Assessing Land Cover Change Projections and Its Suitability with Urban Spatial Pattern Plan Using Google Earth Engine and Cellular Automata in the National Activity Centres on Sumatera Island

Abstract

Abstract The national activity centres (Pusat Kegiatan Nasional – PKN) areas, especially in Sumatera Island, are closely related to the land cover change phenomenon. This study aimed to identify the land cover change projection and assess its suitability with the spatial pattern plan in the “Rencana Tata Ruang Wilayah” (RTRW) which focuses on the study area including Medan, Pekanbaru, and Padang. The integration of remote sensing data, geographic information systems (GIS), and spatial analysis techniques are used to identify and assess the patterns of land cover change over the specific period. The analysis of land cover change patterns is performed using Landsat-7 and Sentinel-2A imagery data which processed in Google Earth Engine (GEE). Furthermore, Cellular Automata and Artificial Neural Network (CA-ANN) modelling is used to identify land cover prediction using transition rules like main road as driving factor then followed by overlay analysis to assess its suitability with the spatial pattern plan. The result shows the different level of each land cover classification in sub-district level. The proportion of the land cover projection suitability in Medan at 2042 was 84,92%; in Pekanbaru at 2040 was only 58,51%; while in Padang at 2030 was 77%. This findings can be able to serve as a suggestion for stakeholders in formulating effective strategies for spatial planning evaluation and managing land cover uses, especially in several national activity centers in Sumatera.