A Cellular Automata Model to Simulate Land-use Changes at Fine Spatial Resolution

Date
2012-12-14
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Cellular Automata have become an important tool for modeling urban growth and land-use changes in recent decades. However, few studies have evaluated the possibility of building a CA model at fine spatial resolution for understanding detailed land-use dynamics. This research was carried out to develop a fine-resolution CA model, with an emphasis on tackling two main challenges. The first challenge is that the model complexity dramatically increases due to the increased number of land-use classes and driving factors at fine resolutions. In this thesis, Rough Set Theory was applied to guide the selection of dominant driving factors for the model calibration, resulting in a similar or improved performance as compared to the use of all factors. The second challenge involves designing a suitable CA model for land-use modeling at fine spatial resolution. Traditional cell-based CA are unable to generate reliable results at such resolutions because single cells often only represent components of land-use entities (i.e. houses or parks in urban residential areas), while recently proposed entity-based CA models usually ignore the internal heterogeneity of the entities. This thesis proposes a novel patch-based CA model to simulate land-use changes where the real-world entities are represented as patches. A patch refers to a collection of adjacent cells that, when combined together, represent an entity differing from its surroundings in nature or appearance. The results reveal that the patch-based CA model generates compact and realistic land-use patterns as found in the historical land-use maps. Calibrated in the eastern Elbow River watershed (adjacent to the City of Calgary), this model was further applied to simulate future land use in this area under three different development scenarios: business-as-usual scenario, protective growth scenario and smart growth scenario. The results reveal that both the protective growth scenario and the smart growth scenario consume less non-developed lands (i.e. agriculture and forest) than the business-as-usual scenario does. The resulting maps generated by the patch-based CA model clearly illustrate that increasing land-use efficiency is an effective way to reduce the impact caused by a rapid population growth.
Description
Keywords
Environmental, Geotechnology
Citation
Wang, F. (2012). A Cellular Automata Model to Simulate Land-use Changes at Fine Spatial Resolution (Doctoral thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca. doi:10.11575/PRISM/24656