A brief overview of some of the projects I've been developing over the course of my graduate studies.

The 2010 Haitian Earthquake: Simulating Aid

A screenshot taken during a model run, showing the
movement of individuals (red) in pursuit of the aid
distributed from aid centers (blue).
Natural disasters occur all over the world, often in places where infrastructure and even maps are inadequate to support a relief effort for the local population. Recently, researchers’ attention has focused on using crowds of volunteers to help map the damaged infrastructure and devastation caused by natural disasters, such as those in Haiti and Pakistan. This data is extremely useful, as it is allows us to assess damage and thus aid the distribution of relief, but it tells us little about how the people in such areas will react to the devastation. I've built an exploratory spatially explicit agent-based model to study the problem. The model is created using crowdsourced geographic information and other sources of publicly available data, which can be used to study the aftermath of a catastrophic event. The specific case modelled here is the Haiti earthquake of January 2010. Crowdsourced data is used to build the initial populations of people affected by the event, to construct their environment, and to set their needs based on the damage to buildings. The foci of the model are the reaction of people within the simulation to information about the distribution of aid and their navigation of the existing transportation network. My hope, and the hope of my collaborators, is to produce a model which could potentially provide a link between socio-cultural information about the people affected and the relevant humanitarian relief organizations.

This project is discussed further in a paper Andrew Crooks and I have had published, available here. For more information about this project, please see here.

Hurricane Sandy and Sentiment

A composite view of all of the sentiment-laden tweets
 produced in the New York area over the course of
Hurricane Sandy

I worked as part of a team to download and visualize geospatially explicit tweets, as well as the negativity/positivity of the messages being tweeted. I was specifically in charge of processing the tweets and analyzing their sentiment content, as well as mapping the tweets and their sentiment. Part of the goal of the project was to analyze the differences between tweets that were geocoded explicitly by the interface through which the tweet was uploaded (e.g. a cellphone, especially iPhones) and those where locational information was provided by the user profile as part of an effort to understand the representativeness of the geocoded population.

The 2012 Colorado Wildfires and the Evacuation of Colorado Springs 

A graph tracking the number of tweets generated
over the course of the Waldo Canyon Wildfire, with the
number of explicitly geotagged tweets highlighted.

The Colorado wildfires of 2012, and in particular the evacuation of Colorado Springs in response to the Waldo Canyon fire, are an example of the kind of complicated environmental and behavioral interactions that define many crisis scenarios. During a crisis, nuanced and heterogeneous behaviors are contextualized within physical space, making it important for responders to synthesize various kinds of information in order to understand the system. During the Colorado Springs evacuation, individual citizens responded to information about the location and movement of fires, making choices based on their own location, resources, and knowledge. To better project how individuals might respond to such fires, I developed a simulation which couples a simple model of forest fires with an agent-based model of citizen response. The model utilizes ambient geographic information to inform agent knowledge and behavior, and represents an example of how aid workers might use such publicly available sources of information in the future to better respond to developing crises.

Sustainability and Traditional Agriculture: Preserving Acequias

A land-use map of the study area used in the
model: Taos, New Mexico. Used to initialize
the model and compare against various
potential future land use patterns.

Water management is an increasingly contentious issue in the American southwest region, not to mention portions of northern China, the Middle East, and Africa. Especially in the southwest, growing populations are stressing local water resources as a result of both consumption and agriculture. Irrigation based agriculture draws especially heavily on these resources and usually cannot survive without them, but irrigation itself is a large-scale coordination problem. Irrigation systems span the properties of many individuals, some of whom do not clearly benefit from them, and furthermore must be maintained, a task individual agriculturalists cannot bear alone. I have taken the lead in developing an agent-based model to investigate the significant interaction and cumulative impact of the physical water system, local social and institutional structures which regulate water use, and the real estate market on the sustainability of traditional farming as a lifestyle in the northern New Mexico area. The regional term for the coupled social organization and physical system of irrigation is “acequias”. The model can be used to explore the sustainability and rigor of the system in light of changing patterns of weather,  crop prices, and government regulations, and can be extended to explore many other questions.

This project is the subject of a paper Andrew Crooks and I have had published, available here. For more information about this project, please see here

GeoMason: Extending the Capabilities of the MASON Toolkit

I've developed a number of the sample models for the GeoMason.

No comments:

Post a Comment