Thursday, November 1, 2012

Mapping Hurricane Sandy

I am sure it is not news to a lot of people that the path, destruction and emergency facilities for Hurricane Sandy have been mapped.  The BBC provides a map of damage and emergecny faclitites (http://www.bbc.co.uk/news/world-us-canada-20154479). Likewise other news organisations provide similar information to readers, but these maps tend to be fairly static in providing a single set of inforamtion. Although you can zoom and change scale on the maps, they have a relatively low level of interactivity. These maps also rely on information being provided to the organisation from another source.

Integrating data sources and providing maps of informtion that is of use to a range of users is another, more difficult task. One of the key players in online mapping has been Google Interactive Maps ( http://google.org/crisismap/sandy-2012). The map for Hurriance Sandy is produced under the umbrella organsiation of Google Crisis Response (http://www.google.org/crisisresponse/) a Google project that collaborates with NGOs, government agencies and commerical organisations to provide information on things such as storm paths and emergency faciltities. Integrating these data sources to provide spaitally located information of use to responders, locals affected by the hazard, interested news readers and authorities involves a clear understanding of the requirements of each target audience and clear planning of the nature of information presentation for each of these. It is worth having a look at the interactive maps to assess for yourself whether this complex task has been achieved. In particular look at the type of information provided about particular resources, the nature of the resources mapped and the scale or level at which this information is displayed and to which part of the audience you beleive this information will be of use (why you think so is the next question).

It is also useful to compare this structured information with the less structured flow of information from Twitter (http://www.guardian.co.uk/news/datablog/2012/oct/31/twitter-sandy-flooding?INTCMP=SRCH). This analysis by Mark Graham, Adham Tamer, Ning Wang and Scott Hale looked at the use of the term 'flood' and 'flooding' in tweets in relation to Hurricane Sandy (see also blog at http://www.zerogeography.net/2012/10/data-shadows-of-hurricane.html. Although they were initially assessing if there was a difference in English language and Spanish language tweets about the storm, their analysis pointed out that tweets were not that useful at providing information aobut the storm at a spatial resolution of smaller than a county (although it isn't clear to me at least if they were mapping the location of the tweets or the contents of the tweets - in some cases the latter might provide some more detailed spatial information on flooding and its impact but would require extraction and interpretation from the tweet itself). This lower spatial resolution and the unfiltered, personal, subjective and unco-ordinated nature of this information source means that it is more difficult to quickly translate into information that is 'useable' by other audiences. Mark Graham is a researcher at the Oxford Internet Institute (http://www.oii.ox.ac.uk/) and his webpage (http://www.oii.ox.ac.uk/people/?id=165) and blog are definitely worth look for anyone interested in mapping and internet and mobile technologies.