A recent study analyzing the land use and land cover (LULC) in the United Arab Emirates (UAE) over the last 50 years has unveiled significant changes that could have far-reaching implications for various sectors, including mining. Conducted by Mubbashra Sultan from the Department of Geosciences at the United Arab Emirates University, the research utilized 72 multi-temporal Landsat satellite images to assess how urbanization, agricultural expansion, and megaprojects have reshaped the landscape.
The findings are striking: the study shows that desert and mountainous regions, which dominated the UAE’s geography, have seen a decline from over 97% coverage in 1972 to nearly 91% in 2021. In stark contrast, built areas have increased from less than 1% to almost 6%, while vegetation cover has more than tripled, albeit from a low base. Water bodies, too, have fluctuated, reflecting the dynamic nature of the region’s development.
Sultan pointed out, “This long-term analysis not only captures the rapid urbanization but also highlights the environmental changes that accompany such growth. Understanding these shifts is crucial for sustainable resource management.” This insight is particularly relevant for the mining sector, which often operates in similar arid environments. The ability to monitor land changes using advanced machine learning techniques can provide mining companies with precise data on land availability, potential environmental impacts, and regulatory compliance.
The research employed three machine learning classifiers, ultimately selecting the Random Forest (RF) algorithm for its superior performance in classifying the data into four distinct LULC classes: built areas, vegetation, water, and desert/mountainous regions. The model’s accuracy ranged impressively from 85.11% to 98.4%, showcasing the effectiveness of machine learning in processing complex geospatial data.
As the UAE continues to diversify its economy beyond oil, understanding land use changes is paramount for strategic planning in sectors like mining. The insights from this study can guide mining companies in making informed decisions about where to operate, how to minimize environmental impacts, and how to align with national sustainability goals.
This research, published in ‘Frontiers in Earth Science’, underscores the transformative potential of integrating machine learning with remote sensing technologies. As Sultan noted, “The results serve as a vital tool for policymakers to manage land resources, urban planning, and environmental conservation.” The implications extend beyond immediate commercial interests; they pave the way for a more sustainable and responsible approach to resource extraction in the region.
For more information about Mubbashra Sultan’s work, visit lead_author_affiliation.
