bioRxiv preprint doi: https://doi.org/10.1101/669523; this version posted June 12, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. 1
1 Full Title: Spatial distribution of epibionts on olive ridley sea turtles at Playa Ostional, Costa Rica
2
3 Nathan J. Robinson1,2,*, Emily Lazo-Wasem3, Brett O. Butler4, Eric A. Lazo-Wasem5, John D. Zardus6,
4 Theodora Pinou7
5
6 1The Leatherback Trust, Goldring-Gund Marine Biology Station, Playa Grande, Guanacaste, Costa
7 Rica
8 2Cape Eleuthera Institute, The Cape Eleuthera Island School, Cape Eleuthera, Eleuthera, The
9 Bahamas
10 3Lyles School of Civil Engineering, Purdue University, West Lafayette, Indiana, U.S.A
11 4Museo de Zoología “Alfonso L. Herrera”, Facultad de Ciencias, Universidad Nacional Autónoma de
12 México, A.P. 70¬–399, Ciudad de México CP 04510, México
13 5Division of Invertebrate Zoology, Peabody Museum of Natural History, Yale University, New Haven,
14 Connecticut, U.S.A
15 6Department of Biology, The Citadel, Charleston, South Carolina, U.S.A
16 7Department of Biological and Environmental Sciences, Western Connecticut State University,
17 Danbury, Connecticut, U.S.A
18 *Corresponding author: [email protected]
19
20 Short title: Spatial distribution of epibionts on sea turtles bioRxiv preprint doi: https://doi.org/10.1101/669523; this version posted June 12, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. 2
21 ABSTRACT
22 There is a wealth of published information on the epibiont communities of sea turtles, yet many of
23 these studies have exclusively sampled epibionts found only on the carapace. Considering that
24 epibionts may be found on almost all body-surfaces and that it is highly plausible to expect different
25 regions of the body to host distinct epibiont taxa, there is a need for quantitative comparative
26 studies to investigate spatial variation in the epibiont communities of turtles. To achieve this, we
27 measured how total epibiont abundance and biomass on olive ridley turtles Lepidochelys olivacea
28 varies among four body-areas of the hosts (n = 30). We show that epibiont loads on olive ridleys are
29 higher, both in terms of number and biomass, on the skin than they are on the carapace or plastron.
30 This contrasts with previous findings for other hard-shelled sea turtles, where epibionts are usually
31 more abundant on the carapace. Moreover, the arguably most ubiquitous epibiont taxon for other
32 hard-shelled sea turtles, the barnacle Chelonibia spp., only occurs in relatively low numbers on olive
33 ridleys, while the barnacles Stomatolepas elegans and Platylepas hexastylos are far more abundant.
34 We postulate that these differences between the epibiont communities of different sea turtle taxa
35 could indicate that the carapaces of olive ridley turtles provide a more challenging substratum for
36 epibionts than do the hard shells of other sea turtles. In addition, we conclude that it is important to
37 conduct full body surveys when attempting to produce a holistic qualitative or quantitative
38 characterization of the epibiont communities of sea turtles.
39
40 KEYWORDS
41 Epibiosis, Lepidochelys olivacea, East Tropical Pacific, arribada, barnacles, turtle-scape, settlement
42 substratum, site selection
43
44 INTRODUCTION
45 In the marine environment, almost any non-toxic, non-protected surface will eventually be colonized
46 by an array of microorganisms, plants, algae, or animals [1,2]. This is true for non-living substrata, bioRxiv preprint doi: https://doi.org/10.1101/669523; this version posted June 12, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. 3
47 such as rocks, sand grains, or the shells of dead molluscs, as well as the bodies of living marine
48 animals. Those organisms that live on the surfaces of other organisms are referred to as epibionts
49 [3], a grouping of diverse taxa with a wide-range of life-histories, physiological tolerances, and
50 mechanisms for attaching to their host [4]. It is therefore not surprising to discover that epibiont
51 communities vary among different host species [5,6], populations [7], and even between different
52 body-areas on a single host [8,9]. Understanding the factors driving spatial variation in the
53 distribution of epibionts, be it on the scale of an ocean or a host’s body, can shed light the habitat
54 requirements of these varied associates. In turn, this information can be used to understand habitat
55 preferences or behaviour of the host [e.g. 10,11] and even epibiont-related disease transmission
56 [e.g. 12].
57
58 Sea turtles arguably host the most abundant and diverse epibiont communities of all large marine
59 megafauna. In an effort to characterize these epibiont communities, researchers have generated
60 hundreds of species-lists reporting the presence of epibiont taxa for sea turtle populations around
61 the world [13]. As these datasets have grown, there is now an impetus to synthesise these data and
62 elucidate epibiont community spatial patterns globally and make quantitative assessments of
63 epibiont loads [14]. Yet to help draw robust conclusions about geographic differences in epibiont
64 communities, it would be beneficial to understand spatial patterns in epibiosis at a finer-scale – the
65 scale of a turtle’s body. Indeed, it is highly likely that epibionts are non-uniformly distributed over a
66 turtle host for several reasons. For example, the physical properties of certain body-parts, such as
67 the carapace, differ from those of the plastron or skin [15], which likely affects attachment or
68 settlement of different epibiont species. Different locations on the host’s body may also present
69 different opportunity for feeding. For example, some epibiont species feed preferentially on certain
70 tissues and thus will be found predominantly in those areas [16] whereas filter feeders might prefer
71 to be in areas where water flow over the host is optimal [17]. Either of these factors, as well as bioRxiv preprint doi: https://doi.org/10.1101/669523; this version posted June 12, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. 4
72 attachment mode, crypsis, mating habits, and others, could lead to distinct variation in the
73 distribution of epibionts on sea turtles.
74
75 While it may be clear that epibiont distribution is likely to vary over the host’s body, many studies on
76 sea turtle epibiosis have either not provided any quantitative assessment of where on the host the
77 epibionts were found [e.g. 7,18], and/or have simply sampled epibionts from a single location on the
78 animal, commonly the carapace [e.g. 19,20]. Tellingly, even some of the most seminal studies
79 investigating the distribution patterns of epibionts on the loggerhead turtle Caretta caretta and
80 green turtle Chelonia mydas hosts have focused exclusively on the carapace [8,21]. There are only
81 two studies of which the authors are aware that have compared the abundance of epibionts from
82 over the entire body: Devin and Sadeghi (2010) [22] and Razaghian et al. (2019) [23]. Interestingly,
83 both studies, which focused on hawksbill turtles Eretmochelys imbricata nesting in Iran, identified
84 that epibiotic barnacles were actually more common on the plastron than the carapace. If this is true
85 for all sea turtle species, then this finding could have important implications for epibiont sampling.
86 Specifically, if epibionts are not uniformly distributed on the hosts’ body then the actual biodiversity
87 of epibionts might be underestimated if only sampled from a single region of the body.
88
89 To better understand spatial variation in epibiont communities on sea turtle hosts, we made a
90 quantitative assessment of the abundance and biomass of several epibiotic taxa from different
91 regions of the body of olive ridley sea turtles Lepidochelys olivacea. This species was chosen because
92 several studies have focused on characterizing the epibiont communities of this species in recent
93 years [e.g. 6,11,18,24] yet no previous studies have specifically looked at epibiont distributions for
94 this host species. Our null prediction was that there would be no variation in the abundance or total
95 biomass of the various epibiont taxa among the different body-areas.
96
97 METHODS bioRxiv preprint doi: https://doi.org/10.1101/669523; this version posted June 12, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. 5
98 Study Site
99 Olive ridley turtles were sampled from Playa Ostional, which is located on the Pacific coast of the
100 Nicoya Peninsula in northwest Costa Rica (9° 59’ N, 85° 42’ W). We selected this site as it is one of a
101 few beaches worldwide that has mass-nesting assemblages or “arribadas” of olive ridley turtles ,
102 thus making it feasible to sample large numbers of turtles relatively easily. These arribadas generally
103 occur once a month and last three to eight days [25]. At Ostional, over 500,000 turtles have been
104 estimated to nest in a single arribada event [26] at densities of over 4 clutches m-2 [27]. For this
105 study, we sampled epibionts from a total of 30 turtles spanning three separate arribada events.
106 Specifically, we sampled 10 turtles on 6-Dec-2015, 5 turtles on 1-Jan-2016, and 15 turtles on 7-
107 Feburary-2016.
108
109 Sampling Protocol
110 During arribada events, we patrolled the beaches of Ostional to encounter nesting turtles. To avoid
111 any possibility that sampling would interrupt the nesting process, we only examined turtles after
112 they had completed oviposition. In addition, we sampled only animals that appeared to be in good
113 health with no visible injuries, lesions, or tumours as certain epibionts are often found in association
114 with certain injuries [28,29].
115
116 When a suitable turtle was encountered, we measured its Curved Carapace Length using a flexible
117 tape measure. The turtle’s flippers were then restrained by hand and the animal was moved onto a
118 plastic tarp. This limited the ability of the turtle to flick sand on its carapace, which would impede
119 efforts to sample epibionts. We then exhaustively collected epibionts by scraping or prying them off
120 the turtle using a knife or tweezers. After all visible epibionts were collected from the dorsal surfaces
121 of the turtle, we would briefly (< 10 mins) flip the animal onto its carapace to collect epibionts from
122 its ventral surfaces. All epibionts were divided between separate containers by the region of the
123 body where they were collected. We divided the body into four regions: (1) the head, shoulder, and bioRxiv preprint doi: https://doi.org/10.1101/669523; this version posted June 12, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. 6
124 fore flippers (subsequently termed head), (2) the tail, cloaca, hind flippers, and inguinal cavities
125 (subsequently termed tail), (3) the carapace, or (4) the plastron (Figure 1). Epibiont samples were
126 preserved in 75% non-denatured ethanol following the protocols outlined in Lazo-Wasem et al.
127 (2011) [18]. No attempt was made to quantify the abundance or distribution of micro-epibionts,
128 such as diatoms, even though it is known that they exist in high abundances on sea turtles [6,24].
129 We did not tag any of the turtles, yet due to high numbers of individuals at each arribada it was easy
130 to ensure that the same individual was not sampled repeatedly.
131
132 All epibionts were identified to the lowest taxonomic level by consulting appropriate literature. The
133 samples were then enumerated, dried, and weighed to the nearest 0.00001g using a Mettler B6
134 Semi-Micro Balance. Samples were catalogued and deposited in the Yale Peabody Museum of
135 Natural History, U.S.A. Specimen records are available at http://peabody.yale.edu/collections.
136
137 To test our null prediction that there would be no variation in the abundance or total biomass of the
138 epibiont taxa between the different regions of the turtles’ body, we used Kruskal-Wallace tests with
139 ad-hoc pair-wise Mann-Whitney tests. Separate tests were run for each major epibiont taxa. Kruskal-
140 Wallace and Mann-Whitney tests were conducted using PAST V.3.13, confirming significance when p
141 ≤ 0.05. We chose these non-parametric tests, over their parametric counterparts, because the data
142 included many zeros and were not normally distributed.
143
144 RESULTS
145 Of the 30 olive ridley turtles that were sampled (CCL: 66 – 69 cm range), epibionts were found on all
146 but one individual. In total, we collected 1614 individual epibionts, which collectively weighed 14.01
147 g. The mean number of epibionts per host was 53.80 with a mean cumulative mass of 0.47 g. The
148 epibiont load by body region (considering all epibiont taxa) was highest on the tail with 59.23 % of
149 epibionts being found in this location and constituting 64.11 % of the total epibiont biomass (Figure bioRxiv preprint doi: https://doi.org/10.1101/669523; this version posted June 12, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. 7
150 2). The location with the next highest epibiont load was the head with 31.69 % of epibionts being
151 found in this location and constituting 24.83 % of the total epibiont biomass. For both the carapace
152 and plastron, epibiont load was very low and never exceeded more than 8 % of the total epibiont
153 load.
154
155 We conservatively estimate that the diversity of epibionts represented 20 different taxa (see Table 1
156 for a full list). Most of these taxa were relatively rare and only five were found in mean abundances
157 exceeding one individual per host. In alphabetical order these were: Balaenophilus manatorum,
158 Chelonibia testudinaria, Platylepas hexastylos, Stomatolepas elegans, and small individuals of family
159 Platylepadidae that cannot as yet be identified. This platylepadid may comprise more than one
160 species, but for now the most conservative approach is to view it as a single operational taxonomic
161 unit.
162
163 The most common taxon was Stomatolepas elegans, which constituted 38.85 % and 63.30 % of the
164 total epibiont load in terms of abundance and mass respectively. Platylepas hexastylos was almost
165 as abundant, constituting 29.30 % of all epibionts but individuals were generally lower in mass than
166 Stomatolepas elegans and so constituted only 17.24 % of the total epibiont biomass. In contrast,
167 Chelonibia testudinaria was not very abundant, comprising only 5.27 % of all epibionts, yet it
168 constituted 16.54 % of total epibiont biomass. Balaenophilus manatorum was relatively abundant,
169 constituting 17.66 % of all epibiont individuals, yet the combined biomass of this species was less
170 than 0.01 g and so relatively negligible. Similarly, Platylepadidae spp. constituted 5.70 % of the total
171 epibiont abundance yet only 0.8 % of the total epibiont biomass, further indicating that these
172 individuals were very small. The remaining 15 taxa constituted only 3.34 % and 2.09 % of the total
173 epibiont load in terms of abundance and biomass respectively (Figure 3).
174 bioRxiv preprint doi: https://doi.org/10.1101/669523; this version posted June 12, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. 8
175 Kruskal-Wallis Test tests indicated significant differences between both epibiont abundance and
176 biomass for the different body regions (abundance: H(χ2) = 23.42, d.f. = 3, p < 0.0001; biomass = H(χ2)
177 = 16.42, d.f. = 3, p < 0.0001). Thus, we can reject the null prediction that epibionts are uniformly
178 distributed on their hosts (Figure 4).
179
180 When we tested for differences in the abundance or biomass of the five major epibiont taxa
181 between the different body region, we observed significant differences in the abundance and
182 biomass of Stomatolepas elegans (abundance: H(χ2) = 25.49, d.f. = 3, p < 0.0001; biomass = H(χ2) =
183 15.67, d.f. = 3, p < 0.0001) and Platylepas hexastylos (abundance: H(χ2) = 16.87, d.f. = 3, p < 0.0001;
184 biomass = H(χ2) = 15.67, d.f. = 3, p < 0.0001). Ad-hoc pair-wise Mann-Whitney tests revealed that for
185 both Stomatolepas elegans and Platylepas hexastylos there were no statistical differences in either
186 abundance or biomass between head or tail sections (p < 0.0001 for all tests). Instead, statistical
187 differences were observed in abundance or biomass between both the head and tail when
188 compared to either carapace or plastron (p > 0.5 for all tests).
189
190 No significant differences were observed in the abundance or biomass of Chelonibia testudinaria
191 (abundance: H(χ2) = 2.94, d.f. = 3, p = 0.1973; biomass = H(χ2) = 1.613, d.f. = 3, p = 0.4650),
192 Balaenophilus manatorum (abundance: H(χ2) = 0.2748, d.f. = 3, p = 0.5139; biomass = H(χ2) = 0.2727,
193 d.f. = 3, p = 0.5173), and Platylepadidae spp. (abundance: H(χ2) = 0.4883, d.f. = 3, p = 0.5830; biomass
194 = H(χ2) = 0.4834, d.f. = 3, p = 0.5873) between the different body areas.
195
196 DISCUSSION
197 Spatial distribution of epibionts
198 Knowledge of how sea turtle epibionts are distributed on their hosts can provide valuable insights
199 into the ecology and habitat requirements of these extracorporeal companions. Furthermore, when
200 epibionts are not uniformly distributed over the host then sampling only from certain body-parts bioRxiv preprint doi: https://doi.org/10.1101/669523; this version posted June 12, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. 9
201 runs the risk of under-sampling and misrepresenting certain epibiont taxa in qualitative assessments
202 and certainly in quantitative evaluations. To test this concern, we measured how the abundance and
203 biomass of epibionts varies between different body regions on host olive ridley turtles. We
204 discovered that epibiont loads on olive ridley turtle are highest, both in terms of number and
205 biomass, on the soft tissues of the head and tail than they are on the carapace or plastron.
206 Specifically, 60 % of the total number of epibionts was found on the tail and 32 % was found on the
207 head. In contrast, only 6 % of the total number of epibionts was found on the carapace and 4 % on
208 the plastron. This finding suggests that studies exclusively sampling epibionts from the carapace, or
209 any other specific, body-part could significantly under sample the total epibiont load and run the risk
210 of not detecting or at least underrepresenting those taxa that are not commonly found on the
211 sampled body-part. For example, the most common epibiont taxon recorded in this study,
212 Stomatolepas elegans was absent from the carapace or plastron even though found in high
213 abundance on the tail and head. Somewhat predictably, due to the intimate association between
214 Balaenophilus manatorum and Stomatolepas elagans [28], Balaenophilus manatorum was also only
215 reported in substantial numbers on the tail section where Stomatolepas elagans was equally
216 common and barely recorded elsewhere on the turtles. Similarly, Platylepas hexastylos was over 5
217 times more abundant on the head and tail than it was on the carapace or plastron.
218
219 It is also interesting to note that the total biomass of Stomatolepas elegans and Platylepas
220 hexastylos individuals exceeded that of Chelonibia testudinaria despite individuals of the latter being
221 heavier. Indeed, individual Stomatolepas elegans and Platylepas hexastylos weighed an average of 0.
222 4 g and 0.2 g respectively, while Chelonibia testudinaria weighed on average 0.8 g. Thus, from an
223 ecological standpoint the small, less conspicuous epibionts, scattered in high densities in the soft-
224 tissues, and consequently more difficult to find, may be the most ecologically important for the host
225 animal.
226 bioRxiv preprint doi: https://doi.org/10.1101/669523; this version posted June 12, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. 10
227 Thus, it is clear that to get a true understanding of the epibionts on sea turtles, studies should
228 sample all body-surfaces. This is necessary to build both an understanding of the habitat
229 requirements of epibiont species and a methodology for using them as biological indicators. For
230 example, Balaenophilus manatorum has been previously suggested as a vector for
231 fibropapillomatosis [28] and this prediction has been further supported by more recent studies
232 [16,30]. The presence of this species could therefore be important to monitor in reference to turtle
233 health. Yet if researchers only sample epibionts from the carapace and plastron, they limit their
234 capacity to identify the presence or absence of this species. We therefore recommend that future
235 studies attempt to survey all body-sections on turtles when either attempting to characterize the
236 entire epibiont community quantitatively or when specifically trying to determine the presence or
237 absence of taxa qualitatively.
238
239 Differences between host species
240 The discovery that epibiont loads are higher for olive ridley turtles on the soft tissues, such as the
241 head and tail, relative to the carapace and plastron was somewhat unexpected. Indeed, comparable
242 studies that conducted quantitative assessments of epibiont distributions on hawksbill turtles found
243 the majority of epibiotic barnacles were found on the plastron and carapace respectively [22,23].
244 Furthermore, it has often been assumed although there is little quantitative data to support it, that
245 for loggerhead and green turtles the highest quantities of epibionts are found on the carapace [see
246 31].
247
248 The distinct difference in the distribution of epibionts between these different sea turtle species
249 begs the question of why such inter-specific differences should occur. One hypothesis could be that
250 these discrepancies are driven by differences in the behaviour of each host species, exposing them
251 to settlement by different larval epibionts with different habitat preferences (i.e. reef, seagrass
252 meadow, or soft sediment), yet we consider this to be unlikely considering the large range and bioRxiv preprint doi: https://doi.org/10.1101/669523; this version posted June 12, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. 11
253 behaviour overlaps between many turtle species. Instead, we propose that differences in the
254 physical properties (e.g. microstructure, chemical composition, physical properties) of the carapace
255 and dermis among different turtle species drive epibiont settlement.
256
257 Supporting the notion that olive ridley carapaces are a challenging substrate for epibionts settlement
258 is our observation of the relative paucity of Chelonibia barnacles associated with this host.
259 Chelonibia spp. are common taxa on the carapaces of loggerhead turtles [19, 20], hawksbill turtles
260 Eretmochelys imbricata [32], and green turtles [6]; however, we observed a mean of only 2.8
261 individuals of Chelonibia testudinaria per host, which is far lower than the observed 16.8 individuals
262 per host on green turtles that share nesting beaches with olive ridley turtles in Costa Rica [6]. The
263 Chelonibia testudinaria found on olive ridley turtles appeared far smaller than those on these green
264 turtles (NJ Robinson, personal observation) suggesting that maybe they are not surviving long
265 enough on this host to grow to their full size. In contrast, the most common epibiont observed on
266 olive ridley turtles was Stomatolepas elegans, which unlike Chelonibia testudinaria, tends to bury
267 into the softer tissues using an array of plates to hold itself in place.
268
269 Geographic patterns in epibiont communities in olive ridleys
270 In recent years, the epibiont communities of olive ridley sea turtles in the East Pacific Ocean have
271 been described in several studies [6,11,18,24]. Collectively between these four studies, a total of 291
272 olive ridley turtles have been sampled for epibionts. From this extensive dataset, it is clear that while
273 the epibiont communities of olive ridley turtles in the East Pacific are diverse and new epibiont
274 associations have been revealed in each study, these epibiont communities are generally dominated
275 by a few main taxa. The taxa include: the barnacles Chelonibia testudinaria, Conchoderma virgatum,
276 Lepas spp. Platylepidaeas decorata (which we describe as Platylepadidae spp. in study for reasons
277 described earlier), Platylepas hexastylos, and Stomatolepas elegans; the leech Ozobranchus bioRxiv preprint doi: https://doi.org/10.1101/669523; this version posted June 12, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. 12
278 branchiatus, the amphipod Podocerus chelonophilus, and the copepod Balaenophilus manatorum.
279 All other taxa tend to be recorded at relatively low frequencies (i.e. less than one per 50 hosts).
280
281 This is not to say, however, that these more commonly observed taxa for East Pacific olive ridley
282 turtle are found on every turtle. In fact, even these commonly observed epibiont taxa are noticeably
283 absent from some studies. For example, Podocerus chelonophilus was observed on more than 25 %
284 of the olive ridley turtles sampled both on the beach of Playa Grande [6] and in the nearby waters El
285 Coco [11]. In this study on Playa Ostional however, we recorded a single individual only of Podocerus
286 chelonophilus among all 30 turtles, even though this beach is less than 50 km away from the
287 previously mentioned sampling locations. Majewska et al. (2015) also reported a similarly low
288 prevalence of Podocerus chelonophilus from olive ridley turtles sampled at Playa Ostional, with only
289 4 % of all the turtles hosting this epibiont [24]. Lepas spp. are also noticeably less prevalent at Playa
290 Ostional, relative to other nearby locations [6,11,24].
291
292 One possible hypothesis to explain differences in the abundances of these common epibiont taxa
293 between turtles sampled on Playa Ostional compared to those sampled at other locations nearby is
294 that Playa Ostional experiences different oceanographic conditions. While previous studies have
295 indeed noted the strong presence of offshore currents on Playa Ostional, relative to other nesting
296 beaches on the Pacific coast of Costa Rica [33], it seems unlikely that this is the primary factor
297 driving the differences in epibiont communities between these locations. This is because nesting
298 olive ridley turtles are known to move over larger distances between nesting events than the
299 distances between these beaches (< 50 km) [34], and thus there is likely significant overlap in the
300 inter-nesting habitats of these turtles even though they nest on different beaches. Instead, we
301 propose that a more robust hypothesis is that the differential prevalence of certain epibiont taxa is
302 linked to differences in behaviour between turtles nesting at each site. Indeed, Playa Ostional is a
303 well-known arribada site where thousands of turtles aggregate in mass nesting events, while Playa bioRxiv preprint doi: https://doi.org/10.1101/669523; this version posted June 12, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. 13
304 Grande and the waters of El Coco likely host turtles that are solitary nesters. Perhaps during such
305 mass mating events there is a much greater chance for epibionts to be scraped off than there is in
306 smaller mating events, meaning the arribada nesting turtles have fewer epibionts. Alternatively, as
307 many epibionts, such as Podocerus chelonophilus and Balaenophilus manatorum are not free
308 swimming and must likely be transferred on contact between hosts [30], it could be that the great
309 potential for the transfer of individuals between hosts means that they even though they might be
310 found on more hosts, they occur in lower numbers per host. Understanding the factors that are
311 driving distinct variation in the presence and absence of these common epibiont species would be
312 an interesting avenue for future research and could reveal several aspects about the behaviour of
313 their hosts that might be challenging to detect using more conventional methods [30,35].
314
315 ACKNOWLEDGEMENTS
316 Epibiont collection was conducted under MINAE permits (ACT-ORDR-099-15 and ACG-PI-058-2015).
317 This research was performed in accordance with the Purdue University Animal Care and Use
318 Committee. Samples were exported from Costa Rica under a SINAC permit (DGVS-183-2016). Pilar
319 Santidrián Tomillo, Christian Diaz Chuquisengo, Elizabeth Solano, and Myriam Norori helped with
320 permitting. Lourdes Rojas aided with identifying taxa.
321
322 AUTHOR CONTRIBUTIONS
323 NJR wrote the paper with input from all authors. Conceived and designed the experiments: NJR, EAL-
324 W, JDZ, and TP. Collected samples in the field: NJR, BOB. Taxonomic analysis of samples: NJR, EAL-W,
325 and JDZ. Laboratory analyses: EL-W, EAL-W. Data analysis: NJR.
326
327 REFERENCES
328 [1] Wahl M. Marine epibiosis. I. Fouling and antifouling: some basic aspects. Marine Ecology Progress
329 Series 1989;58:175-189. bioRxiv preprint doi: https://doi.org/10.1101/669523; this version posted June 12, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. 14
330
331 [2] Harder T. Marine epibiosis: concepts, ecological consequences and host defence. In: Springer
332 Series on Biofilms. Springer, Berlin, Heidelberg. 2008;pp. 219-231.
333
334 [3] Taylor PD, Wilson MA. A new terminology for marine organisms inhabiting hard substrates.
335 Palaios. 2002;17:522-525.
336
337 [4] Wahl M. Epibiosis. In: Marine Hard Bottom Communities. Springer, Berlin, Heidelberg. 2009;pp.
338 61-72.
339
340 [5] Hayashi R. A checklist of turtle and whale barnacles (Cirripedia: Thoracica: Coronuloidea). Journal
341 of the Marine Biological Association of the United Kingdom 2013;93:143-182.
342
343 [6] Robinson NJ, Lazo-Wasem EA, Paladino FV, Zardus JD, Pinou T. Assortative epibiosis of
344 leatherback, olive ridley, and green sea turtles in the Eastern Tropical Pacific. Journal of the Marine
345 Biological Association of the United Kingdom 2017;97:1233-1240.
346
347 [7] Domènech F, Badillo FJ, Tomás J, Raga JA, Aznar FJ. Epibiont communities of loggerhead marine
348 turtles (Caretta caretta) in the western Mediterranean: influence of geographic and ecological
349 factors. Journal of the Marine Biological Association of the United Kingdom 2015;95:851-861.
350
351 [8] Pfaller JB, Bjorndal KA, Reich KJ, Williams KL, Frick MG. Distribution patterns of epibionts on the
352 carapace of loggerhead turtles Caretta caretta. Journal of the Marine Biological Association of the
353 United Kingdom Biodiversity Records 2008;5381:1-4.
354 bioRxiv preprint doi: https://doi.org/10.1101/669523; this version posted June 12, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. 15
355 [9] Carrillo JM, Overstreet RM, Raga JA, Aznar FJ. Living on the edge: settlement patterns by the
356 symbiotic barnacle Xenobalanus globicipitis on small cetaceans. PLoS ONE 2015;10:e0127367.
357
358 [10] Reich KJ, Bjorndal KA, Frick MG, Witherington BE, Johnson C, Bolten AB. Polymodal foraging in
359 adult female loggerheads (Caretta caretta). Marine Biology 2010;157:113-121.
360
361 [11] Robinson NJ, Figgener C, Gatto C, Lazo-Wasem EA, Paladino FV, Santidrián Tomillo P et al.
362 Assessing potential limitations when characterising the epibiota of marine megafauna: effect of
363 gender, sampling location, and inter-annual variation on the epibiont communities of olive ridley sea
364 turtles. Journal of Experimental Marine Biology and Ecology 2017;497:71-77.
365
366 [12] Greenblatt RJ, Work TM, Balazs GH, Sutton CA, Casey RN, Casey JW. The Ozobranchus leech is a
367 candidate mechanical vector for the fibropapilloma-associated turtle herpesvirus found latently
368 infecting skin tumors on Hawaiian green turtles (Chelonia mydas). Virology 2004;321:101-110.
369
370 [13] Frick MG, Pfaller JB. Sea Turtle Epibiosis. In: The Biology of Sea Turtles 3. 2013;p.p. 399.
371
372 [14] Pinou T, Domènech F, Lazo-Wasem EA, Majewska R, Pfaller JB, Zardus JD, Robinson NJ.
373 Standardizing sea turtle epibiont sampling: outcomes of the Epibiont Workshop at the 37th
374 International Sea Turtle Symposium. Marine Turtle Newsletter 2019;157:22-32.
375
376 [15] Wyld JA, Brush AH. Keratin diversity in the reptilian epidermis. Journal of Experimental Zoology
377 1983;225:387-396.
378 bioRxiv preprint doi: https://doi.org/10.1101/669523; this version posted June 12, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. 16
379 [16] Crespo-Picazo JL, García-Parraga D, Domènech F, Tomás J, Aznar FJ, Ortega J, Corpa JM. Parasitic
380 outbreak of the copepod Balaenophilus manatorum in neonate loggerhead sea turtles (Caretta
381 caretta) from a head-starting program. BMC Veterinary Research. 2017;13:154.
382
383 [17] Meischner D. Seepocken auf einer Meeres-Schildkröte, ein ökologisches Idyll. Nature und
384 Museum 2001;131: 1–7.
385
386 [18] Lazo-Wasem EA, Pinou T, Peña de Niz A, Feuerstein A. Epibionts associated with the nesting
387 marine turtles Lepidochelys olivacea and Chelonia mydas in Jalisco, Mexico: a review and field guide.
388 Bulletin of the Peabody Museum of Natural History 2011;52:221-240.
389
390 [19] Caine EA. Carapace epibionts of nesting loggerhead sea turtles: Atlantic coast of U.S.A. Journal
391 of Experimental Marine Biology and Ecology 1986;95:15-26.
392
393 [20] Pfaller JB, Frick MG, Reich KJ, Williams KL, Bjorndal K. Carapace epibionts of loggerhead turtles
394 (Caretta caretta) nesting at Canaveral National Seashore, Florida. Journal of Natural History
395 2018;42:1095-1102.
396
397 [21] Fuller WJ, Broderick AC, Enever R, Thorne P, Godley BJ. Motile homes: a comparison of the
398 spatial distribution of epibiont communities on Mediterranean sea turtles. Journal of Natural History
399 2010;44:27-28.
400
401 [22] Devin ML, Sadeghi P. Barnacles on hawksbill sea turtles, Eretmochelys imbricata, in Hormoz
402 Island, Iran: (Reptilia: Cheloniidae). Zoology in the Middle East 2010;49:45-48.
403 bioRxiv preprint doi: https://doi.org/10.1101/669523; this version posted June 12, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. 17
404 [23] Razaghian H, Esfandabad BS, Hesni MA, Shoushtari RV, Toranjzar H, Miller J. Distribution
405 patterns of epibiotic barnacles on the hawksbill turtle, Eretmochelys imbricata, nesting in Iran.
406 Regional Studies in Marine Science 2019;27:100527.
407
408 [24] Majewska R, Santoro M, Bolaños F, Chaves G, De Stefano M. Diatoms and other epibionts
409 associated with olive ridley (Lepidochelys olivacea) sea turtles from the Pacific Coast of Costa Rica.
410 PloS ONE 2015;10:e0130351.
411
412 [25] Fonseca LG, Murillo GA, Guadamúz L, Spínola RM, Valverde RA. Downward but stable trend in
413 the abundance of arribada olive ridley sea turtles (Lepidochelys olivacea) at Nancite Beach, Costa
414 Rica (1971–2007). Chelonian Conservation and Biology 2009;8:19-27.
415
416 [26] Valverde RA, Orrego CM, Tordoir MT, Gómez FM, Solís DS, Hernández RA et al. Olive ridley mass
417 nesting ecology and egg harvest at Ostional Beach, Costa Rica. Chelonian Conservation and Biology
418 2012;11:1-11.
419
420 [27] {Bézy VS, Girondot M, Valverde RA. Estimation of the net nesting effort of olive ridley arribada
421 sea turtles based on nest densities at Ostional Beach, Costa Rica. Journal of Herpetology
422 2016;50:409-415.
423
424 [28] Lazo-Wasem EA, Pinou T, De Niz AP, Salgado MA, Schenker E. New records of the marine turtle
425 epibiont Balaenophilus umigamecolus (Copepoda: Harpacticoida: Balaenophilidae): new host
426 records and possible implications for marine turtle health. Bulletin of the Peabody Museum of
427 Natural History 2007;48:153-156.
428 bioRxiv preprint doi: https://doi.org/10.1101/669523; this version posted June 12, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. 18
429 [29] Júnior JCR, Pfaller JB, Corbetta R, Veríssimo L. Parasitic isopods associated with sea turtles
430 nesting in Brazil. Journal of the Marine Biological Association of the United Kingdom 2015;95:973-
431 981.
432
433 [30] Domènech F, Tomás J, Crespo-Picazo JL, García-Párraga D, Raga JA, Aznar FJ. To swim or not to
434 swim: potential transmission of Balaenophilus manatorum (Copepoda: Harpacticoida) in marine
435 turtles. PloS ONE 2017;12:e0170789.
436
437 [31] Bjorndal KA. Roles of loggerhead sea turtles in marine ecosystems. In: Loggerhead sea turtles.
438 Smithsonian Books, Washington, DC, USA. 2003; pp. 235-254.
439
440 [32] Corrêa GV, Ingels J, Valdes YV, Fonsêca-Genevois VG, Farrapeira CM, Santos GA. Diversity and
441 composition of macro-and meiofaunal carapace epibionts of the hawksbill sea turtle (Eretmochelys
442 imbricata Linnaeus, 1822) in Atlantic waters. Marine Biodiversity 2014;44:391-401.
443
444 [33] Richard JD, Hughes DA. Some observations of sea turtle nesting activity in Costa Rica. Marine
445 Biology 1972;16:297-309.
446
447 [34] Hamel MA, McMahon CR, Bradshaw CJ. Flexible inter-nesting behaviour of generalist olive ridley
448 turtles in Australia. Journal of Experimental Marine Biology and Ecology 2008;359:47-54.
449
450 [35] Pfaller JB, Alfaro-Shigueto J, Balazs GH, Ishihara T, Kopitsky K, Mangel JC et al. Hitchhikers reveal
451 cryptic host behavior: new insights from the association between Planes major and sea turtles in the
452 Pacific Ocean. Marine Biology 2014;161:2167-2178. bioRxiv preprint doi: https://doi.org/10.1101/669523; this version posted June 12, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. 19
453 TABLES
Mean # of Mean epibiont Systematic Epibiont Taxon epibionts per biomass per Group host host (g) Annelida: Hirudinea Ozobranchus branchiatus 0.1 < 0.1 Annelida: Polychaeta Polychaeta < 0.1 < 0.1 Annelida: Polychaeta Serpulidae 0.1 < 0.1 Chlorophyta Algae / Hydroid 0.3 0.1 Chelicerata: Pycnogonida Pycnogonida < 0.1 < 0.1 Crustacea: Amphipoda Caprellidea 0.1 < 0.1 Crustacea: Amphipoda Corophiidae < 0.1 < 0.1 Crustacea: Amphipoda Podocerus chelonophilus < 0.1 < 0.1 Crustacea: Brachyura Planes spp. < 0.1 < 0.1 Crustacea: Cirripedia Chelonibia testudinaria 2.8 2.3 Crustacea: Cirripedia Chelonibiidae < 0.1 < 0.1 Crustacea: Cirripedia Conchoderma virgatum < 0.1 < 0.1 Crustacea: Isopoda Corallanidae < 0.1 0.1 Crustacea: Cirripedia Balanoiodea sp. 0.5 < 0.1 Crustacea: Cirripedia Platylepadidae spp. 2.1 0.1 Crustacea: Cirripedia Platylepas hexastylos 15.7 2.4 Crustacea: Cirripedia Platylepas sp. 0.2 < 0.1 Crustacea: Cirripedia Stomatolepas elegans 20.9 8.9 Crustacea: Copepoda Balaenophilus manatorum 8.7 < 0.1 Crustacea: Copepoda Harpacticoida < 0.1 < 0.1 Mollusca: Gastropoda Unknown sp. < 0.1 < 0.1 454
455 Table 1. Mean # and biomass of the different epibiont species found on 30 olive ridley sea turtles
456 sampled from Playa Ostional, Costa Rica.
457
458 FIGURES LEGENDS
459 Figure 1. An illustration of the four different regions of the body that were sampled for epibionts. (1)
460 The head, shoulder, and fore flippers (= head), (2) The tail, cloaca, hind flippers, and inguinal cavities
461 (= tail), (3) The shell dorsum (= carapace), and (4) The shell ventrum (= plastron),.
462
463 Figure 2. Proportion of the total epibiont load found on the four body regions (head, carapace,
464 plastron, and tail) of olive ridley sea turtles. The left graph represents % abundance and the right
465 graph represents % total biomass. bioRxiv preprint doi: https://doi.org/10.1101/669523; this version posted June 12, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. 20
466
467 Figure 3. Proportion of the five main epibiont species relative to the total epibiont load found on the
468 entire body of olive ridley sea turtles. The left graph represents % abundance and the right graph
469 represents % total biomass.
470
471 Figure 4. Mean number of individuals (A) and mean biomass (B) for the five main epibiont species
472 (Balaenophilus manatorum, Chelonibia testudinaria, Platylepadidae spp., Platylepas hexastylos, and
473 Stomatolepas elegans) from the four 4 different body regions (head, carapace, plastron, and tail) of
474 olive ridley sea turtles. bioRxiv preprint doi: https://doi.org/10.1101/669523; this version posted June 12, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. bioRxiv preprint doi: https://doi.org/10.1101/669523; this version posted June 12, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. bioRxiv preprint doi: https://doi.org/10.1101/669523; this version posted June 12, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. bioRxiv preprint doi: https://doi.org/10.1101/669523; this version posted June 12, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license.