Future Resilient Land-Use Visions for Valdivia, Chile

Resilience to climate change attracts the attention of scientists, city planners and politicians alike. Land-use change is a crucial theme in resilience strategies as it will affect the vulnerability of hydrological, ecological, and urban systems. For instance, unmanaged urban expansion may increase the risk of future floods because of changes in hydrological conditions that arise from converting natural lands into artificial lands, and an increase in flood exposure that stems from an increasing population within flood-prone zones.

In this study, we simulate 2080’s land-use envisions of Valdivia (Chile) to generate a more resilient city. Based on a series of workshops with representatives from city authorities, academia, and community-based organizations, we developed four different scenarios for the future land-use in Valdivia. The first scenario, named Eco-wetlands, aims at creating a network of protected wetlands that will become an indispensable component of the city to improve ecosystem services, biodiversity, and hydrologic resources. The main objectives of the second scenario, named Friendly city, are creating decentralized urban neighborhoods, and developing a network of urban green space connected with conserved wetlands to improve biodiversity and ecosystem. The third scenario, named Resilient to flood and drought, aims at connecting wetlands and urban parks, protecting native forests within the watersheds, and restoring the riparian areas. The fourth scenario, named Inclusive city, develops mini equidistant urban centers that include plazas, urban parks and community gardens.

A cellular automata-based model is used to simulate the above-mentioned scenarios. Two historical land-use datasets for Valdivia in 1983 and 2010 with a cell space of 20 m × 20 m have been used to calibrate the model. The model has two modules: demand and allocation. The demand module estimates the quantity of changes per time-step, corresponding to one year, which is obtained by a cross-tabulation between the historical maps. The allocation module allocates new land-use transitions based on two transition rule-sets. The first set is determined through a set of land-use change global drivers (explanatory variables). Eight static variables, representing accessibility and geophysical features, were selected as explanatory variables. In addition, the dynamic distance to urban areas and wetlands are calculated automatically during each time-step and introduced in the model as explanatory variables. The second set was determined based on the local neighbors of each cell. The weights of evidence method is used to calculate land-use transition probabilities.

The 2080 land-use pattern of the Eco-wetlands scenario shows remarkable biological green corridors crossing the city and along the river. The 2080 land-use of Friendly city scenario is mainly characterized by a pattern of continuation of urbanization combined with small dispersed patches of green areas. It also illustrates a network of connected urban green spaces and wetlands throughout the city. Resilient to flood and drought presents a protected and connected network of wetlands and a biological corridor along the river that improves permeability. In the Inclusive city scenario, several mini urban centers were developed in Valdivia that increases quality of life, rights, and access to services.