Abstract: Mastering the changes of urban areas and constructing a provincial-scale ecological security pattern is crucial for maintaining regional ecological security. We collected nightlight data, land use data, and point of interest data of Liaoning Province using the Google Earth Engine. We used a fusion algorithm to extract different urban built-up areas in Liaoning Province. With those data, the spatial and temporal variations of ecological security pattern and urban expansion in Liaoning Province were explored. The results showed that: (1) A fusion extraction method could compensate for single data source deficiencies in provincial-scale built-up area extraction, which increased the average extraction accuracy of built-up areas to 88%, being 33% higher than that of single nightlight data. (2) Based on the morphological spatial pattern analysis model and the minimum cumulative resistance model, the core regions were extracted, 28 ecological source areas were selected, 58 ecological corridors were identified. Ecological security distances with different radii were established as buffer zones to construct the ecological security pattern at the provincial level. (3) Based on the built-up area, population and GDP, the center of gravity of the city was established. With the condition in 2012 as reference, the center of gravity of the city was shifted to the southwest at a small angle. It is of great significance to combine the long sequence time-series changes of built-up areas and the constructed ecological security pattern of ecological sources, ecological corridors and ecological buffer zones as urban ecological planning for improving urban planning and ecological coordination at the provincial scale.

Key words: urbanization, provincial ecological security pattern, morphological spatial pattern analysis, minimum cumulative resistance, ecological corridor