Clark University Research
950 Main Street • Worcester, MA 01610
Tel: 508-793-7711 •

Active Learning and Research
Active Learning and Research
Environmental analyst Gil Pontius and his students use computers and geographic software to simulate and make predictions about the environmental impact of land use change.

Meet the researchers: An opportunity for personalized learning

Interview with Diana Huffacker and Emily Shusas
In a recent interview, Diana Huffaker '03 (a transfer student from the College of the Atlantic) and Emily Shusas '02 discussed their participation in Gil Pontius's research on land use change in Massachusetts. Diana is a Geography major and hopes to focus on land use change for her senior thesis. Emily finished her undergraduate degree with a major in Environmental Science and Policy and is now a student in Clark's fifth-year free master's degree program.

Diana, tell me how you got involved in research here at Clark.

The first semester of my junior year I took Quantitative and Computer Methods with Gil Pontius. I enjoyed the class, and learned a lot, so when he asked if any of the undergrads in the class were interested in working with him further on the Ipswich River Watershed Project, I went to talk to him about it. I had worked with professors before, at the college I transferred from, and I was happy to able to continue doing so at Clark.

What has that involved?

The Ipswich watershed has been the subject of joint research between the Marine Biological Laboratory at Woods Hole [Massachusetts] and Clark for the last couple of years. Clark is helping the Lab to study the long-term impacts of land use on water quality. We are interested in nitrogen-loading in ground water as a result of long-term, heavy fertilizer use. To understand that, you need to know what the landscape looked like in the past. My contribution this past summer was to model historical land use; we created a map of how we think the landscape looked in the watershed in 1951. We can also estimate how accurate our projection is.

What did you use to do the 1951 simulation?

We used Idrisi GIS (geographic information system) software developed here at Clark. The next Idrisi release will contain a land use change module called GEOMOD that Professor Pontius has been working on. The data that we used came from the US Census. We used census data to determine how old subdivisions are. The land use data was from Massachusetts Geographic Information System [MassGIS]. We also have historical data on land cover from the towns in the Ipswich Watershed.

First we ran the GEOMOD simulation to create a map of 1971. We compared this simulated map with the actual land cover map of 1971 from MassGIS to check how well the simulation preformed. Then we ran GEOMOD to create a map of 1951, which we then adjusted to show the level of certainty we have. We don't have an actual map of 1951, so we cannot compare our simulated map to the real thing. Each simulated map was created using a "suitability" map developed from census and land use data. Each suitability map represented our best guess on establishing where deforestation was taking place.

What were the criteria for your suitability maps?

First we compared land use categories to areas that had been deforested. In the Ipswich Watershed, the general trend is that forest gives way to residential development; essentially, the suburbs are eating away at the forest. We ranked the land use categories according to the likelihood that they'd become deforested (that is, had been forested in 1951), with residential at number one, and then we also added census data to help differentiate between old residential areas and newer developments. This map, along with the initial maps and a regions map, are all used by GEOMOD to create a simulation map.

Did you have funding for this past summer?

Yes, it came from a National Science Foundation Research Experience for Undergraduates grant through the Marine Biological Laboratory.

Emily, am I correct that you received a HERO fellowship as an undergraduate?

I worked with the HERO program my sophomore year when I took a class with Billie Turner. We did a class project on automobile emissions that I wrote up and put on the HERO website. I wasn't funded through HERO for that, but the project was related to the human-impact research that the HERO fellows were doing. Then I officially became a HERO fellow the summer before my senior year.

How did you first become involved with research at Clark? Was it the project in Bill Turner's class?


For your HERO fellowship did you continue the work on automobile emissions or did you go on something different?

I changed to something that Gil was working on: developing astatistical method to assess different signals of transition between land cover types within a landscape. We examined the central Massachusetts landscape to see if transitions between land cover types are systematic or random, while holding the persistence within the landscape constant. We looked at land use maps from MassGIS for 1971 and 1999 and there was 90% persistence in the landscape. If you use regular statistical methods (such as a Chi square test), they would indicate that the change is non-random, because 90% of the landscape stays the same. So we took the persistent, non-changing areas out of the picture. We looked at just the areas that did change and examined them to see if the change they experienced still appeared to be systematic, or if it appeared to be more random.

What did you find out? Was it more systematic?

Some of it was. Residential land was definitely taking over forested areas, but since the transition to residential land is usually a permanent change in the landscape, there was no transition of residential land back to forest within the study area. We found some interesting patterns. For example, when forest re-grows, it systematically takes over a lot of open land. However, the amount of re-growth is small relative to the amount of forest that transitions to other land types such as residential, so that the amount of forest in the landscape is decreasing overall.

Did you do a senior thesis after your HERO fellowship?

Yes, on the same topic.

Are you still working with Gil for your master's program?

No. Now I'm working on a project involving the standard for arsenic levels in ground water. Professors Goble, Brown, and Hattis in the ES&P department got some funding from the EPA to carry out analyses on this topic. When EPA was trying to determine a new standard for arsenic levels in drinking water, they hired some consultants to further characterize the risks associated with a variety of standards. Professors Goble, Brown, and Hattis finished about half of the analysis and then the funding got cut off. I'm picking up where they left off.

Are they trying to look at what are acceptable levels for arsenic in drinking water?

Yes. They're actually focusing more on developing methods to evaluate risk based on variations in susceptibility to toxic effects within human populations. Tolerance thresholds for different chemicals vary from person to person, and this variation can be modeled and used to figure out how many people will probably demonstrate a health effect at a given dose of a chemical. We're looking specifically at two non-cancer effects, diabetes and heart disease.

Have you decided on a master's thesis topic yet?

I'm still fishing around. I know it will be within that realm, but I'm going to try to work some GIS into it, maybe mapping out the locations of population groups that are exposed to different concentrations of arsenic in their drinking water.

Can you both comment on what you feel the advantages and disadvantages are of doing research as undergraduates?

Diana: I think it's an opportunity to work one-on-one with a professor, which is good and bad. It's an opportunity for personalized learning and time that you wouldn't get in a classroom. You can also study things that wouldn't otherwise fit into a class. With Gil I can say 'I'm interested in this' or 'can we look at this' and we can go off on a tangent if it fits within the project. The bad part is I have to do my reading and if I haven't he catches me right away (laughter). There's nobody to hide behind. It's a neat experience. It's a fun thing to do. I've definitely learned a lot of things that I would have never learned in a classroom.

Emily, can you add anything to that?

Yes, I think it's really good in terms of having real world experience, because usually whether you even go into research in an academic institution or if you get a job somewhere in the private sector you're going to be working with other people on projects. It's really good in terms of learning to work with people and seeing how much you need to rely on yourself to do something and how comfortable you are with that. How comfortable you are with depending on other people to do certain things for you. And as Diana said, you get to work one-on-one with a professor. But it can be a lot more work than a class because for a class you can kind of think, oh, I won't bother with the reading, but if you're working one on one the professor will definitely know if you're slacking off. That's really the only disadvantage. I would say that research experience is very advantageous. If you're going to go on to get your master's degree, you'll have to do a thesis and it's really good to have had some experience and some sense of what you're in for. I think that as an undergrad professors tend to hold your hand more and give you more personalized attention and guidance because they know that it's a lot to take on while you're still pursuing your undergraduate studies. So you get a little bit more experience and you know what's going to be coming when you take on another project.


Contact Information Site Search

Additional Resources
Search by student
Search by professor
Search by department
Fund it
Present it

Diana Huffaker and Emily Shusas
Emily Shusas and Diana Huffaker

 Real world experience QuickTime

Download software.

© 2014 Clark University·