General Events

Back to Listing

Examining our changing urban environment: from spatial pattern to sustainable development

Event Type
Seminar/Symposium
Sponsor
Department of Geography & GIS
Virtual
Date
Oct 30, 2020   3:00 - 4:00 pm  
Speaker
Dr. Soe Myint, Arizona State University
Cost
This event is free and open to public
Registration
Registration
Contact
Department of Geography & GIS
E-Mail
geography@illinois.edu
Views
43
Originating Calendar
Geography and Geographic Information Science

We believe that building sustainable cities is a must to achieve a sustainable world. While the relationship between fractional cover of anthropogenic and vegetation features and the urban warming or heat island has been well examined, the effect of spatial pattern (e.g., clustered, dispersed, random) of these features on air and land surface temperatures (LST) are not well understood. This presentation demonstrates if and how spatial configuration of urban landscapes influence surface and air temperatures in urban areas. It will also cover if and how spatial pattern approaches can be used to measure traditional landscape matrices and examine urban dynamics.

 

The data used to classify detailed urban land cover types and generate landscape indices include QuickBird and Landsat imagery. Classification was performed using an object based image analysis (OBIA). The Landsat TM was also used to examine the continuous urban landscape heterogeneity. We employed a spatial autocorrelation approaches (e.g.,  Moran’s I, Getis-Ord G) that measure the spatial dependence of a point to its neighboring points and describes how clustered or dispersed urban features are arranged in space as well as hot and cold spots of these feature values. We used Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) data acquired over Phoenix to examine daytime and nighttime surface temperatures with regard to spatial arrangement of anthropogenic and vegetation features. We compare the traditional landscape metrics to the use of satellite imagery based local spatial autocorrelation measures in quantifying landscape structure over Phoenix urban area. We spatially correlate Moran’s I values of each land cover per surface temperature, and develop regression models. We also integrate image time series, continuous spatial indices, and non-parametric trend analysis into a spatiotemporal study of urban dynamics. Moreover, a space time modeling approach is also used to examine increasing or decreasing trends of urban warming and emissions using MODerate-resolution Imaging Spectroradiometer (MODIS) LST 8-day composite imagery (MOD11A2) and MODIS Terra Atmosphere Aerosol Level 2 Product (MOD04_3K). Based on our results, it is recommended that policymakers, city managers, and urban planners incorporate and optimize the spatial configuration of urban landscapes for a sustainable future.

link for robots only