Combined vegetation volume and "greenness" affect urban air temperature

Cities are often substantially warmer than their surrounding rural areas. This 'urban heat island effect' can negatively affect the health of urban residents, increase energy usage, and alter ecological processes. While the effect of land use and land cover on urban heat islands has been extensively studied, little is known about the role of vegetation volume or built-area volume about this phenomenon. We ask whether the 3-dimensional structure of urban landscapes influences variations in temperature across a city. Using heights-above-ground information derived from LiDAR data and the Normalized Difference Vegetation Index (NDVI) calculated from multispectral (4 band: Blue, Green, Red, and Near Infrared) aerial images, we estimated vegetation volume and built-area volume (non-vegetated) in Chicago, Illinois (USA). Daily minimum temperature data were obtained from 36 weather stations for summer 2011. The differences in urban air temperature across the study area were as large as 3 degrees C. Maximum likelihood models indicated that a combination of NDVI and vegetation volume best predicted nighttime temperature in Chicago, and that vegetation growing within 250-500 m of the weather station was most influential. Our results indicate that vegetation in "the matrix", i.e. the area outside parks and preserves, is important in temperature mitigation since the majority of the vegetation volume in the study area occurs within residential, commercial/industrial, and institutional land uses. However, open space, which covers only 15% of the study area, has nearly as much total vegetation volume as residential land, which covers 61% of the study area. Clearly, both large wooded parks within a city and large trees scattered across residential areas are needed to best mitigate the urban heat island effect. (C) 2016 Elsevier Ltd. All rights reserved.