The Influence of Spatial Extent Shape on LST-NDVI Patterns: A Multi-Scale Perspective

Abstract Understanding the relationship between vegetation cover and land surface temperature is vital for analyzing landscape structure and addressing climate-related challenges. While extensive research has focused on the scale of effect in multi-scale analyses, the impact of spatial extent shape on LST-vegetation correlations remains largely unexplored, representing a global knowledge gap. Most studies rely on circular or square shapes, overlooking how different geometries may alter analytical outcomes. This study aims to investigate how spatial extent shapes influence the correlation between LST and the Normalized Difference Vegetation Index. Using data from western Iran, we analyzed five geometric shapes—circular, square, elliptical, hexagonal, and diamond—at multiple spatial extents ranging from 90 to 990 meters, applying Pearson’s correlation and statistical tests. Our findings reveal that while circular, square, and elliptical shapes yield similar results, hexagonal and diamond shapes introduce significant variations, particularly at smaller extents, with p-values as low as 0.00. Additionally, we observed that the correlation between land surface temperature and the Normalized Difference Vegetation Index strengthens as the spatial extent increases, peaking at 990 meters. These results demonstrate that the choice of spatial extent shape can significantly impact the interpretation of LST-vegetation relationships, highlighting the need to move beyond traditional circular or square extents. This study provides novel insights into spatial data aggregation methods and offers a framework for enhancing landscape analysis globally. By emphasizing the importance of spatial geometry in ecological studies, the findings hold relevance for landscape ecologists, urban planners, and environmental researchers seeking to refine multi-scale analyses and improve landscape-scale decision-making worldwide.