsistence agriculture and small-scale farming. This transitioned to agricultural intensification on the lowlands and small portions of the mountains and the utilisation of mountains for non-economic nature-based recreation and ecotourism. The use of frequent small, low intensity-controlled fires was prevalent in the past. More recently, fires have been actively suppressed resulting in the build-up of biomass and the development of infrequent, extensive, high-intensity wildfires. Land use change in the mountains was driven primarily by socio-economic drivers, including socio-economic benefits related to globalisation and economic growth, and not by concerns over land degradation or resource depletion. Our findings support evidence that shows that people’s responses to economic opportunities drive local determinants of land use change and highlights the importance of perceptions in driving land use transitions. We show that existing models of land abandonment are likely overly deterministic in that they do not consider social and cultural factors that may cause a landowner to continue using their land for semi-economic or non-economic reasons. While there is merit in large-scale remote sensing studies, we emphasise the importance of using mixed remote sensing and social science methods for informing models of land use change.