wbowden/Teaching/Risk_Assessment... · Web viewSome of the data that were collected were done so in the spring (e.g. total phosphorus and hardness). ... pH, hardness, water clarity

Photo of Lake Willoughby captured by Farrah Ashe

A Comprehensive Suitability Assessment of Vermont Waterbodies for Spiny Water Flea (Bythotrephes longimanus), Zebra Mussel (Dreissena polymorpha),

and Starry Stonewort (Nitellopsis obtusa)

Authors: Farrah Ashe, Nikki Boudah, Kelsey Colbert, Kait Jones and Will Sutor

Partnered with:

Vermont Department of Environmental ConservationTable of Contents:

Executive Summary Page 2

Introduction Page 2

Background Page 3

Approach Page 3

Findings Page 6

Discussion and Recommendations Page 11

Literature Citations Page 15

Acknowledgements:We would like to thank our partner Josh Mulhollem of the Vermont Department of Environmental Conservation (DEC) for his guidance throughout the course of this project. He provided our team with the waterbody data used in the report and answers to our many questions during the entire process. We thank him for the time he dedicated to us and the resources which he was able to provide for our team.

1

Executive Summary:

According to the Vermont Department of Environmental Conservation (Agency of Natural Resources, 2017), the waterbodies of Vermont are vulnerable to invasion from zebra mussels, spiny water flea, and starry stonewort. There is currently a lack of state-wide knowledge on which waterbodies are hospitable to these three organisms. By understanding this, the state can assess where the greatest needs are, and focus their efforts on prevention and monitoring in those locations. If no action is taken to determine waterbodies at high risk for invasion, more affordable preventative measures will no longer be an option. This is problematic because invasive species can drastically alter ecosystems, resulting in the need of mitigation or eradication efforts further down the road which can cost the state orders of magnitude more than preventative measures or monitoring in the early stages of invasive species establishment (USGS, 2017).

The goal of this project was to conduct a statewide analysis to determine which Vermont waters are at the highest risk of introduction and establishment of these three invasives. This was accomplished by reviewing relevant scholarly literature on the three focal species and the environmental conditions they can tolerate. Tolerance information was then organized to determine high, moderate, and low priority waterbodies at risk for invasion from each invasive species. This data was compiled and used with the vessel travel data to create an ArcGIS map on the state waterbodies for each invasive species. Using these tools, we were able to recommend to the DEC the waterbodies that should be prioritized for monitoring and spread prevention.

Introduction:

Vermont’s waterbodies are currently inhabited by seventeen invasive species and threatened by fifteen more invasives in neighboring watersheds (Agency of Natural Resources, 2016). The spiny water flea, starry stonewort, and zebra mussels are three of the seventeen species currently found in parts of Vermont and all three are expected to continue invading Vermont’s waterbodies. Preventing further spread of invasives is critical to conserving Vermont’s biodiversity, as invasives can lack natural predators and therefore outcompete native species. Spread of invasive species also poses a risk to human health as invasives can clog water pipes and impact the quality of fish and water humans are consuming. It is also important to recognize that most of these aquatic invaders are spread between waterbodies as a direct result of human activity. Known invaders such as the zebra mussel, can easily cling to boats or trailers. They can also be spread into new locations due to a lack of public knowledge around baitfish regulations (Agency of Natural Resources, 2016). Mitigating the negative impact aquatic invasive species impose can be aided with public education but if it is not addressed, more extensive preventative measures must be implemented. The result of the high cost of eradication can place a great fiscal burden on local communities.

The purpose of our project is to therefore determine which waterbodies in Vermont are susceptible to invasion by one or all of these species, and the level of risk for invasion based on water quality characteristics. The goal of this report is to help the reader better understand the risk invasives pose on the areas they are inhabiting, and to emphasize the importance of conducting a statewide analysis to determine which Vermont waterbodies are most at risk for invasion of the spiny water flea, starry stonewort, and zebra mussel.

2

Background:

Studies have shown that nonindigenous species pose a severe threat to the health of waterbodies throughout the state of Vermont (Modley, 2008). The introduction of an invasive species can change food webs by destroying native food sources, alter ecosystem functions such as changing water chemistry, and degrade the habitat quality of other organisms. Furthermore, the presence of a nonnative species can decrease native species biodiversity and even cause localized extinction in an ecosystem (USGS, 2017). As a result, the spread of invasive species throughout the state is not only detrimental to Vermont’s ecology but are also financially costly to mitigate (Clavero and Garcia-Berthou, 2005).

The zebra mussel (Dreissena polymorpha), spiny water flea (Bythotrephes longimanus) and starry stonewort (Nitellopsis obtusa), are three aquatic invasive species that have been identified in waterbodies of Vermont. Each of these invaders has been known to cause harm to the ecosystems that they infiltrate, and therefore present a serious concern to the state. Zebra mussels, for example, can attach themselves to any surface in the water including other native clams and mussels causing asphyxiation and eventually death (Johnson and Padilla, 1996). Due to their aggressiveness, and also effective domination of ecosystems that they invade, zebra mussels have been known to alter the structure and function of waterbodies around the world (Karatayev et al., 2002). According to the USGS Ecological Survey, both the spiny water flea and starry stonewort pose similar ecosystem-altering threats to aquatic ecosystems as the zebra mussel. The spiny water flea has the capacity to completely shift the trophic structure of an ecosystem through its insatiable appetite for crustacean and zooplankton. Such effects have shown significant reductions in these two species, crucial to nutrient cycling within aquatic systems (Kelly, 2013). Similarly the starry stonewort can have negative impacts on fish spawning, foraging, and nesting habitat, altering the ecosystem and abundance of aquatic species (Pullman, 2010; Midwood et al., 2016). As a result, many states have begun to require mitigation and prevention efforts for these three aquatic invaders at the state and local level (e.g. analyses and risk assessments of critical bodies of water) (Kipp et al., 2017).

Vermont has not formulated a comprehensive environmental suitability and vessel-based introduction risk assessment for these three aquatic invasive species. Therefore, a statewide analysis of these three invasives that are currently in and threatening the waters of Vermont is needed. Our assessment will determine the waters in Vermont which are at the highest risk of introduction by vessel travel and establishment in native suitable habitat. This will help the state of Vermont determine which aquatic ecosystems should be prioritized for invasive prevention planning, monitoring, and control measures in the future.

Approach:

We chose to focus on three invasive species for the main portion of this project. Zebra mussels were chosen because of their severe negative impacts in Lake Champlain and Lake Bomoseen in Vermont. Studies have shown that these invaders clog water intake pipes, pose a hazard to recreationalists, and harm the native biota and natural food web of these lakes (Modley, 2008). Due to its natural history, there are large amounts of scientific literature on zebra mussels and the factors controlling their growth, reproduction, and subsequent spread (Cohen, 2017; Feng & Papes, 2017; Karatayev et al., 1998; McMahon, 1996; Wu et al., 2010).

3

We also chose this species from the request of our partner due to these reasons and the need for prioritizing waterbodies to better focus prevention and monitoring efforts. The spiny water flea was chosen because of its recent invasion into Vermont (first observed in Lake Champlain in 2014) (Voorhees et al., 2015). This lack of study poses a problem to the state water managers on understanding the best control and prevention methods to use in Vermont, as well as, understanding the spread paths of the species. Starry stonewort is also a relatively new invasive species to the waters of Vermont found in 2015, with its most recent invasion into the waters of Lake Derby of Derby, VT (VT DEC, 2016). The last two species presented the opportunity to gather a wealth of scholarly information on these species to aid in the knowledge of our partner, the VT DEC, and help in finding limiting factors to their survival in Vermont waters.

In order to best determine the level of risk for invasion of each waterbody in Vermont, information on the habitat requirements of the three species was compiled and assessed. We analyzed the scholarly literature on each of the three species in order to understand which lake characteristics were necessary for survival, growth, and reproduction to determine population establishment if the species were to be introduced into the ecosystem. This included research on water pH, conductivity, salinity, hardness, depth, dissolved oxygen, chlorophyll-a concentration, and total phosphorus, among the other water quality parameters that were pertinent to the required habitat profile of the nonnative species. We then compiled the found tolerance ranges from the literature into a spreadsheet. This allowed for easy comparisons of the lake data we received from our project partner.

Our community partner gave us the lake data in groupings by the environmental parameters. The averages that were not already found in this data were calculated to a daily, yearly, and then overall average for each waterbody. Each of the parameters were set as “meets” or “does not meet” for each of the three species. If the waterbody only had one measurement which was taken to represent the whole, this was noted and marked with an asterisk. The waterbodies that has no data were marked with “N/A”, indicating that the data was not available. Some of the data that were collected were done so in the spring (e.g. total phosphorus and hardness). This was used because spring is one of the times of the year in which dimictic lakes (the lake type of most Vermont lakes) experience turnover and therefore the nutrients that were previously layered throughout the water column are most evenly mixed (Nunberg, 1988). This data was then compiled into one spreadsheet. The risk of an invasion occurring given successful introduction was classified into three different categories: low risk, moderate risk, and high risk. Here we defined this risk being the risk of establishment of the invasive species if introduction were to occur. High risk waterbodies fall within two or more of the environmental tolerance parameters or contain two or more parameters with “meets” for that species. Moderate risk waterbodies contain one environmental tolerance parameter that was met, due to natural lake fluxes. Low risk waterbodies do not have any of the parameters that were met. We also noted the difference between waterbodies with low data (only one year of data) versus those which had multiple years of data collection. These specific low data values are well marked so if they need to be explored further on an individual basis, they can be considering the unknown variance that the data could have with using only one year of data to represent the waterbody.

We utilized vessel travel data from the DEC’s greeter program, a program designed to track boater behavior and inspect boats to better understand the transportation of invasives. The greeter data was listed for two different risk factors: risk from last waterbody and risk from time since in last waterbody. These two factors were first categorized on a 1-3 level of risk (1 being low, 2 being medium, and 3 being high). For the risk of last waterbody: 3=The last waterbody

4

visited contains at least one invasive species that is not currently found at the lake where the inspection occurred. 2=It is unknown if the last waterbody visited contains any AIS of concern consisting mostly of out-of-state waters. 1=The last waterbody visited does not contain any AIS that are not already found in the lake where the inspection occurred. For the time since the last launch the categories were: 3=the watercraft was last used in a lake other than the one where the inspection occurred within the last 5 days. 2=the watercraft was last used in a lake other than the one where the inspection occurred within the last 2 weeks but not within the last 5 days, or, it is unknown when the watercraft was last used. 1= the watercraft was not used in a lake other than the one where the inspection occurred within the last two weeks. This data was then summed and recategorized as 2-3 being low risk, 4 being medium risk, and 5-6 being high risk vessels launching. This data was then used to create an ArcGIS shapefile and subsequently a map. State waterbody data (VT Priority Lake/Pond data from vtanrgis public access data) were added to a map of Vermont. The calculated risk percentages attribute was then chosen to be represented as a pie chart for the three risk categories. This map was created to provide a visual to aid in the understanding of the overall risk including the boat travel introductory risk in addition to our environmental suitability establishment risk.

5

Findings:

Our team created a spreadsheet which contains the environmental parameters and the waterbodies of Vermont listed by name and location (by longitude and latitude). In this spreadsheet, chlorophyll-a concentrations, pH, alkalinity, hardness, and total phosphorus concentrations are listed (refer to Table 1 for a sample of the chlorophyll-a data).

Lake_ID Chlorophyll-a limits in the environment for starry stonewortAMHERST meetsARROWHEAD MOUNTAIN meetsBEEBE (HUBDTN) meetsBERLIN N/AMEMPHREMAGOG meetsMETCALF meetsNINEVAH meetsSHELBURNE does not meetSILVER (BARNRD) meetsSOUTH (EDEN) meetsSOUTH BAY meetsSPRING (SHRWBY) meetsST. CATHERINE meetsSTAR meetsSTRATTON meetsSUNRISE meets*SUNSET (BENSON) meetsSUNSET (BRKFLD) meetsTICKLENAKED meetsVALLEY meetsWAPANACKI meetsWILLOUGHBY meetsWINONA meets*WOODWARD meets

We chose not to include dissolved oxygen as one of the final environmental parameters, because the dissolved oxygen varies with the water depth and the lake layers (Rao et al., 2008). The hypolimnion could be anoxic, while the other more shallower layers are still able to contain

6

Table 1: This is a sample table of the type of results that were obtained from the environmental parameters for each of the three invasive species. Here we are seeing the partial results on whether or not starry stonewort could invade each waterbody. “Meets” specifies that the concentrations of chlorophyll-a found in that waterbody are allowable if starry stonewort were to be introduced to this waterbody. An asterisk indicates that the chlorophyll-a concentration measurements are only based on one year’s data. “Does not meet” indicates that the specific waterbodies are not capable of hosting starry stonewort due to the concentrations of chlorophyll-a. “N/A” means that

sufficient oxygen levels to sustain growth and reproduction of the invasive species. For example, despite their temperature sensitivity, spiny water flea will move vertically in the water column to find the optimal conditions suitable for dissolved oxygen concentration. Table 2 contains a list of the Vermont waterbodies that are stressed due to low dissolved oxygen levels. This omission is unlikely to affect our data, but in the future this parameter could be significant. Especially if there is a moderate match in suitability, where one environmental parameter is not met while others are.

Waterbody Name Town Latitude LongitudeLake Eden Eden 44.72 72.5Elfin Lake Wallingford 43.47 72.99Ewell Pond Peacham 44.37 72.17Fern Lake Leicester 43.87 73.07Great Hosmer Pond Craftsbury 44.7 72.37Lake Parker Glover 44.72 72.23Sabin Pond (Woodbury Lake) Calais 44.4 72.42Shelburne Pond Shelburne 44.38 73.17Spring Lake (Shrewsbury Pond) Shrewsbury 43.5 72.92Stiles Pond Waterford 44.42 71.93Ticklenaked Pond Ryegate 44.18 72.1Valley Lake (Dog Pond) Woodbury 44.43 72.43Walker Pond Hubbardton 43.74 73.14

Furthermore, it is important to note that depth was not assessed in our study because of its variability within a lake’s profile. While we did acquire the maximum and mean depths of each waterbody studied from the state, it wasn’t appropriate for any of our focal species. Depth did not impact the spiny water flea, because conditions in the water column vary and the species is able to move vertically. We also found that depth wasn’t appropriate for starry stonewort and zebra mussel, because both species inhabit shallow waters that are present in all lakes.

Each of the environmental parameter records was then added to a master spreadsheet categorized by each species separately. The number of waterbodies that met the species habitat parameters allowed us to calculate a quantitative value for risk of establishment. This final spread looks similar to Table 1, but instead contains a risk column with the waterbodies marked as “low”,”moderate”, and “high”.

7

Table 2: This table is a list of the Vermont waterbodies which are considered to have “stressed” dissolved oxygen conditions.

Table 4: Denotes our tolerance range data gathered from the literature. All tolerance ranges are listed next to their subsequent source. Ranges that did not possess maximum values were left blank.

8

Table 5: Total counts of Vermont waterbodies from our data (n=374), separated into our categories of population establishment risk for each invader. Risk is classified as low, moderate or high in terms of the waterbody meeting either none, one or two of the water quality parameters we previously identified as necessary for establishment (Table 1). Values within parentheses represent percentages for each count.

9

10

Figure 1: This is a map of Vermont with thirteen of the DEC’s greeter stations that had sufficient data available. These are represented with pie charts for the high, medium, and low risk vessels launching into those waterbodies. This risk metric is based on the previous waterbody that the vessel was in and the time since that vessel was previously in another waterbody.

Discussion and Recommendations:

Spiny Water Flea: The results of our literature review and subsequent data analysis of the abiotic/biotic factors dictating the population establishment for spiny water flea (B. longimanus), largely support the idea that it can spread to almost all of Vermont’s waterbodies. Past research has indicated that the three most important habitat characteristics for this species are salinity, water temperature, and lake area. These factors are important due to the life history of this species, as it originated in the deep, cold water lakes of northern Europe and Asia (Kelly et al., 2013). We omitted salinity from our consideration, because while it is a limiting factor for the spiny water flea, all the lakes and ponds in Vermont are freshwater. Spiny water flea is known to tolerate a temperature range of 4-30 degrees Celsius, which directly coincides to the seasonal temperature variability within the waterbodies of Vermont (Weisz and Yan, 2010; Personal communication: Josh Mulhollem). Although its optimal range for population growth has been repeatedly identified as 10-24 degrees Celsius, recent studies have shown that temperatures outside of the optimal range do not impose a significant source of stress on the development of populations (Weisz and Yan, 2010). Additionally, when temperatures are too high or too low based on these tolerance ranges, spiny water flea can move throughout the water column to find a suitable environment that best fits its needs (Kerfoot et al., 2011). This movement capability directly relates to the second most limiting factor for establishment of this species, lake area. Recent research has shown that spiny water flea may perform better in large, deep lakes due to the wide range of habitat availability. Larger lakes possess a wider range of temperatures and higher availability of zooplankton, the main food source of this voracious predator (Kelly et al., 2013). Unfortunately, we were unable to include these two variables in our analyses because of gaps in the data. The waterbody data provided by the DEC did include temperature, but the values were measured on or near the surface and did not encapsulate the variability within the entire water column. For future research on the establishment of this species, we suggest a more in-depth analysis of the temperature variability within Vermont’s waterbodies both seasonally and at different levels of the water column. For the other limiting water quality characteristics that the research identified, such as: pH, hardness, water clarity and phosphorous content, we found that this species is capable of establishment in most, if not all, of the waterbodies we analyzed (Table 3).

Zebra Mussel:

The results of our literature review and subsequent analysis of the factors controlling the establishment of zebra mussels showed that it presents the least amount of establishment risk to Vermont waterbodies. The two water quality factors that are most prominent to its establishment are pH and hardness- a proxy for calcium content (Hincks and Mackie, 1997). These variables operate in tandem with one another. As pH decreases, the calcium content of the water decreases as well due to its relationship with the naturally-occurring buffer, CaCO3. Essentially, as cellular respiration produces carbon dioxide within the aquatic ecosystem, it binds with the surrounding H2O molecules to create carbonic acid(H2CO3). This weak acid is capable of dissolving limestone, releasing mineral ions such as calcium. Due to its weak nature, carbonic acid degrades over time, releasing two H+ ions and CO3

2-. The positively charged calcium ions then bind with the newly formed carbonate ion to create CaCO3 (Hammes and Verstraete, 2002). The loss of

11

readily available calcium within the water column is especially detrimental to zebra mussels because they are unable to use the mineral in order to form their shells (Hincks and Mackie, 1997). This delicate balance between hardness and pH renders a strict tolerance range that allows this invasive species to become established. Our review of the literature returned a pH range of 7.4-9.4 and a minimal hardness concentration of 57.7 mg/L (Cohen, 2017; Hincks and Mackie, 1997). The literature did not provide a maximum range value for water hardness because it is not applicable in this situation, as zebra mussels would be able to form their shells in aquatic ecosystems with higher mineral content (Wu et al., 2010). These values of tolerance were then applied to our waterbody data that we received from the DEC and returned relatively surprising results. Zebra mussels returned the lowest count of waterbodies that are at high risk for invasion (Table 4). We believe these findings are a consequence of the underlying bedrock geology of Vermont affecting the pH of the waters (Prindle, 1932). Most of the waterbodies we examined had a relatively acidic pH, which does not support the spread of this invasive bivalve.

In addition to pH and hardness, we examined the effects of chlorophyll-a and phosphorous concentrations on the suitability of zebra mussel habitat. Chlorophyll-a was considered as a surrogate for food availability as it needs a minimum amount of available nutrients to filter feed efficiently (Wu et al., 2010). The presence of phosphorous in the aquatic ecosystem operates in tandem with this variable; as phosphorous concentrations increase, phytoplankton growth increases due to the availability of nutrients. Our literature review returned a minimum chlorophyll-a concentration of 4.25 ug/L, which was only met in 34.59% of the assessed waterbodies (Table 3). Although the effects of zebra mussel establishment are extremely detrimental to the native ecosystem, we propose that they are the least threatening of the three invasive species that we analyzed in terms of spread risk due to their relatively stringent habitat requirements.

Starry Stonewort:

One of the main factors that influences the establishment of starry stonewort is water velocity (Schloesser et al., 1986). Schloesser et al. found the species cannot tolerate a water velocity of eleven centimeters per second, which is far above any water velocities that are typically observed in lakes and ponds (1986). This is because of the way the plant ground itself into the substrate, making waters with faster velocities uninhabitable for the plant (Schloesser et al., 1986). Schloesser et al. also found that substrate had an influence on the waters where the plant can be found, however, the DEC did not have the same robust data for each of their waterbodies on substrate as they do on water quality data (1986). We therefore omitted these two factors which could be the reason for the high risks to almost all of Vermont’s waterbodies (see Table 5).

Overall there was a lack in invasive range tolerance studies done on starry stonewort, because of the relatively recent invasion (Midwood et al., 2016; Brown, 2016; Escobar et al., 2016). These studies were not using the large datasets and support as those seen in when researching the zebra mussel tolerance ranges of environmental characteristics. Midwood et al. was a field study of the characteristics in observed areas where the species were found, and the characteristics of the area where the species was not observed (2016). This method is inherently flawed to the other due to the possibilities of regional variation and small sample sizes affecting the ranges that are seen. This also puts into question, is the range observed in the study representative for the whole species and what different populations can tolerate. Escobar et al.

12

attempted to combat this unknown by using a model, however, the model is not perfect due to the relatively small amount of knowledge in general available on this species (2016).

In Table 5, you can see that most Vermont waterbodies met the criteria that we set. Although this material does not have strongly scientifically supported metrics like the zebra mussel data, this is a representation of what the scientific community currently knows. We conclude that starry stonewort has minimal limitation in Vermont’s waterbodies, and therefore needs an increase in monitoring and vessel-checking programs to ensure that spread risk is low.

Further Discussion

After conducting a thorough review of the available literature, there are still significant uncertainties surrounding the extent to which we can characterize risk for an invasion of each of these species. When determining the water quality standards that allowed us to characterize habitats as suitable for invasion, there was substantial variation among reported threshold values. This was especially seen for starry stonewort since its invasion was more recent to North America. For example, the pH tolerance range was found by Escobar et al. (2016) to be 7.54-8.37, while Midwood et al. (2016) reported 7.4-8.5; with the nature of pH being a logarithmic scale, the difference between 8.3 and 8.5 is substantial. The literature is not in consensus, and this therefore has error built into our methods. We also used the watershed data that the DEC has stored in a database. Some of these values have extensive data stores with ten years of study, while others only have one year of data. To make up for this uncertainty, we noted which parameters only had one year’s worth of data and clearly labelled them in our data tables and spreadsheets. We used averaged data to alleviate discrepancies and we confirmed our proposed values with our community partner, Josh Mulhollem.

13

Since our project is set in the analysis and risk characterization phases of the ERA framework (see Figure 2), the primary focus was to provide information that could be used for a future a monitoring plan or policy change, not go as far as to communicate risk management techniques. We are only providing our partner with two different tools of risk characterization (our spreadsheet in the supplemental data link and our map in Figure 1).

The alternatives that we have identified, aims to address areas for improvement that are in the current invasive species monitoring program. The alternative that we believe to have the highest merit is the implementation of new boat transport laws. These laws would attempt to impose stricter regulations focused on a reduction of invasive transport between state waterbodies. Neighboring states like New York have already implemented proactive policies, which require boaters to clean and drain their boats before launching or leaving from waters owned by the Department of Environmental Conservation (Constantakes, 2014). A simple change in boating procedure would provide significant reductions in the spread of invasive species throughout the state, as personal watercraft are one of the main vectors of invasive movement.

To implement our proposed alternative we would need to focus on educating boaters throughout the state. This can be accomplished by assigning members of our team to boat launch points at the most popular waterbodies. By spreading the word about the consequences of not draining and drying boating equipment, we could gain support for the initiative. To ensure compliance, a state game warden could be assigned to focus on aquatic plant and animal transport. His/her primary objective would be to understand which species pose a threat to the

14

Figure 2: This diagram shows the Ecological Risk Assessment (ERA) framework referred to by the Environmental Protection Agency. Our project sits during the analysis and risk characterization phases of the process.This image was obtained from material created by Breck Bowden in a lecture on ecological assessment (Bowden, 2017).

aquatic ecosystems and to eliminate negligent boating practices. Perhaps a more extreme alternative would be the implementation of mandatory boat inspections by the Warden or trained volunteers upon entering or leaving waterbodies. The obvious caveat to this scenario is the sheer number of waterbodies that exist within the state. It would take far too many people and too much money to inspect every watercraft. Our hope is that by focusing on the most frequently visited lakes, we can reach the largest number of people.

Literature Citations:

Agency of Natural Resources. (2016). Gallery of Invaders: Species Currently Found in Vermont. Vermont Official State Website. (http://dec.vermont.gov/watershed/lakes-ponds/aquatic-invasives)

Bowden, B. (2017). Ecological Assessment: A Primer [PowerPoint slide]. Retrived from https://bb.uvm.edu/webapps/blackboard/content/listContent.jsp?course_id=_104916_1&content_id=_2117880_1

Brown, W. (2014). Defining trophic conditions that facilitate the establishment of an invasive plant: Nitellopsis obtusa. Master's thesis (natural resources and environmental sciences); University of Illinois, Urbana Champaign, Illinois.

Clavero, M., & Garcia-Berthou, E. (2005). Invasive species are a leading cause of animal extinctions. TRENDS in Ecology and Evolution, 20(3), 110-110.

Cohen, A., (2017). Potential Distribution of Zebra Mussels (Dreissena polymorpha) and Quagga Mussels (Dreissena bugensis) in California Phase 1 Report. California Department of Fish and Game.San Francisco Estuary Institute Oakland, CA. http://www.sfei.org/sites/default/files/biblio_files/2007-Dreissena_Potential_Distribution.pdf

Constantakes, P. (2014). New state regulations target aquatic invasive species [Press release]

Dodson, S. (1992). Predicting crustacean zooplankton species richness. Limnology and Oceanography, 37(4), 848-856

Escobar, L. E., Qiao, H., Phelps, N. B., Wagner, C. K., & Larkin, D. J. (2016). Realized niche shift associated with the Eurasian charophyte Nitellopsis obtusa becoming invasive in North America. Scientific Reports, 6.

15

Feng, X., & Papes, M. (2017). Physiological limits in an ecological niche modeling framework: A case study of water temperature and salinity constraints of freshwater bivalves invasive in USA. Ecological Modelling, 346, 48-57. doi: 10.1016/j.ecolmodel.2016.11.008

Hammes, F., & Verstraete*, W. (2002). Key roles of pH and calcium metabolism in microbial carbonate precipitation. Reviews in environmental science and biotechnology, 1(1), 3-7.

Hincks, S. S., & Mackie, G. L. (1997). Effects of pH, calcium, alkalinity, hardness, and chlorophyll on the survival, growth, and reproductive success of zebra mussel (Dreissena polymorpha) in Ontario lakes. Canadian Journal of Fisheries and Aquatic Sciences, 54(9), 2049-2057. doi: 10.1139/cjfas-54-9-2049

Johnson, L. E., & Padilla, D. K. (1996). Geographic spread of exotic species: ecological lessons and opportunities from the invasion of the zebra mussel Dreissena polymorpha. Biological conservation, 78(1-2), 23-33.

Karatayev, A. Y., Burlakova, L. E., & Padilla, D. K. (1998). Physical factors that limit the distribution and abundance of Dreissena polymorpha (Pall.). Journal of Shellfish Research, 17(4), 1219-1235.

Kelly, N. E., Young, J. D., Winter, J. G., & Yan, N. D. (2013). Dynamics of the invasive spiny water flea, Bythotrephes longimanus, in Lake Simcoe, Ontario, Canada. Inland Waters, 3(1), 75-92. doi:10.5268/w-3.1.519

Kerfoot, W. C., Yousef, F., Hobmeier, M. M., Maki, R. P., Jarnagin, S. T., & Churchill, J. H. (2011). Temperature, recreational fishing and diapause egg connections: dispersal of spiny water fleas (Bythotrephes longimanus). Biological Invasions, 13(11), 2513.

McMahon, R. F. (1996). The physiological ecology of the zebra mussel, Dreissena polymorpha, in North America and Europe. American Zoologist, 36(3), 339-363.

Midwood, J. D., Darwin, A., Ho, Z. Y., Rokitnicki-Wojcik, D., & Grabas, G. (2016). Environmental factors associated with the distribution of non-native starry stonewort (Nitellopsis obtusa) in a Lake Ontario coastal wetland. Journal of Great Lakes Research, 42(2), 348-355.

Modley, M. D. (2008). Aquatic invasive species rapid response planning partnerships in the Lake Champlain basin: Bridging international, political, social, and economic gaps. Water Sa, 34(4), 476-480.

Nunberg, G.K. (1988). A simple model for predicting the date of fall turnover in thermally stratified lakes. Limnology and Oceanography, 33(5), 1190-1195.

Pimentel, D., Zuniga, R., & Morrison, D. (2005). Update on the environmental and economic costs associated with alien-invasive species in the United States. Ecological economics, 52(3), 273-288.

16

Prindle, L. M. (1932). Geology of the Taconic quadrangle. American Journal of Science, (142), 257-302.

Pullman, G. D., & Crawford, G. (2010). A decade of starry stonewort in Michigan. Lakeline, summer, 36-42.

Rao, Y. R., Hawley, N., Charlton, M. N., & Schertzer, W. M. (2008). Physical processes and hypoxia in the central basin of Lake Erie. Limnology and Oceanography, 53(5), 2007-2020.

Rey-Boissezon, A., & Joye, D. A. (2015). Habitat requirements of charophytes-Evidence of species discrimination through distribution analysis. Aquatic Botany, 120, 84-91. doi: 10.1016/j.aquabot.2014.05.007

Schloesser, D. W., Hudson, P. L., & Nichols, S. J. (1986). Distribution and habitat of Nitellopsis obtusa (Characeae) in the Laurentian Great Lakes. Hydrobiologia, 133(1), 91-96.

United States Geological Survey. (2017). Nonindigenous Aquatic Species. USGS: science for a changing world. Retrieved from <https://nas.er.usgs.gov/>

Vermont Department of Environmental Conservation. (2016). Vermont Aquatic Invasive Specie Program: 2016 Update. 2017, from http://dec.vermont.gov/sites/dec/files/wsm/lakes/ans/docs/2016.pdf

Voorhees, J., Region, U., & Paula, M. (2015). Eric Howe Technical Coordinator Lake Champlain Basin Program 54 West Shore Road Grand Isle, Vermont 05458.

Weisz, E. J., & Yan, N. D. (2010). Relative value of limnological, geographic, and human use variables as predictors of the presence of Bythotrephes longimanus in Canadian Shield lakes. Canadian Journal of Fisheries and Aquatic Sciences, 67(3), 462-472.

Wu, Y. G., Bartell, S. M., Orr, J., Ragland, J., & Anderson, D. (2010). A risk-based decision model and risk assessment of invasive mussels. Ecological Complexity, 7(2), 243-255. doi: 10.1016/j.ecocom.2010.02.010

17