Abstract

As populations and economies expand, driving the inevitable growth of urban centers and the emergence of urban sprawls, the World Bank (2023) projects a global addition of approximately 1.2 million km2 of new urban built-up area by 2030, with around 60% of the global population expected to reside in urban areas by 2050 (UNDESA, 2018). Concurrently, Sustainable Development Goal 11.3 aims to enhance inclusive and sustainable urbanization by 2030, advocating for participatory, integrated, and sustainable human settlement planning and management. Given these critical implications and targets, comprehensive measurement and monitoring of urban growth have become imperative. Our study seeks to develop models for identifying urban growth patterns, utilizing built-up area data from the Global Human Settlement Layer (GHSL), which offers periodic and global insights into built-up areas. Leveraging a deep learning approach known as Convolutional Gate Recurrent Unit (Conv-GRU), these advanced models not only facilitate the identification and prediction of urban growth patterns but also provide insights into new urban sprawls, informing informed urban planning and sustainable development initiatives. Applied to analyze urban growth dynamics within Gerbangkertasusila, one of Indonesia's largest metropolitan areas, second only to Jabodetabek, our study aims to derive actionable insights to foster sustainable urban development within this burgeoning metropolitan region.