Date of Award




Document Type

Master's Thesis

Degree Name

Master of Science (MS)


Department of Geography and Planning

Content Description

1 online resource (xiii, 69 pages) : illustrations (some color)

Dissertation/Thesis Chair

George Robinson

Committee Members

Steven Pearson, Alexander Buyantuev


Invasive plants, Aquatic plants

Subject Categories

Biology | Public Policy


Aquatic invasive plants can often be ecologically devastating to ecosystems where they are introduced. This devastation has been widely apparent in the United States, including many locations in New York State over many decades, following anthropogenic transport of species from their native ranges. Due to the difficulty involved in eradicating these plants once they have become fully established, there is keen interest in strategies for early detection to assist with prevention and monitoring to mitigate future management costs. Invasive species detection research aims to find methods for detecting new invaders more effectively. I evaluated three detection methods for their effectiveness and efficiency in locating new aquatic invasive species. The first method, a subjective visual survey, consists of observations made by following shorelines, identifying, and collecting evidence of all aquatic plant species accessible. It involves the least time, training and equipment, and can be performed independently by a citizen scientist. The second method is a full lake point-intercept survey, which requires creating a map of grid points (typically 50m apart) and determining GPS coordinates for each point. These points are then reached with a boat, and sampled using a sampling rake, which consists of two rake heads attached together with the teeth facing outward and tied to the end of a rope. The rake heads are tossed from both sides of the watercraft at each sampling point, and the specimens are collected, identified, and quantified. A third technique is a stratified point-intercept rake-toss, which aims to streamline the full lake survey by focusing only on certain points suspected to contain plant species of interest. The stratification pattern is determined based on biotic and abiotic features known to correspond with invasive plant presence, such as boat launches and beds of native vegetation. In many cases, this latter method is preferred, due to its potential to strike an efficient balance between effort and rigor, however there are no widely agreed upon protocols for stratified point selection. The primary objective of my research was to test methods of survey stratification aimed to maximize species detection relative to the amount of effort required to perform a survey. Greater efficiency will permit more frequent surveys over larger numbers of water bodies. To achieve the objectives of my research I employed a combination of fine-scale field studies and analyses of records from larger surveys performed across the state. Four small local lakes (<12.5 ha) were sampled with both a shoreline visual survey and a full lake point-intercept survey between July and September 2017. Mapped results were examined for patterns on which to base post-hoc stratified point selection and compared to random draws. I applied similar techniques to a collection of datasets from six larger lakes that were previously surveyed within the past decade, some on multiple occasions, in collaboration with NYS DEC scientists. I had aimed to demonstrate generally effective means of sample point reduction by testing detection limits for subsets of full lake point-intercept surveys. This process was guided by relative detection success with different varieties of stratification. Point selection was adjusted according to pre-determined habitat information, sequences of previous grid surveys when available, and subjective visual surveys for the same waterbodies. The results did not identify a generally reliable means of stratification to detect aquatic invasive plant species more rapidly. After examining whether wind direction and access points consistently correlated with the presence of target aquatic plant species, it was concluded that these were not dependable indicators. The abbreviated tests were not able to reliably detect 100% of the known species at the study sites, with the wind direction stratum only detecting all species in 66.7% (four out of the six examined) of lakes and only 50% (two out of the four examined) of lakes in the case of the access point stratum. Additionally, invasive plant presence did not seem to correlate with the presence of native plant species or water depth. For example, Pearson correlations between Eurasian watermilfoil (Myriphylum spicatum) and native aquatic species richness were insignificant (Pearson R = 0.003). The results of the tests performed suggest that there may be unique survey strategies that can be used in the management of specific water bodies, but that these strategies are location dependent and cannot be prescribed generally. Other possibilities to explore in future research include focusing on areas of peak light exposure, potentially factoring in water turbidity, as well as examining areas with the heaviest boat traffic, such as known fishing spots, when applicable.