Surveys for non-indigenous tunicate species in Newfoundland , Canada ( 2006 – 2014 ) : a first step towards understanding impact and control

Fisheries and Oceans Canada initiated survey and monitoring programs in 2006 to determine presence, geographic range, and subsequent spread of non-indigenous tunicates in Atlantic Canada. Although non-indigenous tunicates have impacted aquaculture industries in Atlantic Canada for over a decade, aquaculture operations in Newfoundland and Labrador (NL) have not yet been affected. In this report, we document and explain results from biofouling collectors and various underwater techniques to survey and monitor for non-indigenous tunicates in NL between 2006 and 2014. During early surveys (2006–2007) we only observed low-impact invaders, Botryllus schlosseri and Botrylloides violaceus, at locations along the south coast of NL. Botryllus schlosseri became the most widely distributed species (18 locations within Placentia Bay) and evidence demonstrated spread to locations on the northeast (2011) and west (2013) coasts of NL. In 2012, we detected Ciona intestinalis in a single harbour in southern NL (Placentia Bay). In contrast to southern regions of Atlantic Canada, Styela clava and Didemnum vexillum, two high-impact invasive tunicates, were not observed during these surveys. Newfoundland and Labrador represents the northern range limit of non-indigenous tunicates in Atlantic Canada, but seawater temperatures are unlikely to prevent further expansion or introduction of non-indigenous tunicates throughout the province.


Introduction
Non-indigenous species, defined as organisms living outside of their native distributional range, classically spread either intentionally or accidentally by human activities.In certain cases, species become invasive after establishing themselves in new environments and have detrimental ecological or economic consequences.With escalations in global transport, focus on monitoring, risk, and research activities related to invasive species has increased, which requires accelerating documentation of introductions and range expansions of aquatic invasive species (AIS; Carlton and Geller 1993;Cohen and Carlton 1998;Mack et al. 2000).Invasive tunicates fall under the category of biofouling organisms, which include organisms that attach to underwater surfaces of vessels (e.g.hulls or niche areas such as rudders, sea chests, or propellers) and hitch a ride from one region to another (Carlton and Hodder 1995;Clarke Murray et al. 2011;Frey et al. 2014).Once transported, these biofouling species attach to local substrates (natural or artificial), reproduce, and can establish populations.Although biofouling vectors are primarily human mediated (e.g.movement of recreational and commercial vessels, slow moving barges, or transfer of gear), the likelihood of spread through natural vectors, such as rafting, drifting, or attachment to mobile organisms is an additional concern (Ruiz et al. 1997;Lewis et al. 2005;Thiel and Gutow 2005;MacFarlane et al. 2013).
In total, 7 non-indigenous tunicate species have been confirmed within the Canadian Maritimes: Styela clava Herdman, 1881, Ciona intestinalis (Linnaeus, 1767), Botryllus schlosseri (Pallas, 1766), Botrylloides violaceus Oka, 1927, Ascidiella aspersa (Müller, 1776), Diplosoma listerianum (Milne-Edwards, 1841), and most recently (2013), Didemnum vexillum Kott, 2002(Ramsay et al. 2009;Martin et al. 2011;Sephton et al. 2011;Moore et al. 2014).Non-indigenous tunicates that become invasive have the ability to rapidly foul natural and man-made structures and compete with native species for food and space.On mussel lines at aquaculture sites, they compete with mussels, contributing to decreased mussel size and condition, increased mortality, and increased drop off rates from lines (Daigle and Herbinger 2009).For example, mussel aquaculture in Prince Edward Island (PEI), Canada has experienced serious impacts from the establishment and spread of invasive clubbed, S. clava, and vase, C. intestinalis tunicates (Thompson and MacNair 2004;Locke et al. 2009).The added cleaning of mussel lines and significant increases in weight from tunicates has strained aquaculture operations by increasing loss of product and requiring additional time and labor (Thompson and MacNair 2004).Although the first threat to PEI shellfish aquaculture was S. clava, most recently, the rapid invasion of C. intestinalis replaced S. clava as the foremost invasive tunicate species (Ramsay et al. 2009).In PEI, management responses have included regulation of aquaculture transfers and harvest dates, which have been relatively successful in controlling spread of solitary tunicates (e.g. S. clava and C. intestinalis; Locke et al. 2009).The recent discovery of D. vexillum within the Bay of Fundy (Canada; Moore et al. 2014) further validates the continued spread and range expansion of non-indigenous species in Atlantic Canada and the importance of sustained monitoring programs.
By definition, monitoring is the gathering of data using standardized methods intended to deliver information on species characteristics and their changes over time, while surveys are the collection of information to provide an instantaneous view of a certain area at a specific time (Lehtiniemi et al. 2015).Furthermore, the combinations of monitoring, surveys, and subsequent research are necessary fundamentals for marine environmental management (Lehtiniemi et al. 2015).Throughout Atlantic Canada, the Department of Fisheries and Oceans Canada (DFO) has initiated zonal AIS monitoring programs (e.g.Martin et al. 2011;Sephton et al. 2011;Simard et al. 2013), including this study in Newfoundland and Labrador (NL).These programs use survey and monitoring strategies as the first step to understand the spread and effects of AIS and to guide projects on non-indigenous tunicates through 1) early detection, 2) rapid response and mitigation, and 3) prevention.Information provided by these strategies have informed risk assessments and provided foundations for research on species ecology, prevention practices, and mitigation methods (McKenzie et al. 2016).Prior to initiation of the DFO AIS monitoring program in NL ( 2006), records existed of only one non-indigenous tunicate species (B.schlosseri), which was observed on the west coast of NL in the 1970s (Hooper 1975;Ma et al. 2010) and Argentia in the 1940s (Placentia Bay; US Navy 1951).Since formation of the AIS monitoring program, established populations of two additional non-indigenous tunicates (B.violaceus and C. intestinalis) have been detected in NL (Callahan et al. 2010;Sargent et al. 2013).This document summarizes DFO AIS surveys and monitoring conducted to determine distribution and abundance of non-indigenous tunicates (2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014) throughout insular NL.Furthermore, we will discuss the range and abundance of invasive tunicates and subsequent actions in the context of rapid response, mitigation, and focused research.

Materials and methods
Province-wide AIS surveys were conducted throughout insular NL led by the Department of Fisheries and Oceans Canada in partnership with the Department of Ocean Sciences, Memorial University of Newfoundland (MUN).Survey locations (Appendix 1) were targeted to focus on marine sites defined as high risk for introduction and spread of non-indigenous tunicates (Baines 2007;Lambert and Lambert 1998).Such locations included marinas, public wharves, aquaculture and potential aquaculture sites, and other locations determined to have relatively high volumes of vessel traffic (e.g.ferry terminals or commercial ports).Moreover, survey locations were selected in close proximity to structures with high biofouling potential, such as floating docks and man-made wharves (Airoldi et al. 2015).
A combination of methods was used (2006 -2014) to assess the distribution and spread of non-indigenous tunicates in NL, including biofouling collectors, SCUBA surveys, underwater video, and shoreline observations.The number of survey locations varied year to year (Table 1), because of travel distance between locations (up to 1000 km) and sites identified by industry or stakeholders as areas of concern.Between 2006 and 2009, surveys were conducted primarily using biofouling collectors suspended at 37 different sites in NL.These surveys encompassed the entire island of NL and included 12 harbours, 2 ferry terminals, 8 mussel aquaculture sites, and 15 sites designated as potential candidates for future aquaculture mussel seed sites.Following initial surveys, six locations found to have established populations of non-indigenous tunicates were targeted for monitoring (e.g. annually or biannually).This monitoring included the extent of tunicate cover, species growth, and secondary spread using a combination of collector plates and underwater surveys conducted between 2010 and 2014 (Foxtrap and Arnolds Cove, B. schlosseri; Belleoram, B. violaceus; and Little Bay, Burin, and Marystown, C. intestinalis).At three of these sites, experimental mitigations were conducted to evaluate control methods to prevent the spread of the non-indigenous tunicate (Belleoram, B. violaceus [2008 and2009];Foxtrap, B. schlosseri [2011]; Little Bay, C. intestinalis [2013,2014]; Deibel et al. 2014;McKenzie et al. 2016).
Collectors at harbour and ferry sites consisted of three horizontally oriented, lightly sanded settling PVC plates (10 x 10 cm) suspended from the surface (e.g.attached to wharves, buoys, etc.) in a vertical series at 37.5 cm intervals.For another research project on the west coast of NL, PVC plates were spaced by only 10 cm.Collectors were placed in shaded areas with relatively high biofouling growth, positioned ~ 1.0 m below the surface at low tide, and weighted at the bottom to ensure they remained vertical in the water column.Collectors deployed at aquaculture and potential aquaculture sites varied slightly from collectors at harbour sites to better survey the biofouling potential of mussel mooring lines used at NL aquaculture sites in concurrence with a mussel seed research project.These collectors were hung at the same depth as mussel lines and consisted of six PVC plates (10 × 10 cm) at depths of 1.5, 3.5, 4.0, 4.5, 5.0, and 5.5 m.The three shallowest PVC plates were oriented horizontally (i.e.plate surface was parallel to water surface) in the water column whereas the deepest three PVC plates were positioned vertically (i.e.plate surface was perpendicular to water surface).Mussel spat collection ropes (4 ropes in total of 0.64 cm polyrope held in place at either end by two bucket lids) were attached to each collector between plates at 1.5 and 3.5 m depth.Temperature recorders (Vemco Minilog II-T or Hobo data logger) were attached opportunistically to collectors in several locations to record temperature during deployment.
One to three collector lines were suspended at each harbour site, while three collectors were deployed at each aquaculture and potential aquaculture site.Plates were generally deployed in late spring/early summer (~June) and retrieved in the fall (between November and December) each year.In years following 2006, new plates were often redeployed during site visits in the fall of the year and remained in the water over winter.In some scenarios, additional site visits were possible and allowed plates to be analyzed mid-season (~August) to supplement analyses on full-season plates retrieved in the fall.Collectors were in the water at least during the warmer months of the year and deployed at minimum 2-3 months and in many cases closer to one year.
When collectors were retrieved, PVC plates were fixed in 10% formalin buffered with sodium borate.For laboratory analyses, plates were rinsed with fresh water to remove formalin and both sides of plates were examined under dissecting (Leica M80) and compound (Zeiss Axio AX10) microscopes.The presence and area of nonindigenous tunicates was determined and recorded.Only data from the bottom side of PVC plates are presented here, as sedimentation on top side of plates characteristically reduce recruitment, and in particular, species of interest in Atlantic Canada typically exhibit photonegative behaviors and settle in low light environments (Svane and Young 1989;Lambert 2005;Ruiz et al 2010).In 2006 and 2007, the area of each plate occupied by non-indigenous tunicates was determined by overlaying a 10 × 10 cm grid divided into 1 cm 2 quadrats.Analysis conducted from 2008 onwards was completed using image analysis software (ImageJ; public domain software by W. Rasband, National Institute of Mental Health, USA, http://imagej.nih.gov/ij/index.html).The area occupied was converted to percent cover and categorized based on the following categories (see Sephton et al. 2013): 0 (Absent), <25% (low), 25-50% (moderate), 51-75% (high), >75% (very high).All biofouling organisms were identified to the lowest taxonomic level possible, but information on species other than non-indigenous tunicates is not presented here.
To supplement biofouling collectors, SCUBA and underwater video surveys (e.g.pilings on perimeter and underneath wharf, floating docks, and transects 50-100 m perpendicular and parallel to shore) targeted sites with confirmed presence of non-indigenous tunicates on collectors and often occurred opportunistically in combination with other research projects, which greatly increased  In April 2014, we observed relatively low abundances of C. intestinalis in Little Bay (Site A), Burin, and Marystown.Most specimens in Little Bay appeared physically damaged and not alive (potential evidence of predation or ice scour), but all observed individuals were removed.We did not detect any recruits on PVC collectors (<1 % coverage) until August 2014 on PVC plates.No C. intestinalis was observed on the Little Bay wharf throughout 2014, but a couple of specimens were found on one boat and were removed.In contrast to previous years, 2014 SCUBA surveys detected greatest abundances in Burin on wharf structures and adjacent boats.At the Marystown wharf, a large number of specimens were observed attached to blue mussels fouling a non-active vessel, which were removed.PVC plates retrieved during October 2014 from an eelgrass meadow in Little Bay (Site B) exhibited high coverage (up to 80%) of C. intestinalis.Between 2013 and 2014, monthly seawater temperatures at locations where C. intestinalis was present ranged from -0.4 ºC (February) to 17.2 ºC (August).

Discussion
Survey and monitoring programs for nonindigenous species provide baseline information to assess change, which is necessary to deliver advice for environmental management decisions (Lehtiniemi et al. 2015).Non-indigenous tunicates have historically been reported throughout southern regions of Atlantic Canada (Carver et al. 2006 a;b) including population outbreaks at mussel aquaculture sites in PEI (S. clava) and Nova Scotia (C.intestinalis) since 1997 and 1998, respectively (Cayer et al. 1999;Clarke and Therriault 2007).Generally, DFO AIS monitoring programs between 2006 and 2009 observed trends of expanded ranges, heavier infestations, and increased tunicate biodiversity at stations in these regions of Atlantic Canada (Martin et al. 2011;Sephton et al. 2011) Botryllus schlosseri was the lone non-indigenous tunicate confirmed during initial surveys (2006) and only on the hull of a recreational boat in Argentia (Placentia Bay), which was the same location as historical reports of this species (US Navy 1951).Interestingly, B. schlosseri is now absent from historical west coast locations (Deibel et al. 2014).These results demonstrated a contrasting situation to the widespread and relatively long-term presence of non-indigenous tunicates in more southern regions of Atlantic Canada.However, the presence of B. schlosseri on only one recreational boat hull and its absence on collectors in Argentia the subsequent year suggested either a small population at this location or that B. schlosseri is present elsewhere in the area.In 2007, surveys confirmed B. schlosseri at four additional locations in northern regions of Placentia Bay, which corroborated a more widespread population and supports the suggestion that the original introduction of B. schlosseri may have occurred several decades ago (Deibel et al. 2014).
The lack of baseline information on the presence and distribution of non-indigenous tunicates in NL prompted surveys covering vast expanses of coastlines, which targeted public harbours and aquaculture sites.Monitoring only selected sites at this stage would have narrowed the comprehensive geographic distribution of the surveys and ineffectively targeted non-indigenous tunicates (Lehtiniemi et al. 2015).Although survey locations and strategies varied across years, we observed evidence of local and regional population expansion with all non-indigenous tunicate species in NL (B. schlosseri, B. violaceus, and C. intestinalis).Following most recent surveys (2014), B. schlosseri was observed at the most locations (24), and by 2013, new, discontinuous populations were detected in Conception Bay, Hermitage Bay, and the southwest coast.In Placentia Bay during 2012, B. schlosseri was found fouling natural substrates including eelgrass at numerous locations.Such findings have led to collaborations with a latitudinal study (New Jersey, USA to NL, Canada) examining distribution and diversity of invasive tunicates on eelgrass (Carman et al. 2016).These locations in Placentia Bay have little to no vessel traffic, which indicates that natural dispersal has occurred from presumable sites of introduction (e.g.public harbours) leading to relatively continuous populations within Placentia Bay.Planktonic tunicate larvae generally have limited dispersal potential (Svane and Havenhand 1993;Lambert and Lambert 1998;Shanks 2002), however colonies rafting on natural substrates (e.g.eelgrass) can travel long distances (Worcester 1994;Carman and Grunden 2010).While continuous populations within Placentia Bay likely resulted from combinations of natural dispersal and point to point anthropogenic transport, discontinuous populations elsewhere in NL (e.g.Conception Bay) are expected consequences of human-mediated movements.
Botrylloides violaceus remained restricted to Belleoram harbour, but coverage on collectors and various substrates (wharf, floating docks, and rocks) increased substantially between 2007 and 2010.Interestingly, this expansion also coincided with the enlargement of the fixed wharf in 2009 (Deibel et al. 2014).Such alterations create new, bare substrates, which allows occupation by non-indigenous species better adapted to compete for bare space (i.e.rapid reproduction and growth) than native biofouling organisms (Byers 2002;Altman and Whitlatch 2007;Tyrrell and Byers 2007).Early detection of such relatively isolated populations created opportunities for rapid response and experimental mitigation trials by DFO in 2008and 2009, which used encapsulation methods to remove B. violaceus in Belleoram harbour (McKenzie et al. 2016).Early results were encouraging, but the use of plastic wrap to encapsulate structures also provided new exposed substrate, ideal for tunicate larvae and fragments to settle and grow unimpeded (Deibel et al. 2014).
Although tunicates generally select shaded areas to settle (Svane and Dolmer 1995) Newfoundland and Labrador represents the northern invasion front in the northwest Atlantic for non-indigenous tunicates as these species continue to expand geographically within Atlantic Canada (Martin et al. 2011;Sephton et al. 2011).Environmental parameters, including cold water temperatures, may limit further expansion, but following our initial surveys, related research documented constraints of water temperature on life cycle events for B. schlosseri in Placentia Bay and revealed relatively high survival of overwintering colonies (Ma 2012;Deibel et al. 2014).Therefore, although non-indigenous tunicates remain absent from most regions of NL, further range expansion is possible.Surveys are increasingly essential to document irregular range expansions symptomatic of organisms hitchhiking through human-mediated transport.For example, although non-indigenous tunicates have been predominantly detected on the south coast of NL, in 2011 and 2013, B. schlosseri was detected >100 km from previous recorded locations attached to public wharves on the northeast (Foxtrap, Conception Bay) and southwest (Codroy) coasts of NL, respectively.Furthermore, in 2013, B. violaceus was also detected in Codroy, 100s of km west of the only previous sighting in NL.Such discontinuous range expansions have led to the development of ongoing research on the population dynamics of non-indigenous tunicates across a ~270 km latitudinal gradient on the west coast of NL.
In Atlantic Canada, NL has a relatively high number of small craft harbours encircling the island (Therriault and Herborg 2008), which represent high risks for introduction and spread of biofouling communities (Clarke Murray et al. 2011).Although B. schlosseri is customarily perceived as a nuisance species, C. intestinalis has caused havoc within the aquaculture industry in PEI (Cayer et al. 1999).C. intestinalis is currently relatively isolated within the area of Mortier Bay, but this area experiences relatively high vessel traffic to all regions of NL which increases the risk of human-mediated transport throughout NL.Risk assessment models have indicated that favorable environmental conditions exist along the southwest and south coasts of NL and are at risk for establishment and spread of C. intestinalis and other non-indigenous tunicates currently in Atlantic Canada (Therriault and Herborg 2008).The presence of B. schlosseri on the northeast coast of NL, in combination with its high overwintering survival and short generation times indicates further spread by non-indigenous tunicates to more northern regions of NL may occur (Deibel et al. 2014).Seawater temperatures observed across survey locations were well within the ranges of growth and reproduction for all non-indigenous tunicates detected in NL (Carver et al. 2006a;Therriault and Herborg 2008).Therefore, the threat of non-indigenous tunicates to nearby shellfish aquaculture operations, particularly within Placentia Bay, is of concern.
Survey and monitoring programs have discovered non-indigenous tunicates new to Atlantic Canada, including the high impact, invasive species, D. vexillum within the Bay of Fundy in 2013 (Moore et al. 2014).Vessel traffic between NL and the rest of Atlantic Canada fosters apprehension for further introductions and northward expansion of non-indigenous tunicates.D. vexillum can grow in seawater temperatures between -2 and 24 ºC (Lambert 2009), and therefore is another threat to NL waters.Although not all coastal regions of NL may be suitable for large, invasive populations of all tunicates species (Therriault and Herborg 2008), changes in climate may make insular NL waters more susceptible to invasion.Through early detection or range expansions, the NL DFO AIS survey and monitoring program has been the first step in an aquatic invasive species response plan for prevention, early detection, rapid response, mitigation trials, and research on the ecology of these species at their current northwest Atlantic range limit (McKenzie et al. 2016).Results from past projects have continued to promote research on the vulnerability of early-life stages to treatments (e.g.ultrasound generation, suspended sediments, and bubble streams) to prevent biofouling and minimize risks of introduction (Lowen et al. 2016).Survey and monitoring programs not only provide baseline data and early detection, but further encourage diversification and expansion of research, which ultimately expands the scope of available strategies and science advice for prevention and management.

Table 1 .
Number of stations surveyed using collectors or underwater techniques (e.g.SCUBA and underwater video) in Newfoundland, Canada between 2006 and 2014.

Table 2 .
Summary of locations where non-indigenous tunicates were observed on collectors between 2007 and 2014 (Site type: M -Marina, C -Coastal).
sites.Generally, the rate of occurrence of nonindigenous tunicates was low, but the number of tunicate species and occurrence increased year to year (Table1).For example, in the first survey year (2006) B. schlosseri was the only non-indigenous tunicate found and only at one site.In 2007, B. violaceus was confirmed in Fortune Bay.In 2012 (a particularly warm year), C. intestinalis was detected in southwest Placentia Bay, which also coincided with the greatest number of observations of B. schlosseri (18 sites; Table 1).Throughout all surveys, B. schlosseri was the dominant tunicate found in NL with the most occurrences (18 of 22 positive sites) within Placentia Bay.B. schlosseri was confirmed at one site within adjacent Fortune and Hermitage Bays, and more recently in Conception Bay (2011) and on the southwest coast of NL (2013; Figure 1).growth on plates in 2011 (Table 2).Surveys verified B. schlosseri at one new site (Baine Harbour) in southwestern Placentia Bay in 2008 and for the first time in Conception Bay in 2011 at Foxtrap marina in northeastern NL (Table 3).Follow-up SCUBA surveys in Foxtrap during late November determined that B. schlosseri was restricted to one side of the marina on fouled floating wharves and Laminaria sp.attached to the floating wharves.A small colony of B. schlosseri was observed growing on the no B. violaceus anywhere on or near the wharf structure.In 2013 and 2014, B. violaceus was observed during underwater surveys and on collectors in southwest NL (Codroy; Figure 1).Water temperature loggers recorded seawater temperatures in Belleoram for most of 2008, 2011,

Table 3 .
Summary of underwater survey locations where non-indigenous tunicates were observed between 2007 and 2014 (Site type: M -Marina, C -Coastal, F -Ferry terminal).
(Ruis et al. 2010989;Lambert 2005)Atlantic Canada of interest to this study (e.g. C. intestinalis and B. schlosseri;Svane and Young 1989;Lambert 2005), some tunicate species are exceptions to this rule and prefer well-lit environments(Ruis et al. 2010).