Savanna ecosystems cover an ever-increasing proportion of the Earth’s surface, and house a rapidly growing proportion of the world’s population and livestock. An often overlooked but increasingly urgent threat to savannas is posed by the process of woody encroachment, by which woodland encroaches upon grasslands used for livestock production, the principal means of subsistence in these areas. Grass and trees interact with each other through a positive feedback loop mediated by fire: more grass means more fuel load for the bushfires that keep woody vegetation at bay, which in turn enables further grass growth. Livestock grazing removes fuel load and hence indirectly stimulates woody encroachment. Livestock holders are thus faced with balancing cattle stocking rates and the long-term sustainability of the savanna ecosystem. In our paper we aim to show how livestock management and human-lit bushfires affect the resilience of savanna ecosystems against the process of woody encroachment, using dynamical systems theory. Our research serves as a foundation for future optimal control analysis.
This paper examines the effect of land subsidence, the sinking of the land surface, on property values. Land subsidence can affect properties by damaging the foundation or increasing their exposure to flood risk. Based on detailed property transaction data from the Netherlands and high-quality geodata on current and future subsidence and flood risks, we find that property that is currently subsiding trades at a 0.8% discount if it is built on a foundation prone to damage. Flood-prone properties predicted to subside in the future trade at 1.5% lower prices. The estimates indicate that buyers underestimate foundation risk but overestimate flood risk. Furthermore, an increase in Google searches for subsidence by one standard deviation leads to an additional price discount of 0.8% for houses in subsiding areas. This discount is lower for the properties with the highest subsidence risk, further indicating that risk is not adequately priced.