This is the code for Lucas Ritzdorf's 2019-2020 (originally 2018-2019) science fair project.
In 1986, zebra mussel larvae – known as “veligers” – arrived in the Great Lakes from Europe, having been unwittingly pulled into oceangoing ships’ ballast tanks and subsequently released as this ballast was discharged. Since their introduction, they and their close cousins, quagga mussels, have spread to 29 states (according to the National Wildlife Federation) and doubled the water clarity of the Great Lakes. The latter effect may, at first glance, seem to be an improvement. However, clearer water means less plankton for native fish to eat, as well as increased sunlight penetration, which creates ideal conditions for potentially toxic algal blooms. On their own, these effects would be negative enough, but there are more striking consequences as well. Six-inch pipes become completely clogged and beaches are coated with tiny, razor-sharp shells, affecting economic sectors as diverse as power generation and tourism. Furthermore, Montana’s position at the headwaters of the Columbia River Basin makes it highly likely that, if our state were to contract an infestation, the remainder of this unique ecosystem would shortly follow suit.
Understandably, a large amount of research has already been done on the bivalves, their environmental effects, and the ways in which they spread. Strayer (1991), for example, analyzed the possible distribution of the mussels in North America based on temperature and water hardness. Also, Mackie et al. (1996) investigated the calcium concentrations for which quagga mussels could continue to grow, as well as the pH necessary for effective reproduction. Constraints such as these are critical when evaluating the suitability of lakes to infestation by invasive species.
It should not be surprising that, when mussels were detected in Tiber Reservoir, Montana’s government mandated inspections for watercraft passing predetermined stations and began advertising the use of the “Clean, Drain, Dry” method – boats are washed off and cleaned out, drained of their ballast and live-well water, and dried completely. However, little has been done in the way of predictive analysis, potentially an extremely valuable protection measure. Using an accurate predictive model, the most at-risk waterways, as well as “key lakes” which, if infested, would enable the spread of mussels over a large area, can be better identified and protected.
Accordingly, this study aims to create an accurate, computerized risk-assessment model for invasive mussel infestations of Montana lakes, as well as a quantitative simulation of mussel spread through the state. The latter functionality, when combined with travel route information for trailered boats, also enables determination of the optimal arrangement of watercraft inspection stations to intercept the highest number of contaminated boats. The model is extensible to other water systems simply by the usage of corresponding input data. Such a model would be a useful tool for Montana’s administrative agencies, as well as those of other states facing mussel infestation. It would allow the identification of both high- and low-risk lakes, and resources – mobile inspection stations, routine eDNA testing, warning signs, etc. – could be more efficiently distributed.
These scripts were written for Python 3.8, but should continue to work for future versions.