Steve Condie, PhD; Illinois Mathematics and Science Academy
There is a natural inclination for humanity to view the Earth as our steadfast, never-changing home. After all, it seems as permanent as the ground beneath our feet. However, nothing could be further from the truth, especially in today’s rapidly changing climes. Global temperatures are on the rise, and with them comes an increase in sea levels, wildfires, and hurricanes, three serious, expensive issues. Unfortunately for the National Park Service (NPS), who guard ninety-seven coastal units, coastlines are especially vulnerable to such problems. Thus, the NPS must take extra precautions and expenditures to maintain and protect them. With limited budget, however, a critical problem presents itself: where exactly should these efforts be focused?
To answer this question, our group developed a model addressing the issue of rising sea levels in the future for all five coastal parks in question: Acadia National Park, Cape Hatteras National Seashore, Kenai Fjords National Park, Olympic National Park, and Padre Island National Seashore. Our model determines “high”, “medium”, and “low” thresholds for sea levels by drawing lines one-half standard deviation above and one-half below the mean, dividing the data roughly into thirds. It then uses an exponential moving average five years around every data point, which weighs sea level fluctuations as they occur. We linearly extrapolated the data 10, 20, 50, and 100 years into the future, and found that our results correlated extensively with existing research of ocean current behavior, confirming to us that results had sound basis in reality.
We were also tasked with the creation of a climate-related vulnerability model to relay to the NPS through a scoring system the risk of climate-influenced damage to any given coastal unit over the next 50 years. The formula of our scoring system was inspired by an existing Climate Vulnerability Index (CVI). Although its name indicates similarity to own model, its purpose is in fact predicting people’s vulnerability to the climate. First of all, we changed the variables to four specific climate-related events found to be highly influential monetarily and in some instances, through lives lost, to National Parks and the nation as a whole: tropical storms, wildfires, lowering of air quality, and sea level. We then converted each of these events into scorable quantities, and through analysis of the past frequency and severity of these events, extrapolated the variable scores into the future using our own adapted model. These final scores would be beneficial to the NPS in determining the vulnerability of any given coastal unit to climate-related damage and expenditure many years into the future.
Finally, we were tasked with recommending the NPS of where funding should be allotted to repair climate-related damage in the future, incorporating our vulnerability model. Aware of the Service’s sometimes limited financial resources, we incorporated visitor statistics into our recommendation model, to help them serve the greatest possible amount of people. Through a machine-learning algorithm, we incorporated all four variables present in our climate-related vulnerability model and correlated them with visitor count data over the past twenty years. The artificial intelligence program then built a predictive algorithm from the data to predict visitor count, which was a Gaussian process featuring a numerical vector, which was refined through repeated testing to 77% accuracy. Based on this algorithm, for the first ten years, the NPS should prioritize funding to Cape Hatteras, at twenty years should prioritize Acadia, and finally, at 50, should invest most heavily in Padre Island. The Service should certainly take these predictions into consideration to best honor the beautiful areas that they are charged to protect.
Gupta, Abhay '18; Wei, James '18; Kovach, Istvan '18; Sai, Shyam '18; and Hong, Darius '18, "2017: "From Sea to Shining Sea: Looking Ahead with the National Park Service"" (2017). Distinguished Student Work. 7.