Seventh Symposium on the Urban Environment
Seventh Conference on Coastal Atmospheric and Oceanic Prediction and Processes

J2.2

1984-2005 UHI trend over Montreal, Canada, using high resolution Landsat imagery and surface station data

Philippe Martin Sr., EC, Montreal, QC, Canada; and Y. Baudouin and M. Beauchemin

Urban heat islands (UHI) have intensified in most north-American cities in past decades and will continue to be on the rise due to climate change and increasing urbanization. Such a trend is a concern for public health authorities which have to cope with excess mortality and heat related health problems. Heat also affects air quality in big cities due to tropospheric ozone formation and greater pollen emissions. Health administrations in several cities need to better understand and quantify these trends. The present study aims at the quantification of the UHI trend in Montreal, Canada, over a 20 year period (1984-2005). Two methods were chosen and compared: high-resolution Landsat thermal imagery and surface weather data. The UHI "signal" used is the classical urban-rural temperature difference, but also intra-urban spatial variability. The urban-rural difference is calculated using satellite windows of varying pixel sizes centered on weather stations, as well as regional satellite windows representing “urban” and “rural” domains. The intra-urban differences are calculated between urban districts, as well as between targeted areas which have undergone intense urbanization and those which have not. Before these analysis techniques can be applied, appropriate satellite images must be selected. A specific methodology was developed for this which uses weather conditions (wind speed and cloud cover) at image time and for the previous 24 hours, synoptic situation and images equally distributed in time. In order to quantify the contribution of urbanization to the UHI trends, the correlation between surface temperature anomalies and two urbanization indices were calculated. These indices are the Normalized Difference Vegetation Index and the Normalized Difference Build-up Index (NDVI and NDBI). Influential factors other than vegetation and man-made environment are assessed. These results will help to better plan health prevention measures to cope with heat stress.

Joint Session J2, Characterizing the Urban and Coastal Climate: Thermal and Boundary Layer Structure and Atmospheric Responses
Monday, 10 September 2007, 1:00 PM-3:00 PM, Kon Tiki Ballroom

Previous paper  Next paper

Browse or search entire meeting

AMS Home Page