Urban ecological security assessment and forecasting, based on a cellular automata model: A case study of Guangzhou, China

Abstract

Forecasting changes in urban ecological security could be important for the maintenance or improvement of the urban ecological environment. However, there are few references in this field and no landmark research work has been reported, particularly quantitative research. A forecasting model for ecological security based on cellular automata (CA) was developed using preliminary spatial data from an ecological security assessment of Guangzhou conducted previously (1990–2005). The model was constrained using transformation rules based upon proposed planning for 2010–2020. A simulation accuracy of 72.09% was acquired. Using a one-bit assessment grid for 2005 as the starting state for the simulation, the model was used to forecast ecological security for 2020. This revealed that although the ecological security status would be improved relative to current trends, there would still be an overall decline in ecological security over the next 15 years. Even if new urban plans were implemented, landscape pattern analysis suggested a more scattered and homogenous distribution in the urban landscape of Guangzhou and significant variation in landscape characteristics among districts. This suggests that further measures must be adopted to reverse the current trends in Guangzhou's ecological security. The model highlights the need to make ecological protection an integral part of urban planning. This study demonstrates the potential of CA models for forecasting ecological security. Such models could make an important contribution to decision-making for regional governors and to the development of urban planning incorporating assessment and prediction of ecological security.