Relationships between satellite-measured thermal features and Alexandrium-imposed toxicity in the Gulf of Maine

https://doi.org/10.1016/j.dsr2.2005.06.025Get rights and content

Abstract

Relationships between satellite-derived sea-surface temperature (SST) patterns and the occurrence of paralytic shellfish poisoning (PSP) toxicity events caused by Alexandrium fundyense in the western Gulf of Maine are examined. Comparisons between surface A. fundyense cell distribution patterns and SST images indicate that highest cell concentrations are associated with colder waters of the eastern segment of the Gulf of Maine coastal current (EMCC) and that frontal zones at the edges of the EMCC often act as boundaries to surface distributions. Surface thermal patterns coincident with a May 2000 PSP toxic event and shellfish harvesting closure on the western Maine coast show enhanced connectivity between the EMCC and the western Gulf of Maine, suggesting transport linking A. fundyense cells in the EMCC to inshore areas of the western Gulf of Maine. Surface drifter data support such transport. Thirteen years (1990–2002) of toxicity data from eight monitoring sites along the coast of Maine and concurrent SST data show that in years of either large or very reduced toxicity, a consistent relationship exists between the timing and strength of fronts, taken as an indicator of alongshore connectivity, and the occurrence and strength of toxic events. Years with weak fronts and/or fronts that become established relatively late in the summer growing season are years of the strongest toxicity events in western Gulf of Maine. Years of early and strong fronts are years with few and/or weak toxicity events. Our results suggest that advective connections exist between cells present in the EMCC and toxicity along the western Gulf of Maine coast and that large-scale hydrographic processes, characterized here as surface thermal patterns, influence A. fundyense populations in the western Gulf of Maine, either through delivery of actual cells or advection of advantageous conditions into the region. These data point to the utility of satellite and other coastal observing system data for the monitoring and prediction of conditions linked to toxic events in coastal waters.

Introduction

The saxitoxin associated with the dinoflagellate Alexandrium fundyense2 causes paralytic shellfish poisoning (PSP) in the Gulf of Maine (Fig. 1), with direct health and therefore economic implications (Blaxter and Southward, 1997). The toxin can be fatal to humans in doses as small as 500 μg/100 g ingested tissue (Hallegraeff, 1995). Shellfish toxicity events in the Gulf of Maine have been recorded annually since 1972. To ensure public safety, the Maine Department of Marine Resources (DMR) monitors toxin levels in shellfish at an extensive series of coastal sites using a mouse bioassay. If toxin levels approach or exceed 80 μg toxin/100 g tissue, shellfish beds in the local vicinity are closed to further harvesting. Economic losses resulting from shellfish closures are difficult to quantify over the entire Gulf of Maine; however, the cost of closures in 1994 in Casco Bay alone was estimated at $1.6–$1.7 million (Maine Environmental Properties Project (MEPP), 1996).

In 1997, a multi-institution, multi-investigator program to study the oceanographic ecology of A. fundyense in the region was initiated (ECOHAB, Gulf of Maine). Here, we use results from this program and the historical record of toxicity events along the Maine coast to examine the relationship between satellite-measured sea-surface temperature (SST) patterns, A. fundyense distribution, and the occurrence of toxicity events in the western Gulf of Maine.

Previous work using data from broad-scale ECOHAB ship surveys in 1998 and 2000 established that large A. fundyense cell densities can occur in offshore waters in the Gulf of Maine during summer months (Townsend et al., 2001), spatially continuous with elevated concentrations in the Bay of Fundy reported by Martin and White (1988). Townsend et al. (2001) note that the locations of highest cell concentrations (ca. 5.5×103 cells l−1) are coincident with the cold, nutrient-rich eastern segment of the Gulf of Maine coastal current (EMCC) (Pettigrew et al., 2005), possibly due to increased light penetration and elevated nutrient concentrations in surface waters. These cells are hypothesized to originate from a seed population in the Bay of Fundy, with highest cell densities found in the southeastern part of the Bay near the coast of Nova Scotia (Martin and White, 1988). White and Lewis (1982) also found the highest concentration of A. fundyense cysts in the winter in offshore waters north and east of Grand Manan Island. High vegetative cell concentrations in the Bay of Fundy suggest that these cysts could be a major source for summer bloom initiation (Martin and White, 1988).

In the eastern Gulf of Maine, the westward-flowing EMCC (Fig. 1) is considered the dominant hydrographic feature (Brooks and Townsend, 1989). Extensive tidal mixing keeps surface waters relatively cold and nutrient-rich throughout the year. In the vicinity of Penobscot Bay, the EMCC turns cyclonically offshore. At this point, portions of the EMCC contribute either to the cyclonic gyre over Jordan Basin or feed into the western segment of the Gulf of Maine coastal current (WMCC) continuing down the coast (Fig. 1). The amount of EMCC water that contributes to either branch is seasonally dependant and is related to vertical density structure over Jordan Basin (Brooks and Townsend, 1989). As summer approaches, more water contributes to the gyre over Jordan Basin than continues into the WMCC. At this divergence point of the EMCC from the coast (the Penobscot Bay region), Shumway et al. (1988) identified a toxin-free portion of the coast with sites both east and west of this region exhibiting reoccurring toxicity events.

Warming and the increase in stratification over both offshore and western Gulf of Maine waters in spring/summer makes the (at least surface) offshore deflection of the relatively cold EMCC easily visible in satellite infrared images. At this time, two surface frontal zones become prominent. In the vicinity of Penobscot Bay where offshore-flowing cold EMCC surface waters lie adjacent to the warmer, more stratified waters of the western Gulf of Maine, a strong surface thermal gradient and frontal zone oriented in the cross-shore direction results (Xue et al., 2000). Additionally, upstream of this position within the eastern Gulf of Maine, an alongshore-oriented surface thermal front forms along the offshore margin of the EMCC, where it lies next to more stratified Jordan Basin waters (Fig. 2B).

Anderson (1997) hypothesized that the cyclonic offshore branching of the EMCC in the vicinity of Penobscot Bay makes it unlikely that cells present in the western Gulf of Maine result from transport from an eastern Gulf of Maine source. This scenario necessitates an independent population of A. fundyense in the western Gulf of Maine having its own source and transport pathways (Anderson, 1997). Townsend et al. (2001), however, suggest that A. fundyense in the EMCC can be transported into the western Gulf of Maine, meaning that a single A. fundyense population, possibly originating in the Bay of Fundy, could influence the entire study region. The degree of connection between eastern and western Gulf of Maine appears highly variable (Pettigrew et al., 1998) making generalizations about connections between observed high offshore cell densities and toxicity within inshore shellfish beds difficult based on sparse or episodic data.

Here, we first use coincident ship and satellite SST data to show the relationship between cell surface spatial patterns and surface thermal features linked to circulation in the Gulf of Maine. Using satellite and surface drifter data, we then document a direct connection between surface EMCC water and the western Gulf of Maine coast that was coincident with a toxic event and shellfish harvesting closure. We then use a 13-year timeseries of toxicity measurements from Maine coastal sites and coincident satellite SST imagery to examine the relationship between toxicity events along the western Gulf of Maine coast and the apparent connectivity between the EMCC and the western Gulf of Maine. We hypothesize that interannual differences in the occurrence and magnitude of toxicity events along the coast of western Gulf of Maine are associated with interannual variability in the strength and timing of the seasonally modulated EMCC contribution to the WMCC.

Section snippets

Data and methods

Toxicity, determined by mouse bioassay, is measured at up to 300 possible stations along the coast of Maine each year by the Maine DMR. If the toxin level at a station is at, or approaching, 80 μg/100 g tissue, the DMR closes the area to the harvesting of that species of shellfish. Toxicity is measured in a varying range of shellfish species at each site. Here, we use toxicity from Mytilus edulis (blue mussel) from the years 1990–2002 as this species was extensively and most consistently sampled

Broad-scale spatial patterns and evidence of east–west connections

Relationships between satellite-derived SST and the distribution of A. fundyense in surface waters of the Gulf of Maine collected during three broad-scale surveys in 1998 are shown in Fig. 3. These data suggest, especially early in the summer (June and July, Fig. 3A and B), that highest cell concentrations are located within the colder waters of the EMCC and that surface cell distributions are modulated by circulation patterns (Fig. 1). These results are consistent with the comparisons of cell

Discussion

One observation from the data is a delayed annual onset of toxicity in the eastern station (East Pond Cove, Fig. 5) compared to those in the west. Although this single eastern station requires analyses of additional stations to verify regional trends, it is consistent with proposed hypotheses explaining east–west Gulf of Maine A. fundyense ecology differences. One hypothesis explaining differences in the annual timing of toxicity events in the western and eastern Gulf of Maine is that the two

Conclusions

Surface A. fundyense cell distributions in the Gulf of Maine are strongly linked to surface hydrographic properties. We show that these are identifiable in satellite SST data such that maximum cell concentrations are within the colder EMCC waters. Comparisons of interannual variability in surface frontal dynamics and toxicity present a consistent story linking interannual variability of toxicity events in the western Gulf of Maine to large-scale hydrographic features. Toxicity events in the

Acknowledgments

A special thanks to the reviewers for this issue for their helpful and constructive criticism. Thanks to Laurie Bean for assistance with the toxicity data, to Jim Churchill for the 2000 drifter data, Peter Cornillon for access to Pathfinder AVHRR data and Ryan Weatherbee for assistance with final figures. We acknowledge the huge effort by the Townsend laboratory at the University of Maine in processing the in situ data from the survey cruises in 1998 and 2000 and our ECOHAB Gulf of Maine

References (24)

  • J.F. Cayula et al.

    Cloud detection from a sequence of SST images

    Remote Sensing of Environment

    (1996)
  • A.C. Thomas et al.

    Satellite-measured phytoplankton variability in the Gulf of Maine

    Continental Shelf Research

    (2003)
  • D.W. Townsend et al.

    Offshore blooms of the red tide dinoflagellate, Alexandrium sp., in the Gulf of Maine

    Continental Shelf Research

    (2001)
  • D.M. Anderson

    Bloom dynamics of toxic Alexandrium species in the northeastern US

    Limnology and Oceanography

    (1997)
  • D.M. Anderson et al.

    Toxic Alexandrium blooms in the western Gulf of Maine: the plume advection hypothesis revisited

    Limnology and Oceanography

    (2005)
  • Bean, L.L., McGowan, J.D., Hurst Jr., J.W., 2005. Annual variations in paralytic shellfish poisoning in Maine, USA,...
  • J.H.S. Blaxter et al.

    Advances in Marine Biology: The Biogeography of the Oceans

    (1997)
  • David A. Brooks et al.

    Variability of the coastal current and nutrient pathways in the eastern Gulf of Maine

    Journal of Marine Research

    (1989)
  • P.J.S. Franks et al.

    Alongshore transport of a toxic phytoplankton bloom in a buoyancy current: Alexandrium tamarense in the Gulf of Maine

    Marine Biology

    (1992)
  • G.M. Hallegraeff

    Harmful algal blooms: a global overview

  • Keafer, B.A., Churchill, J.H., McGillicuddy, D.J., Anderson, D.M., 2005a. Bloom development and transport of toxic...
  • Keafer, B.A., Churchill, J.H., Anderson, D.M., 2005b. Blooms of the toxic dinoflagellate, Alexandrium fundyense, in the...
  • Cited by (35)

    • Projected effects of climate change on Pseudo-nitzschia bloom dynamics in the Gulf of Maine

      2022, Journal of Marine Systems
      Citation Excerpt :

      Third, because Scotian Shelf Water inflows are relatively cool compared to Northeast Channel inflows (Townsend et al., 2015), and because the inflows continue via the coastal current into the Bay of Fundy, increased Scotian Shelf Water inflows could help explain the relatively small increases in SST near the coast south of Nova Scotia, in the Bay of Fundy, and along eastern Maine. In both the hindcast and projected composite years, a mass of cool surface water enters the domain from the Scotian Shelf, wraps around Nova Scotia, and spreads into the Bay of Fundy, similar to what has been seen in satellite images (Luerssen et al., 2005). In the projected composite year, while the average surface temperature in most of the domain was 16°C, this coastal region remained between 8 and 12°C (Supplementary Material, Fig. S-7).

    • Dynamics of an intense Alexandrium catenella red tide in the Gulf of Maine: satellite observations and numerical modeling

      2020, Harmful Algae
      Citation Excerpt :

      Temperature may differ across these salinity fronts, so in certain cases satellite sea surface temperature (SST) may indicate coastal fronts. One study did find that thermal fronts and PSP toxins were spatially related in the coastal region of the Gulf of Maine (Luersson et al., 2005), although such thermal fronts are typically undetectable in spring and summer (Ullman and Cornillon, 1999). Physical/density fronts in the Gulf of Maine are associated with salinity changes (Incze and Yentsch, 1981; Stumpf and Goldschmidt, 1992; Hetland et al., 2015).

    • Decadal variability on the Northwest European continental shelf

      2018, Progress in Oceanography
      Citation Excerpt :

      Coastal blooms of harmful phytoplankton are known to impact fishing and aquaculture industries, and harm human health (Berdalet et al., 2016). In particular, blooms of the destructive ‘red tide’ dinoflagellate Alexandrium fundyense are thought to be mediated by water mass positioning and front strength in the western Gulf of Maine (Luerssen et al., 2005). Red tides have been observed in Scottish waters (Jones et al., 1982) but the nutrient supply is typically from land runoff.

    • Understanding interannual, decadal level variability in paralytic shellfish poisoning toxicity in the Gulf of Maine: The HAB Index

      2014, Deep-Sea Research Part II: Topical Studies in Oceanography
      Citation Excerpt :

      Since 1972, PSP toxicity has been a recurrent problem throughout the region, including western gulf waters, and is responsible for widespread shellfish harvesting closures and significant economic losses virtually every year (Shumway et al., 1988; Jin et al., 2008). Since the major 1972 outbreak, annual toxicity has varied extensively within the region, both in geographic extent and magnitude (McGillicuddy et al., 2005b, 2011; Luerssen et al., 2005; Thomas et al., 2010; Horecka et al., 2014; Kleindinst et al., 2014). An important feature of A. fundyense population dynamics and toxin production is the hypnozygote or resting cyst that is part of this organism's life history (Anderson, 1998).

    View all citing articles on Scopus
    1

    Current address: College of Marine Science, University of South Florida, St. Petersburg, FL 33701, USA.

    View full text