Computers, Society, and Nature

This goes along with a previous post on the application of GIS to tennis. From a student in the Intro to GIS course:

In an attempt to reach Americans as well as the few ashamed Canadian baseball fanatics, I examined Michael Lewis’ “Moneyball”. In this top-selling sports/business hybrid novel, Lewis writes about the elusiveness of an accurate way to measure defense. The typical statistic used to gauge a player’s defensive skill is his number of errors, meaning the mistakes a player makes throughout the season. However, Lewis challenges this simplification of defense, reasoning that a less fleet-footed athlete might not even be afforded the opportunity to make some mistakes due to physical incapability to even get to the ball.

The solution to this conundrum, Lewis says, is a new breakdown of defense, in which the entire field is broken into smaller units of areas, known to GIS people as the implementation of the raster model. With this type of structure, every ball hit to any fielder can be documented with set of geographic coordinates. The fielder’s position prior to the ball being put in play is also noted. Both the velocity with which the ball travels, as well as its trajectory, is also tagged to positional location as attribute data.

The Oakland Athletics, a major league baseball team, have implemented these tactics, along with other scientific methods, and have experienced tremendous success. With one of the smallest budgets in baseball, they have used GIS to efficiently spend their meager funds to acquire players that provide the most production for the amount of money they are willing to pay. Over the past five years, the Oakland Athletics have been among the league leaders in victories per season despite their inescapable “poverty”.

Being from New York, I constantly have it drilled into my head that the Yankees are the team to beat and that rooting for anyone else is futile. I despise the Yankees. Where is the fun in rooting for a team that’s expected to win because of the throngs of hundred million dollar contracts they hand out on a yearly basis? The Oakland A’s, however, have successfully shown that money can’t buy everything. The Yankees are Goliath, and the A’s are David, with their heads buried in science books. The Yankees continue to chase their own tails, failing to win a World Series since their payroll ballooned to over $200 million, compared to Oakland’s $40 million. Baseball may not look like a battle of intellect, but behind the scenes brilliant minds (many of whom are experts in GIS) are quickly gaining respect and snatching up all the jobs previously bestowed upon those who were said to have an immeasurable “baseball sense”. It appears that what you need to win at baseball is not the biggest muscles, but rather some thick coke-bottle glasses, a McGill diploma, and a computer loaded with ArcGIS 9.

So, to all my Canadian friends: before you write baseball off as a useless “American Pastime”, you should realize that there is a growing market demand in this sport for those who have skills in GIS. Major League Baseball is accepting applications now.

References: Lewis, Michael. Moneyball: The Art of Winning an Unfair Game. W.W.
Norton and Company, Inc., New York, 2003.

Thanks to Mongoose girl for this post:

The study of the spread of diseases has tended to focus primarily on human suffering and mortality. In many cases, animals were the vectors of transmission for the virus or bacteria to humans. However, a study from the journal of Emerging Infectious Diseases is attempting to assess human’s negative impact on animals’ health. This study utilized GIS to research the spread of Mycobacterium tuberculosis in free-ranging mongooses in Botswana and suricats in South Africa (also known as meercats).

In these areas of Africa eco-tourism has emerged as a profitable way to support local economies. It has many positive attributes such as encouraging local communities’ autonomy, preserving wildlife habitat, and encouraging responsible resource use. However, increased human presence in what used to be remote and fairly undisturbed areas has resulted in disease transmission to animal populations. This is one of the first studies to consider the transmission of a primarily human pathogen into free-ranging wildlife.

Researchers required a tool to assist them in tracking the geographic locations infected animals as they roamed throughout the day. Researchers also needed to map all of the human infrastructures in the animal’s range, which included national parks and multi-purpose lands. GIS was an ideal solution as it allowed them to overlay multiple layers and look for geographic similarities between human presence and sites of disease transmission. Point layers were used to depict tourist facilities, garbage dumps, and locations of mongoose TB cases. Lines and polygons were used to portray roads and land use. To monitor the disease outbreak, mongoose troops were followed by patrollers in both the morning and evening. They collected and then georeferenced information such as animal sightings, important geographic locations and individual TB cases. This created enough graphical and non-graphical attribute data to determine the rate of infection spread, by calculating the time and distance between new outbreaks. The GIS output was also useful in visualizing the extent of TB’s spread within the mongoose populations and the humans’ role in transmission.

GIS technology has allowed for the domains of epidemiology, geography and wildlife biology to be incorporated into one analysis. Future emphasis can be placed on protecting wild species, either by limiting visitors contact or by simply ensuring that garbage is hygienically disposed of. Ecotourism plays an important role in supporting rural communities. However, animal health cannot be neglected. After all, there needs to be something wild left to encourage visitors to come!

For more information see: Alexander, K., Pleydell, E., Williams, M., Lane, E., Nyange, J. and Michel, A. 2002. Mycobacterium tuberculosis: An Emerging Disease of Free-Ranging Wildlife. Emerging Infectious Diseases. 8: 598-601. Accessed November 25th, 2005.

A post from Solizma in the Intro GIS course.

Baystate Medical Center in Springfield, MA uses GIS applications to treat illnesses, address medical issues, and analyze health abnormalities. The only hospital with a full-time GIS staff, Baystate initiated its health geographics program in 1998. Since then they have been awarded the ESRI “Special Achievement in GIS” Award in 2002 and ESRI “Vision” Award in 2004 at ESRI’s International Health GIS Conference. (The latest conference proceedings are here.)

Baystate’s GIS staff have experience in demographics; cancer, cardiovascular disease, and injury epidemiology; hospital facilities; non-hospital facility siting; healthcare marketing; bioterrorism; and emergency preparedness, planning and support. Current applications vary in scale from single human organs to the whole hospital to multiple states.

The medical center staff are developing a “four-tiered conceptual model for hospital surge capacity planning and response” which they have named the “Healthcare Preparedness Infosphere (HPI).” This GIS-based model is made up of four “health information systems” that provide hospital healthcare, situational awareness, incident management, and decision support at multiple levels ranging from the individual hospital to multiple states. The model allows tracking of patients, resources, and assets, and it can be used to support improved
healthcare and quality. It is applicable to both routine situations or emergencies and disasters.

Another current initiative of the GIS unit at the medical center is flu tracking: staff are mapping the historic geographic distribution of flu and pneumonia patients to identify areas of higher incidence. Staff will use this information to plan vaccination clinics and educational activities accordingly.

The “Rays of Hope” breast cancer program used US Census data to determine areas of breast cancer screening nocompliance by looking at geographic areas with a high advanced-stage-to-case ratio (ASCR). That is, they looked for areas with a high proportion of advanced stage breast cancer cases out of the total number of cases, which indicates that the cases are not getting detected as early as they could or should. The screening noncompliance areas were identified by spatial analysis and their demographic characterisics were evaluated. Based on these results, researchers were able to design screening programs to target high-risk areas, allowing optimal allocation of resources and a maximization of screening yield.

[Of course, the assumption above is that this is a geographic phenomenon when instead it may be more strongly correlated to poverty or availability of health insurance–Sieber]

At a much finer scale, Baystate staff are using GIS to analyze results of Transanal Endoscopic Microsurgery (TEM). This surgical procedure requires parallel positioning of instruments in a 4 by 20 cm long rectoscope. Researchers want to determine whether location of the polyp requiring operation in the rectum correlates with difficulty in performing the surgery. A cylindrical coordinate system and topology is being used as a basis for both 2-D and 3-D visualization and analysis. Findings indicate that polyp location may be relevant to the
limitations of the surgery.

Other initiatives and applications at Baystate include hospital mapping, route optimization management for delivery trucks and drivers serving home-bound patients, trauma surveillance, and development of a regional geodatabase to provide basemap support for regional emergency preparedness, planning, response, recovery, and hazard vulnerability assessment. See the links below for more information.

Overview of Baystate’s Health Geographics Program

Baystate Health

ESRI: Medical Center Improves Community Programs with GIS

Thanks to Simon in the Intro GIS course.

The term gerrymander was coined in 1812 after the governor of Massachusetts, Elbridge Gerry. The term applies to the process by which political districts are reorganized to weigh voting in favor of the dominant political party. In this way the opposition is concentrated into few districts or the minority strength is reduced and diluted over many districts. Gerrymandering is a commonly used legal practice in U.S. states to influence the voting outcome.

The spatial and attribute data supplied by users of GIS now makes the process of gerrymandering easier than ever. With a click of a mouse district boundaries can be remapped according to racial, household income or polling statistics, to name a few. Whereas GIS can be used to ensure fair redistricting, it also facilitates redistricting on a basis of political power.

Of particular concern to me as an environmentalist are the effects of gerrymandering on environmental policies. Gerrymandering not only redistributes voter opinion in unwieldy patterns over counties, but stretches districts over multiple distinct bioregions. How can voters adequately express their opinions on key environmental issues when they are clumped into the same district as other regions that have completely separate environmental concerns?

Without a formal template to assign districts, GIS provides politicians with a tool to perform increasingly sophisticated analyses on voting behavior and assign districts based on a desired outcome. One solution to this problem is to assign districts to watersheds or ecoregions. In this way GIS could be used to define districts based on bioregions rather than voting behavior. The outcome would be a reasonable redistricting system, where the inhabitants of each bioregion could express their opinions on related environmental issues.

For more information of gerrymandering, visit Fair Vote: Program for Representative Government.

Thanks to Enrico from the intro GIS course for this post.

Here’s an interesting use of GIS for history/background information about natural disasters on National Geographic’s website.

It is a spatial application made using Macromedia Flash, so it has vector qualities and doesn’t take as long as ArcGIS to restructure all of the entities when changing the map extent. Once the Flash is loaded, everything is pretty much ready to go. The “Navigator” on the bottom right functions as the inset map that indicates where you’re zoomed in on and also as a means of moving the map extent within North America.

One also can view ALL dates of selected disasters or choose any increment of 25 years by ticking a box and sliding a rule along the bottom timeline.

When rolling over a natural disaster (a point), a small window pops up nearby to give quick facts. For tornatoes and hurricanes, their paths (represented by polygons and a lines, respectively) are highlighted when the mouse is over their point. Larger disasters allow you to click on them to bring up detailed information. These attributes are videos, photos with captions, and large text blocks of information.

There is a nice color scheme that makes it easy to distinguish between types of “forces of nature,” and the scheme is continued when following the links of the larger disasters. Each type of disaster also has a different shape for its “point” representation on the map, making it even easier to distinguish among them.

This application is useful in understanding the spatial trends of hurricane and tornado origins and paths as well as seeing evidence of volcanoes and earthquakes along the “ring of fire.” This example of GIS can be used for public general interest or as a fun way to educate students about the forces of nature.

Thanks to Andrea for this post.

Public health needs are most often at the forefront of international discussion. Disease outbreaks need to be readily dealt with, as to minimize effects. The tragedy of epidemics is to great to risk. Proper planning, with up-to-date and useful information, is needed in any effective decision-making process. GIS has become an invaluable tool to address public health needs.

With the development of the Public Health Mapping and GIS program in 1993, the World Health Organization can now map disease outbreaks, assess epidemic risk and analyze epidemiological data. Analysis of spatial data in tabular format often misses details and trends in space. GIS has facilitated decision-making processes by creating a visual representation of disease to look at outbreak control, monitoring and management.

The roles of GIS in WHO’s Public Health Mapping and GIS program include: determining the geographic distribution of disease, analyzing spatial and temporal trends in disease, determining populations at risk and other risk factors, and planning resources, targets, intervention and monitoring needed to mitigate impacts of the disease.

GIS maps diseases outbreaks in relation to social and ecological variables that may factor into the spread of the disease, such as population demographics, the natural environment and existing health services. Visualization of the problem aids in targeting areas of greatest concern, for more effective disease treatment and control.

Mapping of disease is not a new thing, but GIS can do this faster than ever before. The speed at with GIS can map the distribution of disease and the social and ecological that may play a role in the spread of the disease can get aid to the areas its most needed, fast. There is great hope in the fact that GIS can target areas that lack the health resources to deal with disease outbreak, making aid more readily available to those who do not have the basic, required health resources and to those in areas of high risk.

WHO’s HealthMapper describes applications of the Public Health Mapping and GIS program. Over 500 people in 70 countries have been trained to use GIS software; this has resulted in the mapping of HIV/AIDS risk worldwide, malaria spread in Ethiopia and post-tsunami relief in Indonesia. The visual representation of disease leads, not only to better and more prompt decision-making processes in disease control, but also to better public awareness and understanding of epidemics.

Check out WHO’s Global Atlas for interactive maps on disease distributions. The site also contains excellent encyclopedic descriptions of the use of GIS for health mapping.

Thanks to huds for this post about Dental Topography and Food Deserts: The Role of geographic information systems (GIS) in our Diets

A more scrumptious application of GIS technology allows food scientists to not only “map out our appetites”, give the reasons for why we eat what we eat, but also gives us insight into what our ancestors ate!

At the University of Arkansas, a professor and his team of researchers managed to create GIS based methods to examine fossil teeth to help extract diet information. A combination of GIS software and laser technology gives insight into a dramatic shift in anatomical time, from more herbivorous habits to meat eating.

Why teeth?
“Teeth are perfect for testing diet hypotheses, because they are the best preserved items in the fossil record and are part of the digestive system,” said Peter Ungar, professor of anthropology. “But until now, we haven’t had the technology to pull much information out of them.”

That’s where GIS come in. Teeth shape can tell us what the initial chewing design was capable of, and the “wear and tear” gives clues into food habits and textures. With modern day benchmarks, the research team can get a pretty good idea of what we were all consuming back in the day.

Professor Ungar looked to technology to avoid time consuming manual analysis and gain better, more accurate results. “Dental landscapes” were examined by a high-resolution laser that reads three-dimensional coordinates of the teeth along the surface, which is coupled to GIS software that then calculates them and produces a 3-D map of one tooth.
The team’s analysis showed that Australopithecus afarensis had shallow slopes on their teeth, suggesting a diet of brittle foods like nuts, seeds, roots and tubers, while the teeth of early humans showed steeper slopes with greater shearing power, suggesting a dietary shift to tough meats.

Widely accepted archeological evidence argues the consumption of meat by early human ancestors, but it hasn’t been until Professor Ungar’s research that these dietary hypotheses can be supported!

To analyze more recent food habits, scientists at the UW College of Architecture and Urban Planning in collaboration with the UW School of Public Health have been able to map out the relationships between local environments and health to give insight into why certain population diets vary across regional settings.

Based on data collected via telephone surveys, the team links poor, unhealthy nutrition to socioeconomic status, spending power, and residential transportation accessibility, GIS based environmental variables were measured to highlight distances to fast food joints as opposed to health food stores, annual income and health food costs, and walking/exercise space in neighborhoods.

“The use of GIS offers many exciting ways to map the health-enhancing dimensions of neighborhoods”, the team insists.

I could eat to that!

See
Fat Neighborhoods: Spatial Epidemiolgy Meets Urban Form

Diet Information from Fossil Teeth

Thanks to “a tree lover” in the introductory GIS course for this post.

The Smithsonian National Zoological Park has an informative and interactive website concerning the use of GIS in conservation biology. Of particular interest, is a project called World forests: Biases in forest protection across world biomes. This project aims to assess the decline and protection of forests across the major biomes – temperate, boreal and tropical. The GIS application they created contained layers such as the estimate of the original coverage of forest, the current protected areas, and the modern forested area.

With this application they could then answers 3 questions :

1. Did deforestation in past centuries differ among major global biomes-the boreal, temperate, and tropics? The GIS allowed them to determine that temperate forests declined the most (by 65%), followed by tropical forests (45%) and then by boreal forests (13%).
2. How much forest remains in boreal, temperate and tropic zones? Of the total remaining forests in the world, 51% are in the tropical zones, 45% in the boreal zone and only 4% in the temperate zones.
3. What is the degree of protection in these biomes relative to the degree of threat? Less than 5% of the remaining temperate forests are currently protected, for boreal forests the proportion is less than 4%, and for tropical forest more than 15%.

It is possible to see the resulting GIS application and even to use some of the tools of GIS (e.g., identify, query, and measure) with the conservation atlas of the Smithsonian Institute. We can see from these GIS results that more efforts must be directed towards protecting areas of the temperate forests.