Investigating the neighborhood effect of urban morphological metrics on summertime land surface temperature variations in Istanbul, Turkey.

This study aims to explore three main questions. (1) Which urban morphological metrics are related to temperature variations? (2) Does the correlations between metrics and LST change with varying neighborhood sizes? (3) Are there different metrics related to LST for different-sized urban units? A Spearman's correlation analysis was performed to reveal the relationship between urban morphology metrics and land surface temperatures regarding five different neighborhood sizes, namely 1000 m, 500 m, 250 m, 125 m, and 62.5 m. Thirty-eight urban metrics related to land cover, topography, 3D surface, land use, complexity, and building and road properties have been calculated for Istanbul using Geographic Information System Environment. The results show that while the vegetation index, green area ratio, and the average distance between buildings have negative correlations, the rest have positive correlations. In addition, land cover indices, green area ratio, the number of buildings, and building coverage ratio are highly correlated for all neighborhood sizes. At the same time, other urban morphology metrics have varying correlation levels. The correlation of building coverage ratio, floor area ratio, perimeter–volume ratio, the fractal dimension of buildings, shadow intensity, and sky view factor increase with decreasing neighborhood size. The study indicates that the correlation level of urban morphological metrics with land surface temperature varies with neighborhood size. Therefore, different action strategies should be considered to improve urban microclimatic comfort on different scales. The number of samples can be increased, and ground truths can be used for validation to ensure the accuracy of results. [ABSTRACT FROM AUTHOR]