Developing High-Resolution Descriptions of Urban Heat Islands: A Public Health Imperative
GIS SNAPSHOT — Volume 13 — September 15, 2016
PEER REVIEWED
Two maps of Portland, Oregon, located at the confluence of the Columbia and Willamette rivers, are shown. One shows the range of temperatures at 6 AM; the other shows temperatures at 7 PM. At 6 AM, temperatures range from 13.0°C (55.5°F) to 18.2°C (64.8°F). At 7 PM, temperatures ranged from 21.7°C (71.1°F) to 33.2°C (91.7°F).
Empirically derived 1-m–resolution descriptions of (A) morning and (B) evening urban heat islands in Portland, Oregon, on August 25, 2014. Temperatures were recorded in 1-hour periods at 6 AM and 7 PM. In the morning, low-lying vegetation cover had the strongest effect on temperature; in the evening, temperatures were most strongly affected by variation in building heights. High-resolution data sets such as those used here can inform preparation for extreme heat events and public health interventions.
Top-down view (component A) of a small portion of the Portland, Oregon, canopy; canopy (trees) areas are indicated by green pixels, whereas not-canopy areas are indicated by white pixels. The image is overlaid with a series of 8 increasingly larger red concentric rings (or buffers), depicting ground distances in the calculation of canopy within a specific buffer distance. A yellow star in the center indicates a point of interest for the buffering process. Component B has a similar scene to component A; however the concentric rings have been removed and slightly more area of the raw canopy data are visible. The next components (C, D, E, and F), illustrate the output of a standard focal, or moving window, analysis: the data set from component B begins to blur, as if going out of focus, as each pixel in the scene converts from canopy and not canopy values to a number representing the percentage of ground cover classified as canopy within each buffer distance (component , 50 meters; component D, 100 meters; component E, 200 meters, and component C, 600 meters.
Figure. Example of the process of creating buffers on a raster. The input data (A) is converted from vector to a raster that contains pixel values representing the results for specified distances. In this example, the raw data has the values 1 (green, representing canopy) and 0 (white, representing not-canopy). The output describes the percentage of land cover classified as canopy.
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.