Introduction

Bottlenose dolphins (Tursiops truncatus) have been recognized as marine mammal sentinels in aquatic and coastal environments (Reddy et al., 2001; Wells et al., 2004; Bossart, 2006). Dolphins residing in nearshore habitats are exposed to an increasing variety of persistent pollutants (Berrow et al., 2002; Stein et al., 2003; Irwin, 2005) from anthropogenic sources that degrade their habitat, limit their food resources, and increase their susceptibility to diseases (Fair and Becker, 2000). As a long-lived species, dolphins are apex predators in the Indian River Lagoon (IRL), Florida, exhibit site fidelity to the region (Odell and Asper, 1990; Mazzoil et al., 2005), and serve as important indicators for the effects of exposure to pollutants in the marine environment.

The IRL was declared an Estuary of National Significance in 1990 by the United States (U.S.) Environmental Protection Agency (EPA). This designation was the result of expansive human population growth, habitat, fisheries, and species loss, invasive species, freshwater inflow, excessive nutrients, sedimentation, pollutants, and toxins (U.S. EPA, 1996) negatively impacting the most biodiverse estuary in North America (Gilmore, 1977). Key biological indicators, including dolphins, were identified for implementation of research, management, and protection actions in the IRL Comprehensive Conservation and Management Plan (U.S. EPA, 1996). Dolphins in this ecosystem, which spans over one-third of the east coast of Florida, are exposed to marked variation in water quality and localized pollution.

Pathologic evidence from stranded IRL dolphins showed a high prevalence of infectious and inflammatory diseases, suggestive of immunologic dysfunction (Bossart et al., 2003). Photo-identification surveys conducted in the IRL identified a variety of dermal lesions on dolphins (Mazzoil et al., 2003). Further, increased stranding incidents in 1996 and 2000 (Stolen et al., 2007), and an Unusual Mortality Event in 2001 (Marine Mammal Commission, 2002) prompted the initiation of a Bottlenose Dolphin Health and Risk Assessment (HERA) project in 2003, to investigate anthropogenic contaminants (e.g., trace metals, PCBs, organochlorine pesticides, perfluorinated compounds, and polybrominated diphenyl ethers) and infectious disease agents in this population (Fair et al., 2006).

Dolphins in the IRL form part of a complex mosaic of stocks identified along the eastern seaboard of the U.S. (Waring et al., 2002). For purposes of conservation under the U.S. Marine Mammal Protection Act of 1972, the coastal western North Atlantic bottlenose dolphin population is defined in seven management units from New Jersey to Florida (Waring et al., 2002). Additional resident estuarine stocks exist, but are not differentiated from the coastal management units, and are in need of further investigation (Waring et al., 2008). For example, vessel-based photo-identification research has indicated that year-round resident dolphins occur in the sounds and estuaries within the South Carolina management unit (Gubbins, 2002; Zolman, 2002; Speakman et al., 2006), the northern Florida unit (Caldwell, 2001), and the central Florida unit (Odell and Asper, 1990; Mazzoil et al., 2005). This article describes spatial distributions of dolphins within the IRL during a 3-year period (2002–2005). Knowledge of home ranges in conjunction with ongoing investigations of environmental pollution and health issues will serve to refine management strategies on both broad and fine geographic scales that are relevant to the ecology and behavior of IRL dolphins.

Methods

Study Area

The Indian River Lagoon is a 250-km linear estuary located on the east central coast of Florida. Four distinct water bodies, Mosquito Lagoon, Indian River, and Banana River, interconnected via man-made canals, and the St. Lucie Estuary, constitute the major habitats (Fig. 1). The system is bisected by an Intracoastal Waterway (ICW) that extends from Maine to Miami, Florida. The average depth of the lagoon is 1.5 m, with maximum depths averaging 3.7 m in the dredged channels and harbors; widths vary from a few meters to 8.9 km (Gilmore, 1977). Five inlets and the Canaveral Lock connect the IRL to the Atlantic Ocean.

Figure 1
figure 1

The Indian River Lagoon, Florida, U.S., sub-divided into six segments based on hydrodynamics and geographic features.

Full size image

The IRL ecosystem was divided into six segments based on hydrodynamics and geographic features for purposes of characterization and management (Woodward-Clyde Consultants, 1994) (Fig. 1). The northern segments (1A, 1B, and 1C) have wide expanses of water, small watersheds, and low tidal ranges and flushing. The large surface area results in precipitation and evaporation being the most important parameters of the hydrologic cycle, and also provides a potential for mixing and dilution. However, the lack of inlets or tidal exchange in the northern areas makes these habitats highly susceptible to pollutant loading. Water quality in the central and southern segments of the lagoon (2, 3, and 4) is influenced by infusions of water from flood control drainage canals, e.g., in particular, runoff from agricultural watersheds and freshwater releases from Lake Okeechobee (Sime, 2005). Discharges from these sources introduce higher amounts of nutrients, metals, pesticides, and suspended solids into the system (Woodward-Clyde Consultants, 1994).

Field Program

From September 2002 to August 2005, photographic surveys of dolphins were conducted once monthly using a variety of vessels, ranging from 6 to 8 m in length, operated at a speed of 10–12 km/h. Due to the large expanse of the survey area, multiple days were needed each month to assess the entire system. Surveys were conducted over consecutive days to the extent possible, with weather the most important factor in selecting days to ensure consistent sighting conditions. Monthly survey distance was determined using ESRI ArcGIS 9.0, by measuring the survey route in the ICW. This route allowed visual access to both shallow (<3 m) and deeper (≥3 m) sections of the IRL. Total survey distance for the complex was 344 km (segment 1A—52 km; segment 1B—38 km; segment 1C—82 km; segment 2—36 km; segment 3—35 km; segment 4—101 km). Field and photo-identification protocols have been described previously (Mazzoil et al., 2005).

Photo-identification Analyses

Field data were archived and analyzed with customized Microsoft Access software. Computer-based analyses of the digital images of dorsal fins were described previously (Mazzoil et al., 2004). In addition, the prototype image of each dolphin per sighting was assigned a score, based on a weighted system of photographic quality (Urian et al., 1999; Read et al., 2003) to reduce the chance of missing or making an incorrect identification (Friday et al., 2000). The distinctness of each dorsal fin was assigned a category (Urian et al., 1999; Read et al., 2003), and only those with major and intermediate features were used in the analyses.

Segment and Linear Analyses

Each dolphin (distinct and indistinct) was assigned to one of the six watershed hydrologic segments based on the GPS coordinates of a given encounter. The density of dolphins in each segment was calculated by dividing the total surface area of the segment in km2 by the total number of dolphins sighted in that segment. The re-sighting rate for distinct dolphins was calculated by dividing the total number of sightings for distinct individuals by the total number of distinct dolphins seen in each segment. Linear ranges were determined using a 1-km2 latitudinal grid system, designed in ESRI ArcINFO, which was layered over individual sighting records. The difference in grid numbers between the two most distant points for each dolphin represented the total linear distance traveled for each individual.

Environmental Data

Total surface water area and salinity were defined for each segment from the literature (Table 1). Anthropogenic influences on the lagoon include pollution from non-point sources (e.g., stormwater runoff) that freshen the water and introduce oxygen-demanding materials (plant and animal matter), suspended solids (resulting in higher and longer bacterial survival rates), excessive nutrients (nitrogen and phosphorous), and toxic metals (lead and zinc) into the system. The pollutant loading rates for: biological oxygen demand (BOD), total suspended solids (TSS), total nitrogen (TN), total phosphorous (TP), lead (PB), and zinc (Zn) were calculated for expected land use conditions through 2010 (Woodward-Clyde Consultants, 1994) and quantified for each segment (Table 1). The relationship between total dolphin population and dolphin density per km2 was evaluated by creating a correlation matrix for each demographic and environmental variable using Pearson’s correlation coefficients. A P value <0.05 was considered statistically significant.

Table 1 Total surface water area, salinity, and non-point source pollutant loading rates projected for each segment through 2010
Full size table

Results

Field Program

Photo-identification surveys encompassed a total of 2755 h, with 1821 h spent actively searching for dolphins along the survey route (“on” survey), and 934 h spent traveling to survey destinations and returning to port (“off” survey). If dolphins were encountered more than once while “on” survey or within the same survey month, only the first sighting was used in the analyses. “Off” survey sightings were not included in the analyses.

The mean number of days to complete each monthly survey was 8 (range: 4–12) and the mean number of total days spanned to complete each survey month was 20 (range: 8–28). Thirty-two of 36 monthly surveys, which covered the entire 344-km route, were completed. Four monthly surveys were incomplete due to mechanical failures, high winds, and multiple hurricanes. However, these less rigorous surveys (November 2002, December 2003, Apil 2004, and September 2004) obtained ample coverage of the entire complex (75–86%), and thus were included in the analyses.

Photo-identification Analysis

During the study period, 8686 dolphins were sighted during 1974 encounters and over 263,000 digital images were sorted, analyzed, and archived. Six-hundred and fifteen individual dolphins with distinct dorsal fins were identified.

Segment and Linear Range Analyses

Segment 1C (North Indian River) contained the highest number of dolphins (including indistinct dolphins) observed per area (n = 2262) and segment 3 (South-Central Indian River) the least (n = 966) (Table 2). However, the highest density of dolphins per km2 was found in segment 2 (North-Central Indian River). The density of dolphins per km2 in segment 2 was more than twice that of the density in segment 1B (Banana River).

Table 2 Number of distinct and indistinct dolphins sighted in each IRL segment by km2
Full size table

Distinct dolphins with ≥6 sightings (n = 339) were used for linear range analyses. This approach provided a sufficient number of individuals for analysis of general patterns. Seventy-five percent of recognizable dolphins (253 of 339) were sighted in ≤2 segments, and of these, 32% (109 of 339) were seen only in one segment. The majority of dolphins (98%) were seen in ≤3 segments and no dolphins were seen in all six segments of the system (Table 3). Dolphins in the Mosquito Lagoon (segment 1A) exhibited the strongest site fidelity, with 71% (67 of 94) seen exclusively therein. Dolphins in the Mosquito Lagoon also had the highest re-sighting rate (Table 4).

Table 3 Number and percentage of distinct dolphins with ≥6 sightings observed in one or more IRL segments
Full size table
Table 4 Re-sighting rate per segment for dolphins with ≥6 sightings
Full size table

The mean linear range for dolphins (n = 67) utilizing only the Mosquito Lagoon (segment 1A) was 22 km (range: 5–42 km). Dolphins (n = 27) seen in both the Mosquito Lagoon and Indian River, had average maximum distances of 10 km (range: 1–35 km) and 28 km (range: 4–88 km) per each water body, respectively. Only two dolphins seen in the Mosquito Lagoon were observed in Barge Canal, which connects the Banana River with the Indian River (Fig. 2).

Figure 2
figure 2

The linear home range represented by mean (rectangle) and extreme (line) for dolphins with ≥6 sightings.

Full size image

The mean linear range for dolphins (n = 56) observed in the Banana River (segment 1B) was 14 (range: 1–29 km) or 19 (range: 6–28 km) km in the Banana River, adjusted for entrance at the northern or southern entryways, and 38 km (range: 1–87 km) in the Indian River (Fig. 2). Two dolphins were sighted only within the Banana River. The mean linear range for dolphins (n = 56) observed in the St. Lucie River (segment 4) was 2 km (range: 1–8) and 55 km (range: 5–106) when also sighted in the Indian River. No dolphins used the St. Lucie River exclusively. The mean linear range for dolphins (n = 133) seen only in the Indian River was 54 km (range: 7–106 km) (Fig. 2).

Analysis of Environmental and Population Variables

No statistically significant correlations were found between total population or dolphin density per segment with any of the environmental variables (Table 5). Some environmental variables were highly correlated with each other within IRL segments.

Table 5 Relationship between total dolphins and dolphin density with environmental variables and IRL segment
Full size table

Discussion

Segment Analysis

These results add to the previous findings of site fidelity reported for dolphins from freeze-branding (Odell and Asper, 1990) and photo-identification studies (Mazzoil et al., 2005), and provide documentation for the persistence of a resident population of dolphins within the IRL over a 30-year period. Interestingly, the findings in this study identify geographic partitioning by dolphins within environmentally characterized regions of the system. The majority of dolphins (75%) were sighted in ≤2 adjacent segments, signifying long-term site fidelity (the tendency to remain in an area over an extended period) within relatively restricted home ranges (the area traversed by an individual for food gathering, mating, and caring for young) (Burt, 1943), forming overlapping communities (“distinct assemblages of dolphins that inhabit similar ranges and that interact more with each other than with adjacent assemblages”; Wells, 1986). These findings are similar to those reported for dolphins in Sarasota Bay, FL (Wells, 1991).

Linear Analysis

The mean linear distances traveled by dolphins were similar for dolphins seen exclusively in the Indian River (54 km) and for those found in the triangular configuration of the combined Indian and Banana Rivers (52–57 km), suggesting that these habitats may be similar in resources. Home range dimensions have been linked to habitat differences (Shane, 1990), food resources (Balance, 1992; Defran et al., 1999), and reproductive resources (Scott et al., 1990). Odell and Asper (1990) found that dolphins using both the Indian and Banana Rivers averaged 30–45 km in total range versus 12 km for those dolphins only in the Indian River. The greater linear distances traversed in the current study may reflect a more comprehensive and longer duration of survey coverage.

The smaller mean distance traveled by dolphins in the Mosquito Lagoon can be associated with its non-impaired waterway status (Florida Department of Environmental Protection, 2007). This segment may be richer in prey resources that are uniformly distributed in space and time, and can support more dolphins as seen in South Carolina (Gubbins, 2002). However, travel distance may be an artifact of survey effort in this region, as coverage did not extend past the northern boundary of Ponce Inlet. For example, dolphins seen in the northern Indian River were reported as far as St. Augustine (Odell and Asper, 1990) and a dolphin freeze-branded in the Mosquito Lagoon in 2004 (Fair et al., 2006) stranded alive in St. Augustine [W. Noke, personal communication, 2006], suggesting that at least some of the dolphins in the Mosquito Lagoon range further north.

The linear analyses included single “outlier” sightings, likely overestimating the range size for some dolphins. Even the largest range (106 km) should be considered relatively small, given no geographic barriers preclude movement between segments. Further, while some home ranges overlapped, core areas (areas used more frequently than others) (Ewer, 1968) did not, for the four dolphins seen in the northernmost (segment 1B) and southernmost (segment 4) Indian River. Spatial separation patterns indicated disparate communities for each outermost region: Mosquito Lagoon, and the northern and southern portions of the Indian River. Genetic analyses would be informative to determine the degree of relatedness between these communities.

Environmental Analysis

No statistically significant correlations were observed between total population or dolphin density and surface water area, salinity, or contaminant loads (BOD, TSS, TN, TP, Pb, Zn) within the six segments of the lagoon. These results suggest that other factors determine dolphin distribution patterns in the IRL and have important implications for the development of protection and management strategies. For example, dolphins exhibit strong fidelity to well-defined segments of the system and do not avoid areas with poor environmental conditions (i.e., the high volume of freshwater intrusion in the St. Lucie Estuary and the high concentrations of nitrates and phosphates from agricultural activity). Similarly, resident dolphins in the Moray Firth, Scotland, exhibit persistent geographical segregation, indicating that they do not move freely within the Firth, and thus may be more susceptible to localized disturbance or pollution (Wilson et al., 1997). Along the Texas coast, dolphins consistently did not avoid slick and oil sheens from a major oil spill (Smultea and Würsig, 1995) and did not leave their preferred habitat during a 3-month-long red-tide event (Irwin, 2005), thereby increasing their exposure to chemicals and biotoxins. Exposure to brevetoxins produced by harmful alga blooms has been associated with marine mammal mortalities (Bossart et al., 1998; Flewelling et al., 2005; Fire et al., 2007), suggesting that dolphins are vulnerable to localized health hazards.

The concept that dolphins can simply move if their environment is disturbed reflects a common misperception of the marine environment as a uniform space (Wilson et al., 1997) and does not account for other influences on distribution such as prey availability (Shane, 1990), predation risk (Corkeron, 1990; Heithaus and Dill, 2002), calving events (Mann et al., 2000), gender (Wells et al., 1990), and social organization (Bräger, 1999). Further, when negative impacts are not uniformly distributed, some population units may be overexploited while others remain intact (Wade and Angliss, 1997). Two examples of regional site fidelity and localized disturbances are illustrated for the IRL, whereas a third example demonstrates the impact of disease due to overlapping community boundaries.

Example 1

An Unusual Mortality Event was declared by the National Marine Fisheries Service in 2001 (Marine Mammal Commission, 2002) when a record number of dolphin deaths occurred in the IRL within a 4-month period. Most dolphins (32 of 34) stranded in a small geographic area (Stolen et al., 2007), in segments 1B, 1C, and the northern tip of segment 2. The stranding locations were representative of the home range characteristics of dolphins seen within the Indian and Banana River triangular complex in this study, indicating that the cause of the mortality was confined to a particular geographic range. The etiology of these mortalities remains undetermined, but a link to saxitoxin producing puffer fish poisoning has been suggested (Landsberg et al., 2006).

Example 2

Lobomycosis is a chronic, mycotic disease caused by a yeast-like organism (Lacazia loboi), and found only in humans and dolphins (Rodriguez-Toro, 1993). A 30% (9 of 30) prevalence of lobomycosis was reported among HERA dolphins captured in segment 4 (South Indian River), with no cases detected among 45 dolphins captured in the northern sections (Reif et al., 2006). The localization of photographically identified lobomycosis among dolphins with home ranges in segment 3 (South-Central Indian River) and segment 4 (South Indian River) (Murdoch et al., 2008), characterized by freshwater intrusion and lower salinity, suggests that exposure to localized environmental conditions may contribute to the unusually high prevalence of the disease. The source of the organism for the disease in humans is believed to be an aquatic environment or soil and vegetation associated with the natural habitat of the fungus (Rodriguez-Toro, 1993). The specific environmental factors contributing to the distribution of the disease in dolphins remain unknown.

Example 3

Bossart et al. (2005) reported oral and genital neoplasms among IRL dolphins captured during HERA between 2003 and 2005. Evidence suggests that these tumors contain a papillomavirus as well as a herpesvirus (Bossart et al., 2005; Rehtanz et al., 2006), and are spread horizontally by dolphin-to-dolphin contact. Home ranges for these dolphins included segments 1A, 1B, 1C, 2, and 4. Since dolphins in the IRL exhibit a fluid social structure (Kent et al., 2008), are promiscuous breeders, and use adjacent segments of the IRL as shown in this report, these and other infectious disease agents have the potential to be transmitted throughout the ecosystem.

Conclusion

Overall findings suggest that at least three different dolphin communities exist within the IRL as important functional elements of the ecosystem and should be managed, at minimum, as a separate stock along the western North Atlantic seaboard. IRL dolphins appear to be susceptible to localized environmental disturbances as well as far-reaching epizootic events. The limited ranging behavior of these dolphins suggests that applying the appropriate spatial scales to assess deleterious environmental impacts will be necessary to fully address localized conservation and management strategies.

Recent studies have highlighted the significance of long-term monitoring data on dolphin habitat use, environmental characterization, and health (Wells et al., 2005; Adams et al., 2008; Litz et al., 2008). As additional data regarding the health status of IRL dolphins become available from capture/release health assessments (Bossart et al., 2006), they should be evaluated by incorporating the ranging characteristics of individuals in order to explore environmental factors that may adversely impact dolphin health. The approach outlined here will be used in the future to explore fine-scale variation in the distribution of contaminants which may, in turn, provide clues as to their origins and inform management strategies to mitigate these exposures. Conversely, pooling health data among all IRL dolphins may obscure intra-regional variability and obscure exposure to small-scale environmental hazards.

Future investigations to elucidate dolphin population structure and habitat use within the IRL will be essential for developing appropriate strategies to protect and monitor their health. Such investigations should include using photo-identification in conjunction with other field techniques (e.g., telemetry) as tools to study population dynamics and life history, and monitor spatial and temporal trends, the prevalence/occurrence of dermal disease, reproductive success, and mortality patterns. Furthermore, field studies should be conducted in concert with biopsy sampling—to delineate genetic structure within the various communities, and measure individual fitness and population viability—and capture-release sampling of dolphins for health assessment.

Longitudinal trends in home range patterns and health status should be combined and used to measure the results of restoration efforts undertaken by the U.S. EPA National Estuary Program (1996), the Comprehensive Everglades Restoration Plan (2000), and others to reduce pollutants and restore water quality in the IRL. As key components in monitoring the environment (Wells et al., 2004; Bossart, 2006), dolphins play an integral role in mankind’s stewardship of the irreplaceable and valued resources of the Indian River Lagoon.