Return to Initiatives

PEOPLES Resilience Framework for Mapping Recovery in Haiti

Community resilience is defined as the ability to prepare for and recover from extreme events (both natural and man-made disasters). The PEOPLES Resilience Framework was designed to enable researchers from different disciplines to measure, model and describe seven different categories of community resilience indicators (http://mceer.buffalo.edu/quakesummit2011/program/presentations/08_d-Renschler.pdf).

As part of Dr. Renschler’s Landscape Modeling with GIS class at UB, graduate and undergraduate students were asked to review a series of historic aerial images for a number of city blocks within the capital of Port au Prince Haiti.  They were instructed to look for evidence of recovery on aerial images in Google Earth using the image time slider tool to inspect pre- and post-earthquake images over the period of time represented by the suite aerial images.

The students faced a series of technical challenges due to the variability of the image quality (resolution, lighting, shadows, clouds, and image registration issues), as well as the student’s inexperience interpreting photos to identify resilience indicators within the PEOPLES Framework.  On top of these technical issues, the post-earthquake aerial images only span through 2010 limiting the amount of recovery that could be detected within a single year.  Despite these challenges, the students were able to record a handful of parameters for a cluster of city blocks south of the Haitian Palace.

City blocks were created as polygons from street centerlines in ArcGIS desktop, properly attributed to create data fields to record the PEOPLES parameters and converted to KML files.  Students uploaded the KML files to Google Earth and performed their inspections using a predefined set of aerial images available on Google Earth’s image time slider.  Each block was inspected and assigned a numeric value from 0-100% to note each of the assigned parameter’s functionality.  For example, how much debris was visible in the streets surrounding a city block and whether the debris impeded traffic? What percentage of buildings within a given block showed signs of damage?  Was there evidence of commerce (presence or absence of street vendor’s market umbrellas)?

The data was compiled and formatted to be displayed in Tableau Public in order to make interactive maps and charts to compare student’s interpretation of the recovery indicators.  An interactive data investigation dashboard is displayed below which features four map views, each one show the distribution and the observed recovery variables (from 0 to 100%) collected by 6 different students. The students were allowed to choose their parameters of interest. Some chose 1 parameter. Others chose 2 parameters. Some focused their observations on a small cluster of city blocks while others inspected each of the 34 city blocks in this sample study area.

As you move your mouse across the maps, chart, and small-multiple map display, the “hovered” city block will be highlighted in each data view window and a tool tip will display specific information about the observations made within that city block.

Clicking on the checkboxes on the right side of the viz will change the focus and the amount of data being displayed. Clicking the arrows on the Animate Maps by Photo Date control will change each of the map views to reflect the observations made by the students for that specific time-stamped Google Earth aerial photo.

Have fun and explore the data…