A moving window-based spatial assessment method for dynamic urban growth simulations

Abstract

This study proposes a spatial evaluation method for urban growth simulation based on moving windows, where the metrics measured within each window are considered to be those of the central cell. We also applied the generalized additive model to identify the quantitative relationship between the urban growth drivers and the spatial assessment metrics. A case study in Jiaxing city shows that the single-number overall accuracies (OAs) are above 94% and the figure-of-merits (FOMs) are above 27% in both 2010 and 2015. Most regions of the study area yield very high OAs and low FOMs while the regions around the administration centres yield low OAs and high FOMs. The spatial method can well indicate the model’s effects on the urban simulations in different regions. The spatial assessment can report the assessment metrics of each cell to produce assessment maps as well as quantify the relationship between drivers and assessment metrics.

Keywords:

• Single-number assessment
• spatial assessment
• moving window analysis
• cellular automata
• urban growth
• generalized additive model

Data availability statement

The datasets are freely available at https://figshare.com/s/af0a0c3aa21dc139e0a4.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

Supported by the National Natural Science Foundation of China (42071371) and the National Key R&D Program of China (2018YFB0505400).