This update includes over 500 boundaries with several new digitizations. It also includes several boundaries digitized by students that were not available in previous releases. In total, the undergraduates working in the geoLab submitted hundreds of new pieces of boundary data from Spring 2020 to Summer 2021. The undergraduates who submitted to the project are: Lindsey Rogers (‘21, now a CGA fellow at W&M), Joshua Habib (‘21), Sidonie Horn (‘22), Sean Murphy (‘22), Dorian Miller (‘22), Hadley Day (‘23), Lydia Troup (‘23) Dominic Fornatora (‘23), Natalie Spage (‘23), Kristina Pupkiewicz (‘23), Michael Roth (‘23), Caro Rivera (‘23), Charlie Altman (‘22), Isabel Schruer (‘23), Tara McLaughlin, Russ Biddle (‘23), Renee Ritchey (‘24), Emily Topness (‘22), James Turner (‘23), Sam Updike (‘22), Helena Buckman (‘23), Neel Simpson (‘22), Jason Lin (‘22), Sylvia Shea (‘21), and Sydney Fuhrig (‘21).
geoBoundaries is largely staffed and managed by undergraduate students. In the 2020-2021 school year, Sylvia Shea (‘21) and Sydney Fuhrig (‘21) led the project. This year, Lydia Troup (‘23) and Sean Murphy (‘22) transitioned into the co-leadership positions. Team co-lead Sean says “The opportunities for student leadership are unparalleled in other student organizations. We’ve gained skills in human resource management, communication with external partners, and problem solving that can only be learned in practice, not in a classroom.”
For the first time, researchers uploaded their boundaries through a GitHub site. The GitHub sites, designed by Dan Runfola, aimed to streamline and standardize the review process for boundary data. The new website has improved the quality and speed of data collection, and allowed for external research to contribute to our dataset. Lydia Troup (‘23), an undergraduate researcher who worked on the GitHub site, and geoBoundaries co-lead, says “we went from submitting files on a spreadsheet to having a fully functional bot auto-checking each file. The convenience and speed are gamechangers. The system exposes students to a workflow similar to what they will encounter in professional post-graduate settings.”
Boundary data not available online had to be created by hand by undergraduate researchers. Manually digitizing proved challenging but satisfying to student researchers who undertook the task. One student, Renee Ritchey (‘24), digitized several boundaries for 4.0 including Nauru (NRU1), Uruguay (URY2), the Solomon Islands (SLB3), and Latvia (LVA2). Renee described digitizing, saying “When I sit down to digitize, I have a base map open and an overlaid image. Then I manually plot points on the map to match up the reference image.” This process can take dozens of hours depending on how complicated the boundaries are. Despite these challenges, Renee describes the importance of creating these files saying “there are a lot of limitations in licensing, so by putting it in our format, we create a source where anyone can do open analysis with these boundaries.” At geoBoundaries, we believe that it is crucial that academics, researchers, and civilians alike have access to this boundary data without restriction. Ultimately, we hope to improve our field by producing data that is more open and accessible to everyone.
Researchers also created the Humanitarian and Authoritative data sets. The authoritative sets come from the UN SALB databases, and the authoritative data sets contain the boundaries as provided by countries themselves. The opportunity to study authoritative boundary datasets has expanded students’ understanding of boundary conflict. Hadley Day (‘23) says “Authoritarian data sets are created by governments, so that is where the most border disputes become evident because we often have areas that are ‘double-mapped’. This process allowed me to see more clearly what kinds of territories are disputed, and allowed us to create theories on why some land is considered more valuable than others.”
Alternatively, the humanitarian data sets come from NGOs like HDX or HumData which are generally used for aid work. Hadley also offered commentary on the new humanitarian boundaries. “It is also powerful to come to terms with how and why these groups create humanitarian data sets because often they are in response to serious crises that require NGO interventions. Oftentimes, these data sets are dated which can lead to less accurate crisis analysis. I hope that more funding can be provided to these critical mapping organizations in the future so they have the resources to make frequent updates to their data.”
Overall, the geoLab provides unique opportunities to lead, research, and publish, that are otherwise difficult to find in undergraduate settings. Students build technical skills with data analysis and GIS, and enhance their understanding of border conflicts across the world. The Lab gives students the opportunity to work directly with external partners like ESRI, the UN, and the WHO, setting students up for success upon graduation with workforce skills one cannot learn in a classroom setting. Ultimately, the geoLab provides students with excellent research opportunities and in turn those students provide deeply needed research to the field. With that, we are so excited to bring you, finally, geoBoundaries 4.0.
The geoBoundaries 4.0 data can be found on the geoBoundaries website and on the Harvard Dataverse. The corresponding article can be found on PLOS One.
]]>The Global Environment Facility, commonly referred to as the GEF, is an organization that tackles pressing environmental issues and supports community-based and national sustainable development initiatives. In past years, students from the geoLab have worked with the GEF and have co-authored research papers for the organization.
“Moving towards the spring, the GEF offered more opportunities for conducting impactful international research,” Panganiban said. “And that led to the proposal of having four virtual internships related to the environment and GEF mission.”
Beginning in June, Maison, Panganiban, Chen, and Mills researched their four different projects for 40 hours a week. Though under the guidance of GEF, each student was required to pace and manage themselves, only meeting with GEF leadership once per week.
Maison–a recent graduate of the college–worked as the Assistant Program Manager for the geoParsing team’s BRIGHT project. For her summer research project, she worked on a literature review regarding the consequences of human encroachment on natural resources and how these unnatural conditions will lead to more infectious disease outbreaks.
“I applied to GEF to research COVID-19 because the semester before I did labor market research on the implications of COVID on health care access,” Maison said. “I wanted to take a different approach and further my understanding of the pandemic, and I think my work this summer did that. I loved doing the internship. It was enjoyable to lead my own research, as well as meet other people across different teams in the lab.”
Maison believes the knowledge she gained during this internship helped her improve her research on the geoLab’s BRIGHT project.
“I learned a lot about sustainable development goals from the UN, as well as the implications of different environmental interventions,” Maison said. “My research definitely helped me expand my knowledge on sustainability, as well as what contracting for another organization looks like.”
Mills currently works as an analyst for the geoData team and says she was excited to have an internship that revolved around the impacts of COVID-19. Her research utilized satellite imagery and forest change remote sensing data to study the impacts of the COVID-19 pandemic on land degradation and biodiversity.
“I looked at areas specifically within the Democratic Republic of Congo. Along with Brazil and Indonesia, the DRC was a top country in the world for deforestation as of March 2020,” Mills said. “As countries began focusing on lockdown restrictions at the start of COVID, deforestation occurred with decreased regulation. So even as people stay inside, protected areas are at risk of illegal mining and forest clearing.”
In order for Mills to track the high levels of deforestation in the Democratic Republic of Congo (DRC), she learned JavaScript and time series modeling in R over the summer. Prior, Mills had only ever coded in Python.
“It completely changed the way that I worked as a data scientist, and it made me so much more comfortable for classes and other geoLab projects this year,” Mills said. “My experience was just something I feel I would never get in a class.”
Chen is currently on the geoDev team, where she primarily works on the roadRunner project. Her study aimed to quantify the association between GEF interventions and local health conditions of children under five years in Kenya. Chen’s health measures included looking at rates of diarrhea and lower respiratory problems.
“We use the Health Survey data as well as the GPS data,” Chen said. “Then, I put them into a spatial statistical model called QGI, which was developed by Dr. Dan Runfola. And, through that model, we were able to tell the magnitude of the association and also how far away the impact could be. So in the end, we were able to make some recommendations to GEF to help them understand how their projects can impact not only the direct metrics, like the biodiversity or vegetations, but also the socio-economic and health aspect of their projects.”
Chen says she enjoyed being able to use the material she learned in her statistics classes and apply it to real-world problems.
“I learned a lot along the way,” Chen said. “Because in stats classes, you’re just looking at toy models, but in the real world models, there could be a lot of unexpected things happening there. I’m happy to see I’m involved in a project that could have some real-world impact.”
Panganiban, who graduated last Spring, was the former team lead for geoBoundaries but is now working as an accountant at Deloitte. His research analyzed satellite imagery and ancillary data to evaluate the impact of the GEF’s biodiversity projects in the Philippines for the past decades.
Panganiban collected data on the area’s biodiversity by using two metrics: a vegetation index that measured terrestrial biodiversity and fishing catch rates.
“If an area is increasing in fish catch, that means that factors supporting fish are increasing, such as productivity, algae growth, and high amounts of food. If the fish catch is high, then the ecosystem is probably pretty healthy,” Panganiban said. “This was only an analysis based on available data, which is why the interviews with executing agency officials were conducted to verify whether what we’re studying from satellites verifies what’s on the ground. That was the most interesting part.”
Panganiban conducted interviews with Filipino government officials in order to further his collection of data and monitor local sites. With the nation in lockdown, all the interviews were conducted virtually.
“One of the coolest things that I think I ever got to do was hop on Zoom calls and talk with officials from the Department of Environmental Natural Resources and seniors in NGO from different islands who are working remotely,” Panganiban said. “I was pretty much doing the same thing that someone in the Philippines would do, except I’m in the United States.”
In October, each student presented their research over Zoom to members of GEF and the head of the geoLab, Assistant Professor of Applied Science Dan Runfola. Much of the data they collected will be used in GEF’s future reports and papers.
“I think the most important part of that meeting, for me, was talking with people at the GEF about how our findings can impact future studies,” Mills said. “For example, I detected land degradation in one of the protected areas I analyzed. So, I explained that I thought it made sense to go in and do further on-the-ground evaluation to distinguish between seasonal land change versus something more abrupt like deforestation. And they agreed with that. We looked at satellite imagery and saw visual proof of potential deforestation.”
For the upcoming year, Maison, Mills, and Chen hope to use their acquired research skills and collected data to aid with future geoLab projects. All four students praised the geoLab for providing this opportunity, and Chen noted how appreciative she was to have a virtual internship for the summer when she originally expected to be unemployed.
“I think one important thing to note about the geoLab is that if you’re a passionate student, it will always find an opportunity for you,” Panganiban said. “Giving students the opportunity to see their work be used by a large international organization is truly the best gift we can offer.”
]]>“I would say living in Laos really changed everything,” Morin said. “On Sunday mornings, I would go see what the status of construction projects were. That’s just what I did for fun because Laos borders China. It was really transformational for my research being like…this is something I’m really interested in and finally being able to see it.”
Morin has been researching international aid since high school and took a gap year before her freshman year, where she was able to delve more into the topic. When she started at William and Mary, Morin became a research assistant at an on-campus lab that introduced her specifically to Chinese development financing. However, Morin wanted to search for more student-led research initiatives, which she found at the geoLab.
“I spent the summer exploring more topics on Chinese development financing because I was living in Southeast Asia,” Morin said. “And that’s how I met Matt Crittenden. And because he was also in Southeast Asia at the time (at the beginning of sophomore year), we sat down and chatted, and he told me all about the BRIGHT team and what they were doing. And I was like, ‘Oh, this is the exact topic I want to research but from a more student-driven perspective.’”
Matthew Crittenden ’21 is the team lead for geoParsing (one of the four student-led research teams in the geoLab) and introduced Morin to the Belt-Road Initiative Geospatial and Headline Tracking (BRIGHT) team. The Belt and Road Initiative (BRI) is a global infrastructure development plan by the Chinese government that invests in over 70 countries. The BRIGHT team uses satellite imagery to track Chinese development in Latin America and the Caribbean.
The BRIGHT team, which has seven members including Morin, is partnered with the National Geospatial-Intelligence Agency (NGA), a combat support agency under the U.S. Department of Defense. Using NGA-funded commercial imagery, geoParsing is the first undergraduate team the agency has ever paired with. The NGA is also partnered with Columbia University’s international relations graduate program and multiple D.C. think tanks.
“Say we find a port in Cuba, and we read a lot of open-source information about what is happening in the port and what is being constructed,” Morin said. “We can look at the satellite imagery and say ‘OK, this information is wrong,’ or ‘this information is correct.’ Using satellite imagery is a way of going in-depth and holding open-source information accountable because a lot of times, the satellite imagery tells a very different story from what the news articles are publishing.”
Starting January 2020, Morin was tasked with writing the BRIGHT Team’s research on Bahamian ports in a paper, which would be published on the NGA’s platform named “The Tearline Articles.” Finishing the paper in the middle of June after writing approximately 4000 words, the NGA decided not to publish the piece, due to the specific ports falling into a “gray area” of the BRI.
Though Morin described the NGA’s choice as extremely disappointing, she says she and Crittenden were able to learn from the paper’s shortcomings and improve their research methods approach. Starting late June, Morin began writing the next two papers for the Tearline Articles, in which she worked hard to make sure the information was not in gray area and qualified as part of the BRI.
“We still wanted to talk about ports in the Caribbean because the BRI emphasizes transportation infrastructure, which, in the Caribbean, means ports,” Morin said.
The two papers Morin wrote in the latter half of the summer detailed the BRI’s work in seven Caribbean ports and energy infrastructure in Cuba. Due to U.S. sanctions in Cuba and the BRIGHT team’s mission to avoid any further gray areas, Morin recounted how “extremely complicated” the writing process was for the Tearline Articles. At the end of August, the NGA published Morin’s papers.
“It was very surreal….we were on the front page of intel.gov for over a month, which is so cool,” Morin said. “I think I had reached this place where I was like, ‘are they really ever going to get published?’ The hardest part about the papers after the expectations were set was the editing process. Because constantly you’re going back and editing and having to change things.”
For this upcoming semester, Morin is now the program manager of the BRIGHT team and is excited to implement many of her ideas to improve the team’s research process. The NGA recently approved this year’s project, which will involve the BRIGHT team looking into hydropower in South America. Morin hopes to carry out new methods that will improve efficiency and collaboration within the team.
“They’re such fantastic human beings,” Morin said, describing her team. “I think going into this semester, I came in with a lot of ideas and things I wanted to change about the lab. Our publications were phenomenal….But, just because of logistical difficulties, there was so much more time than I guess anyone expected, and Matt and I weren’t anticipating that we were going to work in the summer…but we did. It all worked out, but I want to change how the team interacts with each other.”
]]>