The effect of lighting and temperature on the eggs and hatchlings
of olive ridley turtles at Rushikulya, India
A Thesis Submitted to
The Manipal University
In partial fulfillment for the degree of
Master of Science in Wildlife Biology and Conservation 2008
By Divya Karnad
Post-Graduate Program in Wildlife Biology & Conservation Centre for Wildlife Studies and National Centre for Biological Sciences UAS-GKVK Campus Bangalore – 500 065 ii iii
To those who teach by example, especially my family, Arun and Sashirekha.
iv EXECUTIVE SUMMARY
The olive ridley turtle (Lepidochelys olivacea) nests both sporadically and en masse along
the Indian Coast. Of the three mass nesting sites along the East coast of India, the Rushikulya
rookery may currently have the most regular nesting population of olive ridley turtles and is
therefore likely to play a key role in maintaining the Indian Ocean population of the species.
The sporadic nesting site of Chennai is completely altered by human activity and represents a set of conditions completely different from those in Rushikulya. Olive ridley turtles in India are protected and have been studied for several years but detailed studies on factors affecting nesting and hatching have not been conducted. The present study examines the effect of temperature and lighting on egg and hatchling survival of the olive ridley turtle.
The response of the hatchlings to different lighting regimes on the beach, as well as to
specific combinations of wavelength and intensity of light was studied. Hatchlings responded
to both intense point sources of light at Rushikulya as well as glows from hidden point sources. A mixed age plantation of Casuarina equisetifolia proved to be an effective light
barrier that prevented disorientation of hatchlings. Preference for light of lower wavelength
and higher intensity was observed, although hatchlings responded differently to light in the
violet band depending on its intensity. Olive ridley hatchlings were indifferent to red light
indicating that the use of this wavelength could be recommended as a photo pollution
management measure. Incubation temperature and hatching success of selected nests were
monitored within hatcheries at both sites. Incubation temperature did not have a significant
influence on mortality in nests; however, based on these temperatures, female biased sex
ratios of hatchlings at both sites were predicted.
v Acknowledgements
My advisor Kartik Shanker allowed me the freedom to pursue this research at my own pace, put up with a fair amount of indecision and provided much needed support through this whole process. Ajith Kumar, my course director has been an unending source of inspiration, with his breadth of knowledge and enthusiasm. I would not have achieved half as much without his motivation, kindness and humour. Kavita Isvaran with her kindness and patience is someone I can go to for advice or help. She has helped mould this idea and many others before it.
My classmates have been the most important part of my two years here and it is hard to think of a better set of people to be stuck with for that long. I thank them for helping me learn much about wildlife, music, philosophy, acceptable dinner-time conversation, teamwork, forgiveness, adventure and most of all having fun! In particular I treasure the moments of sanity with Kiran; a true friend, philosopher and guide, of insanity with Umesh, Nandini,
Dharma, Dipti and Swapna, of thought-filled discussion with Kulu and general discussion with
Robin, of laughter and tears with Aathira and of comfort and friendship with Priya, Priyanka,
Nachiket and Kaavya.
I am grateful to the Centre for Wildlife Studies (CWS) and the Centre for Ecological Studies
(IISc) for funding this work, to Manipal Academy of Higher Education (MAHE) and the
National Centre for Biological Sciences (NCBS) for providing the framework for this course, as well as the Forest Department for Orissa for its co-operation and support, in particular Mr
B.K. Patnaik (PCCF Wildlife), Dr CS Kar and Mr Ajay Jena (DFO, Berhampur). A number of local conservation bodies including the Rushikulya Sea Turtle Protection Committee (RSTPC)
vi and the Students Sea Turtle Conservation Network (SSTCN) provided logistic and other support. I am thankful to the generosity of those organisations and their volunteers.
Rabindranath Sahu, Ganapathy Sahu, Somanath Rao, Simhadri and Mohendra Naik in Orissa, as well as Shreya Bhat, Madhavan, Asha, Shravan Krishnan, Karunakaran, Akhila and Arun V. in Chennai helped me with field work. A number of others including including Samjukta Sahu,
Coralie D’lima and especially Suresh Kumar helped with logistics, advice and suggestions.
Devcharan Jathanna, Rashid Raza, Suhel Quader, Nibedita Mukherjee, Geoff Hyde and
Chaitanya Krishna added valuable comments and inputs that helped shape my thesis. I am grateful to all these people for the time and effort they spared for me.
My parents, grandmothers and brother have long supported my decisions and encouraged my work. That I have come this far is due credit to them and their love.
vii CONTENTS
INTRODUCTION ...... 2
BACKGROUND ...... 4
LITERATURE CITED ...... 6
ORIENTATION OF OLIVE RIDLEY TURTLE HATCHLINGS IN RESPONSE TO LIGHT IN
RUSHIKULYA, ORISSA...... 10
ABSTRACT ...... 10
INTRODUCTION...... 11
METHODOLOGY...... 14
Study area...... 14
Experiment 1: Impact of photic regions on the beach ...... 15
Experiment 2: Testing light quality ...... 16
Model of expected orientation ...... 17
Analysis...... 18
RESULTS...... 18
Experiment 1 ...... 18
Experiment 2 ...... 19
Orientation model...... 22
DISCUSSION...... 23
LITERATURE CITED ...... 27
TEMPERATURE DEPENDANT EFFECTS ON THE EGGS AND HATCHLINGS OF OLIVE
RIDLEY TURTLES ALONG THE EAST COAST OF INDIA ...... 31
ABSTRACT ...... 31
INTRODUCTION...... 32
METHODOLOGY...... 34 Study area...... 34
Field methods...... 36
Analysis...... 37
RESULTS...... 38
DISCUSSION...... 43
LITERATURE CITED ...... 47
CONCLUSION...... 52
1
INTRODUCTION
There are seven species of sea turtles found across the world – the leatherback (Dermochelys coriacea), green (Chelonia mydas), loggerhead (Caretta caretta), hawksbill (Eretmochelys imbricata), Kemp’s ridley (Lepidochelys kempii), olive ridley (Lepidochelys olivacea) and
Australian flatback (Natator depressus) (Frazier 2002). The study of factors affecting the mortality of these species is important since they are all threatened species (IUCN 2007). Of the five that are known to inhabit the coastal waters of India, the olive ridley sea turtle is the most common. It nests along much of the Indian coast, but is most numerous along the east coast (Shanker and Choudury 2006).
Of all the species of sea turtles, only the ridley turtles nest en masse. This phenomenon is known as an arribada. These events are characterized by the sudden simultaneous mass emergence of females onto a nesting beach in order to nest (Pritchard 2007). The event is distinguished by its startling nature and rapidity with which it concludes. Mass nesting occurs at a few sites around the world including Costa Rica, Mexico and India.
There are three known olive ridley mass nesting sites in India; Gahirmatha, Devi River mouth and Rushikulya (Bustard 1976, Kar 1982, Pandav et al. 1994), with only Gahirmatha designated as a protected area. The nesting season at these three sites extends from December till April every year. Mass nesting is known to take place more than once at a given site. The arribada is generally expected to occur earlier at Gahirmatha than the other two sites
2 (Shanker et al. 2003). However in recent years, mass nesting at Gahirmatha has become
erratic, due in part to the developmental activities and fishing in the area. Mass nesting at
Devi has declined to the point where arribadas are no longer expected to occur there. The
death toll of adult females has been steadily increasing in the region (Pandav and Choudhury
1999). Rushikulya currently has the most consistent record of arribadas in the last decade along with fewer dead adults washing ashore. Hence it may contribute significantly to future populations of the olive ridley turtle in the Indian Ocean.
The female olive ridley turtle lays between 100 - 150 eggs on average, in a single clutch. The nest is a pot shaped chamber in the sand that is approximately 1.5 to 2 feet deep (Pandav et al. 1998). Females from the Indian Ocean population have been shown to shift nesting site and lay eggs more than once in a season (Tripathy and Pandav 2008, Pritchard et al. 1990 cited in Rostal 2007). Hatchlings emerge from the nest about 50 days later and move towards the sea using light as a cue (Pandav et al 1998). The probability of survival of an embryo into adulthood is very low (p < 0.001) as they face most of their mortality during the egg and hatchling stage (Crouse et al. 1987). Physical factors such as temperature, rainfall and inundation; chemical factors such as oxygen concentration within the nest (Ackerman 1980), presence of chemical pollutants; and biological factors such as disease and predation, determine the survival of eggs (Fowler 1979). Artificial light pollution and predation affect hatchling survival (Peter and Verhoeven 1994).
The present study focuses on two important physical factors; (1) effect of temperature on egg mortality, and (2) effect of lighting on hatchling mortality. The study is presented in the form of two chapters; the first chapter titled “Orientation of olive ridley turtle hatchlings in
3 response to light at Rushikulya beach, India” is formatted in accordance with the guidelines
of the journal Marine Ecology Progress Series. The second chapter titled “Temperature
dependant effects on eggs and hatchlings of Lepidochelys olivacea along the East coast of
India” is also formatted in accordance with the guidelines of Marine Ecology Progress Series.
In dealing with endangered species, conservation initiatives based on science are essential.
Research is essential not only to learn about the biology of the species, but more importantly, to inform conservation activities. The current study examines (1) the influence of temperature on egg mortality and hatchling sex ratios, and (2) the effect of light on hatchling orientation. The results form this study will form the basis for new and better informed conservation measures for the species in India.
BACKGROUND
When adult mortality (Crouse et al. 1987) does not contribute greatly to mortality in the population, the important factors affecting the populations of sea turtles are factors affecting nesting and hatchling success. Populations of leatherback turtles have shown an increase after their nesting beaches were protected (Dutton et al. 2005). The crucial factors affecting eggs and hatchlings on nesting beaches are predation, lighting and temperature. Conservation work at some sites in India has ensured that predation is controlled to some extent (Tripathy
2005, Shanker 2003), but the more intangible factors of lighting and temperature have not been addressed at all.
4 Many sea turtle conservation initiatives attempt to protect nests by relocating them to a hatching corral or an artificial hatchery. Careful relocation of sea turtle eggs to a hatchery has been shown to increase hatching success in species like the loggerhead (Wyneken et al.
1988). However, changing the location of the nest results in modifying the incubation environment of the eggs, which may have unforeseen consequences. Incubation temperature needs to be within a critical range for successful incubation of eggs (Miller 1996). Incubation temperature also plays a role in sex determination of the embryos (Mrosovsky 1972; Wibbels
2007). These temperature thresholds might vary between populations as they evolved to conditions at different nesting sites (Mc Coy et al. 1983, Wibbels et al. 1998 cited in Wibbels
2007). Further, although these thresholds have been established under the constant conditions of laboratory studies, the effects of temperature in the wild have not been studied in detail.
Sea turtle hatchlings orient towards the brighter horizon during sea-finding (Salmon and
Witherington 1995). Both intensity and wavelength are known to affect orientation
(Witherington and Bjorndal 1991). Studies have been conducted on hatchlings of a number of species at many sites including olive ridley turtles in Costa Rica (Witherington 1992).
However, the genetic distinctiveness of the Indian Ocean population (Shanker et al. 2004) may account for behavioral variations in the hatchlings that can only be discerned by studying this population. Further site – specific lighting issues can be addressed by such a study, which is important because the Rushikulya mass nesting site is a significant contributor to the Indian Ocean population.
In this study, I examine the influence of incubation temperature on hatchling mortality and predicted sex ratios. I also address the issue of lighting, examining in detail the response of
5 hatchlings to different wavelengths and intensities of light. Based on the responses of hatchlings, a map indicating the levels of disorientation hatchlings along this stretch of coast adjoining the study site has been developed. This study will inform existing conservation measures for the Indian Ocean population of the olive ridley turtle.
LITERATURE CITED
Ackerman, R. A. 1980. Physiological and ecological aspects of gas exchange by sea turtle
eggs. American Zoologist. 20(3): 575 – 583
Bustard, H. R. 1976. The world’s largest sea turtle rookery. Tigerpaper. 3 (3): 25
Crouse, D. T., L. B. Crowder and H. Caswell. 1987. A stage-based population model for
loggerhead sea turtles and implications for conservation. Ecology 68: 1412 – 1423.
Dutton, D. L., P. H. Dutton, M. Chaloupka and R. H. Boulon. 2005. Increase of a Caribbean
leatherback turtle Dermochelys coriacea nesting population linked to long-term nest
protection. Biological Conservation. 126 (2): 186 – 194.
Frazier, J. 2002. Marine Turtle of the past: A vision for the future. In: Lauwerier, R.C.G.M.
and I. Plug (eds.) The future from the past. pp 103 – 116. Oxbow Books, Durham.
IUCN. 2007. http://www.iucnredlist.org/search/details.php/11534/all. Accessed on 25th May
2008.
6
Kar, C.S. 1982. Discovery of second mass nesting ground of the pacific olive ridley sea
turtles Lepidochelys olivacea in Orissa, India. Tigerpaper 9: 6-7
Miller, J.D. 1996. Reproduction in sea turtles. In: Lutz, P. L., J. A. Musick and J. Wyneken
(eds.) The Biology of Sea Turtles: Volume 1. CRC Press, Taylor and Francis, Oxford,
U K.
Mrosovsky, N. 1972. The water finding ability of sea turtles. Behavioral studies and
physiological speculations. Brain Behavior and Evolution. 5(2): 202 – 225.
Pandav, B. and B. C. Choudhury. 1999. An update on the mortality of the olive ridley sea
turtles in Orissa, India. Marine turtle newsletter. 83: 10 – 12.
Pandav, B., B. C. Choudhury and C. S. Kar. 1994. A status survey of olive ridley sea turtle
(Lepidochelys olivacea) and their nesting beaches along the Orissa coast, India.
Unpublished Report. Wildlife Institute of India, Dehradun.
Pandav, B., B. C. Choudhury and K. Shanker. 1998. The olive ridley turtle (Lepidochelys
olivacea) in Orissa: An urgent call for an intensive and integrated conservation
programme. Current Science. 75 (1212): 1323 – 1328.
Peters, A. and Verhoeven K. J. F. 1994. Impact of artificial lighting on the seaward
orientation of hatchling loggerhead turtles. Journal of Herpetology. 28(1): 112 – 114.
7
Pritchard, P. C. H. 2007. Arribadas I have known. In: Plotkin, P. (ed.) The biology and
conservation of ridley sea turtles. pp 7 – 21. The Johns Hopkins University Press,
Baltimore, Maryland.
Rostal, D. C. 2007. Reproductive physiology of the ridley sea turtle. In: Plotkin, P. (ed.)
Biology and conservation of ridley sea turtles. The Johns Hopkins University Press,
Baltimore, Maryland.
Salmon, M. and B. E. Witherington. 1995. Artificial lighting and seafinding by loggerhead
hatchlings: Evidence for lunar modulation. Copeia. 1995(4): 931 – 938.
Shanker, K. 2003. Thirty years of sea turtle conservation on the Madras coast: a review.
Kachhapa. 8: 16 – 19.
Shanker, K. and B. C. Choudhury. 2006. Marine turtles in the Indian subcontinent: a brief
history. In: K. Shanker and B.C. Choudhury (eds.) Marine turtles of the Indian
subcontinent. pp 3-16. Universities Press, Hyderabad, India.
Shanker, K., B. Pandav and B. C. Choudhury. 2003. An assessment of the olive ridley turtle
(Lepidochelys olivacea) nesting population in Orissa, India. Biological Conservation
115: 149 – 160
8 Shanker, K., R. K. Aggarwal, J.Rama Devi, B. C. Choudhury and L. Singh. 2004.
Phylogeography of olive ridley turtles (Lepidochelys olivacea) on the east coast of
India: implications for conservation theory. Molecular Ecology. 13:1899–1909.
Tripathy B (2005) Lighting and sea turtle hatchlings in Rushikulya: Letters to the editor.
Indian Ocean Turtle Newsletter 1:26-27
Tripathy, B. and B. Pandav. 2008. Beach fidelity and interesting movements of olive ridley
turtles (Lepidochelys olivacea) at Rushikulya, India. Herpetological Conservation and
Biology. 3(1): 40 – 45
Wibbels T (2007) Sex determination and sex ratio in ridley turtles pp 167 – 189. In: Plotkin
PT (editor). Biology and Conservation of Ridley Sea Turtles. Johns Hopkins
University press, Baltimore, Maryland
Witherington B. E. and K. A. Bjorndal. 1991. Influences of wavelength and intensity on
hatchling sea turtle phototaxis: Implications for sea-finding behaviour. Copeia 1991
(4): 1060 – 1069.
Witherington, B. E. 1992. Sea-finding behavior and the use of photic orientation cues by
hatchling sea turtles. Unpublished PhD thesis, University of Florida.
Wyneken, J., T. J. Burke, M. Salmon and D.K. Pedersen. 1988. Egg failure in natural and
relocated sea turtle nests. Journal of Herpetology. 22(1): 88 – 96.
9 ORIENTATION OF OLIVE RIDLEY TURTLE HATCHLINGS IN
RESPONSE TO LIGHT IN RUSHIKULYA, ORISSA
ABSTRACT
Sea finding behavior in sea turtle hatchlings is mediated by visual cues. Hatchlings move
towards the brighter horizon and also show wavelength preferences. Previous studies on
olive ridley (Lepidochelys olivacea) hatchlings indicate that they prefer low wavelength and
high intensity light. These preferences were confirmed in the Indian Ocean population by the
present study. Hatchlings at the study site responded both to visible point sources of light and
to mere glows of light. The introduced plantation tree Casuarina equisetifolia acted as an
effective light barrier. The potential impacts of its use as a light barrier are discussed. A map
of the region showing the likely disorientation of hatchlings in different photic regions is also
presented. The implications of this study in developing ‘turtle-friendly’ coasts are also discussed.
10
INTRODUCTION
Sea turtle hatchlings are known to be positively phototropotactic and hence orient towards the brighter of two horizons (Mrosovsky & Kingsmill 1985). As most hatchling emergence is at night, positively phototropotactic behaviour plays a critical role in sea-finding (Mrosovsky
1972, Verheijen 1985, Witherington 1990). The naturally brighter horizon is seaward, but in recent decades, increasing beach front lighting has created an artificial light horizon on the landward side along many parts of the coast. Subsequently, hatchlings have been observed to orient wrongly away from the sea, resulting in considerable mortality (McFarlane 1963).
Increasing coastal development is likely to amplify hatchling mortality due to light pollution and therefore, finding suitable mitigation measures is vital.
Hatchlings use multiple visual cues during sea-finding, of which intensity, wavelength, background illumination and landward silhouettes are important factors (Witherington &
Bjorndal 1991, Witherington 1992, Godfrey and Barretto 1995, Salmon & Witherington
1995, Tuxbury & Salmon 2005). While background illumination and landward silhouettes reduce the relative effects of coastal artificial light, light of low wavelength and high intensity tend to increase a hatchling’s probability of disorientation (Lohmann et al. 1997).
However response of hatchlings to intensity and wavelength of light show species specific differences (Witherington & Bjorndal 1991). Variation in response at the population level may also be possible.
11 No prior research has quantified the wavelength and intensity preferences of the Indian
Ocean population of olive ridley (Lepidochelys olivacea) hatchlings. The genetic distinctiveness of this population (Shanker et al. 2004) could account for differences in behavior, besides the obvious conservation implications of such research. In this study, I
examined the response of hatchlings from the Indian Ocean population of L. olivacea to light
quality, in order to address the issue of beach front lighting due to coastal development and
industrialization at an important nesting site on the east coast of India.
On the mainland coast of India, L. olivacea nests in significant numbers along the east coast of India, although sporadic nesting also occurs along the west coast (Kar & Bhaskar 1982,
Shanker & Choudhury 2006). The three main mass nesting sites along the east coast are
Gahirmatha, Devi River Mouth and Rushikulya (Bustard 1976, Kar 1982, Pandav et al.
1994). Although the site at Gahirmatha is historically best known and has received formal
protection, mass nesting events have been erratic at this site in the past decade (Shanker et al.
2003). This is believed to be due to the fragmentation of the nesting beaches due to cyclones
in 1988 and 1999, and due to an increase in mechanized fishing along the Orissa coast that
has resulted in the stranding of over 100,000 dead turtles in the last decade (Shanker et al.
2003). Growing industrialization in surrounding areas is also a matter of considerable
concern at this site. Nesting at the Devi river-mouth has declined over the years, with only
the Rushikulya river-mouth showing consistent mass nesting, and perhaps an increase in
numbers (Pandav 2000, Shanker et al. 2003, Tripathy 2005a, D.K & K.S pers obs.). Relative
to Devi and Gahirmatha (several thousand per year), fewer dead turtles are washed up along
this coast. The relatively low impact of mechanized fishing here makes this a critical mass
nesting site for olive ridley turtles at this point. However, this makes the issue of artificial
12 lighting critical in determining the survival of this population of L. olivacea over the long
term. I address the issue of existing beach front lighting at this site, as well as potential mitigation measures in the form of biological light barriers that occur along some stretches in
Rushikulya.
To understand how different sources of light pollution will influence hatchling orientation
and to test the efficacy of light barriers, I need to know how hatchlings react to important
attributes of light – wavelength and intensity. Light of high intensity is known to attract
hatchlings of all species and L. olivacea hatchlings have shown preference for light in the
violet to orange regions of the spectrum, during trials elsewhere (Witherington 1992).
Besides this study of light at the finer scale, I also addressed the issue of hatchling orientation
as influenced by major sources of light pollution and potential barriers to this pollution at the
mass nesting site. These field conditions form a good representative of the light pollution sources and potentially mitigating barriers that this species encounters along the east coast of
India. Finally, I use our findings concerning hatchling response to experimental and coastal light conditions together with spatial information of previous nesting records to develop a
model of the expected impact of the light regime at Rushikulya on hatchling orientation and
mortality.
13
METHODOLOGY
Study area
The study site is a 5 km stretch of beach encompassing the Rushikulya mass nesting site in
Orissa, India (19º 22’ 56.30 N, 85º 5’ 0.23 E to 19º 24’ 35.74 N, 85º 05’ 59.48 E. This is the southernmost mass nesting beach in India. Siltation and flooding of the Rushikulya River causes the beach profile to vary both seasonally and annually. The three villages
(Purunabandha, Gokharkuda and Kantiagada) (Map), along the beach have a human
population of about 9000 of which around 3000 persons are involved in fishing activity
(unpublished data).
Plantations of Casuarina equisetifolia flank the northern portion of the study area. These
plantations support predators of L. olivacea including the golden jackal (Canis aureus),
striped hyena (Hyaena hyaena), jungle cat (Felis chaus), jungle crow (Corvus splendens) and
white-bellied sea eagle (Haliaeetus leucogaster). Large sand dunes and disused prawn farms run parallel to the coast close to the river mouth. The main sources of light pollution are the three villages along the beach, the Chennai - Kolkata National Highway (NH5), which cuts across the landscape from south to north and the chemical factory at the southern end of the beach.
14
Figure 1: Map showing (a) olive ridley mass nesting sites: Gahirmatha, Devi and Rushikulya (b) Study area at Rushikulya
Experiment 1: Impact of photic regions on the beach
By visual classification, the beach was divided into four different photic regions based on the light shading effect of the plantation tree Casuarina equisetifolia as well as intense lighting from surrounding sources of light pollution.
In the photic region one (PR 1) C. equisetifolia plantations were 50 m from the high tide line
(HTL), which effectively cut out illumination from the surroundings. The second photic
15 region (PR 2) C. equisetifolia plantations were about 500 m away from the HTL, so the
region experienced a glow from the surrounding areas. Photic region three (PR 3) had no C.
equisetifolia barrier and was exposed to well spaced point sources of light from the highway.
The fourth photic region (PR 4) also had no light barrier and experienced light of high
intensity from the highway as well as the nearby chemical industry.
The movement of hatchlings seawards and landwards was experimentally evaluated at the
four different photic regions using circular arenas of 1.5 m radius, well above the high tide
line, which were divided into eight sectors. Newly emerged hatchlings were tested for
activity and then placed in complete darkness before being released ten at a time into the
centre of the arena. The hatchlings were placed in such a way as to be able to view both the
light source and the sea. The hatchlings were allowed to orient and move to the periphery of
the arena from where they were collected, re-oriented and released into the sea. The number
of hatchlings in each of eight sectors was counted at the end of every trial. Altogether nine
trials were conducted at each of the sites. All trials were conducted during the waning
gibbous moon.
Experiment 2: Testing light quality
A circular arena was setup as before, but with light barriers along the periphery to prevent
external illumination entering the experimental chamber. The light was shone vertically
downward from a height of 30 cm through a hole created in the light barrier. Lights of two intensities (eight and four point sources) were chosen in four bands of wavelength (Red: 580
– 800 nm, Yellow: 475 – 600 nm, Blue: 375 – 575 nm, Violet: 300 – 450 nm). Ten active
16 hatchlings from a nest were randomly chosen, and introduced at the centre of the arena. They were allowed to orient to the particular wavelength and intensity of light chosen for the trial, and move to the walls of the arena, which denoted the end of the trial. The hatchlings were re-oriented and released at the end of each trial. Hatchlings from another nest were selected for the next trial. Nine such trials were conducted for each combination of wavelength and intensity.
Based on the results obtained from the above experiments, choice experiments involving a T tube setup where the hatchlings had to choose between certain pairs of wavelengths and intensities to resolve the differences in hatchling orientation were conducted.
Model of expected orientation
The beach was classified based on the exposure to light in four different photic categories (as above). A model of the orientation profile of hatchlings was developed based on the probability of their disorientation conditional on their emergence from different photic regions of the beach (Table3).
Conditional probability = p (disorientation in a given PR)
p (disorientation in all PR)
The model was extended to include those parts of the beach known to have harbored mass nesting events in the past and those that were likely to do so in the future. Based on this model a map of the site showing photic regions and probability of disorientation was
17 developed. This model was used to predict hatchling response only for that period of the lunar cycle, around the full moon.
Analysis
Analysis was conducted using the software R 2.7.0 (R Development Core team 2008), SPSS v
10 (SPSS Inc, Chicago USA) and ORIANA v 2 . In the experiments evaluating field
conditions, the response of hatchlings, measured as the number of hatchlings moving into
each of the eight sectors of the arena, was taken as the response to the predictor of photic
regions using a generalized linear model with binomial errors. The evaluation of the
experiments with wavelength and intensity was conducted similarly where the response of
hatchlings was compared to the predictors of wavelength and intensity. The interaction terms
between the two predictors were evaluated against purely additive effects. Hatchling
response was also tested for significance of directionality by finding the mean vector and
performing Rayleigh’s test of significance of directionality.
RESULTS
Experiment 1
Both concentrated point sources of light and visible glows caused greater disorientation of
hatchlings than spaced sources or parts of the beach flanked by C. equisetifolia 50 m away
(Fig1: Generalized linear model; Deviance = 108.6, p <0.001).
18 The greatest deviance from the expected seaward orientation (0º or 360º) was seen in photic region four. Rayleigh’s test for circular uniformity indicates that the hatchlings showed significant seaward orientation only in Photic region one (Table 1).
Table 1: Significance of directionality in seaward orientation across photic regions. PR – Photic region
Photic Mean Mean vector Circular Rayleigh’s region Vector ( ) length (r) SD (p) PR 1 326.19° 0.68 46.97° 0.01 PR 2 337.5° 0.19 104.49° 0.71 PR 3 355.94° 0.42 75.06° 0.17 PR 4 112.5° 0.19 104.49° 0.71
Figure 2: Landward movement of hatchlings due to disorientation in different photic zones along Rushikulya beach. cas 50 represents PR 1, cas.500 is PR 2, sp. point is PR 3 and cl.point is PR 4
Experiment 2
Wavelength and light intensity interacted to influence hatchling orientation (Generalized
linear model; Deviance = 84.603, df = 69 p = 0.00). Hatchlings responded to high intensity
19 more than to low intensity for all wavelengths except violet (Fig 2). Although the response to
low intensity light was greater in arena trials, hatchlings consistently chose higher intensity
light in choice based trials (χ2 = 4.5988, df = 1, p = 0.03). The response of hatchlings to red
light was very similar to their movements in complete darkness, in arena trials (Fig 3). They
also preferred lower wavelengths (violet and blue) in the choice based trials than higher ones
(χ 2 = 15.2191, df = 3, p < 0.001).
Figure 3: Interaction plot of hatchling response to wavelength and intensity
Directional responses of hatchlings to light indicated greatest dispersion of hatchlings in
response to the red band of wavelength (580 – 800 nm) followed by that of the yellow band
(475 – 600 nm) (Table 2). Significant orientation towards the source was observed at all bands of wavelength except red (p < 0.01).
20
Table 2: Directionality of hatchlings in response to bands of wavelength
Colour Mean Mean Circular Rayleigh’s (wavelength vector Vector SD (p) band) length (r)
Violet 4.16° 0.99 2.86° <1 e-12 Blue 360° 0.96 15.05° <1 e-12 Yellow 11.87° 0.85 32.47° 9.97 e-5 Red 345.36° 0.57 60.35° 0.03 Dark 112.5° 0.079 129.24° 0.94
Figure 4: Hatchling response to wavelength (vio - violet, blu - blue, yel - yellow and dar – dark).
21
Orientation model
The probabilities of hatchling movement towards land were calculated, based on response of
hatchlings in different photic regions location at the beach (Table 3). The map (Fig 4)
developed based on these probabilities indicated that hatchlings along the northern portion of the beach were less likely to be disoriented due to artificial lighting than those on the southern part of the beach. As mass nesting took place along these parts of the beach during the years 2004 and 2008, only about 1.13% of hatchlings were likely to be disoriented, if hatching took place around full moon. Mass nesting on the southern end of the beach, which occurred in the years 2005 and 2006, would result in close to 50% of the hatchlings being disoriented during full moon.
Table 3: Conditional probabilities of hatchling movement in different photic regions of the beach
Photic regions Conditional probability of disorientation PR4 0.54
PR 3 0.193
PR 2 0.393
PR 1 0.013
22
Figure 5: Map indicating regions of the coast where hatchlings showed different levels of disorientation due to artificial lighting. Locations and years of mass nesting have also been marked.
DISCUSSION
A number of studies have reported the impacts of different aspects of light on hatchling orientation (Mrosovsky 1972, Verheijen 1985, Witherington 1992). Findings from the present study confirm that both wavelength and intensity of light are important orientation cues for hatchlings from the Indian Ocean population of L. olivacea. In this paper I argue that wavelength and intensity have an interactive effect on hatchling orientation. Our findings also suggest that light barriers form an effective means of preventing hatchling disorientation on turtle nesting beaches. With an orientation model, I identified the areas along the study site that need to be targeted for light spillover prevention measures.
23 Artificial lighting was reported to have a high impact (80% disorientation) in the study site
(Tripathy et al. 2003). As a result, local conservation organizations resorted to a number of measures to rescue the disoriented hatchling, ranging from premature digging up of hatchlings to setting up mosquito net barriers to trap disoriented hatchlings (Tripathy 2005b).
The collection and release of disoriented hatchlings is a labour intensive process that often results in the loss of hatchlings to predators and dehydration, due to the inability of the volunteers to cope with the numbers of hatchlings needing rescue (D.K pers obs). Our study reports a lower probability of disorientation, as most of the nesting occurred on parts of the beach shaded by an existing light barrier, plantations of C. equisetifolia.
Hatchling response to the existing lighting regime at the study site was probably confounded by moonlight as the study period coincided with the peak phase of the lunar cycle. Sea turtle hatchlings are known to respond to the background illumination of the moon (Salmon &
Witherington 1995). I believe that during the study, illumination from the moon reduced the intensity gradient caused by artificial light sources, allowing many hatchlings from photic regions 2 and 4 to show a seaward orientation. Still a large number of hatchlings were disoriented and trapped in the net barriers, although they were a small proportion of the total number of hatchlings on the beach. The best solution to this is to set up light barriers along the entire stretch of the beach. Artificial light barriers over such large stretches are neither practical nor economically feasible. Hence plantations of coastal vegetation such as C. equisetifolia could serve as suitable ‘natural’ light barriers to prevent artificial illumination on the beach.
24 C. equisetifolia is a nitrogen fixing plantation crop that has been introduced to the coasts of
India both as a shelter belt and for the use of its wood (Pinyopusarerk & Williams 2000). The
plantation in our study was a mixed age plantation which supported predators of L. olivacea
eggs and hatchlings. Still, it formed a more effective light barrier than the single age stands that are planted along much of the East coast of India. However, being an introduced species, C equisetifolia’s impact on the environment is yet to be completely studied. A study has suggested that changing sand erosion patterns and slope of the beach coupled with
decline in nesting in those areas could be linked to plantations of C. equisetifolia (Choudhari, thesis).
C equisetifolia cannot be recommended as a conservation tool over large areas of the coast before further research quantifying its impacts are carried out. Instead coastal lighting could be modified to be more ‘turtle-friendly’. Witherington (1992) showed that hatchlings of four species of turtle including L. olivacea from Costa Rica showed a preference for light of low wavelength. They were indifferent to light in the red region (630 – 700 nm) of the spectrum.
This preference is theoretically associated with sea-finding under low light conditions, such as on moonless nights. Therefore, one expects that as intensity of light increases, wavelength ceases to be a discriminating factor. This was shown by preferential orientation of hatchlings to light at high intensity even if they do not prefer the wavelength at lower intensities
(Witherington & Bjorndal 1991). I find that the hatchlings from the Indian Ocean population of L. olivacea show the same patterns of light quality preference. They were found to react more to wavelength at lower intensities, indicating that although wavelength might be the primary cue, high intensity light could out-compete all other visual orientation cues.
Hatchlings in the light proof arena showed a peculiar response to high intensity light,
25 choosing to circle rather than orient uni-directionally. This indicates that light of high
intensity could have very complex effects on orientation that require more detailed studies to
resolve.
Hatchling orientation across the study site varied depending on the type of light illuminating
parts of the study area. Hatchlings reacted significantly to both the glow from hidden sources and to the compounded effect of point sources and their glow. This suggests that hatchlings make decisions comparing the gradient between the two horizons, rather than move towards the brightest point in their visual range. The threat of increasing development in terms of oil pipelines, ports or industries at Rushikulya and other olive ridley turtle nesting sites only
increases the probability of disorientation of hatchlings at those sites. Finding solutions to the
conflict over coastal resources is imperative and might involve a combination of the
measures examined in this paper.
Managing the photic environment at mass nesting sites of L. olivacea could prove to be
critical to the continued survival of the species. Where light cannot be shaded out high
wavelength and low intensity light should be chosen since they are least likely to disrupt the
natural orientation mechanism of hatchlings. However the best technique to prevent disorientation of hatchlings in response to artificial beach front lighting is to screen the light from view. Since artificial light barriers can be used only over short stretches, natural light barriers such as sand dunes or trees are preferred. However, the species used as biological light barriers must be chosen with care in order to avoid unwanted consequences on the very species they attempt to conserve.
26 LITERATURE CITED
Bustard HR (1976) World’s largest sea turtle rookery? Tigerpaper 3(3): 25
Chaudhari S (2008) Impact of casuarina (Casuarina equisetifolia) plantation on nesting of
olive ridley (Lepidochelys olivacea) sea turtle along the Chennai – Pondicherry coast.
Unpublished Master’s thesis, Pondicherry University.
Godfrey MH, Barreto R (1995) Beach vegetation and seafinding orientation of turtle
hatchlings. Biol Conserv. 74: 29-32.
IUCN (2007) http://www.iucnredlist.org/search/details.php/11534/all. Accessed on 25th May
2008.
Kar CS (1982) Discovery of second mass nesting ground of the pacific olive ridley sea turtles
Lepidochelys olivacea in Orissa, India. Tigerpaper 9: 6-7
Kar CS, Bhaskar S (1982) The status of sea turtles in the Eastern Indian Ocean. In: Bjorndal
K (Ed) The Biology and Conservation of Sea Turtles. Smithsonian Institution Press,
Washington, D.C., 365-372.
Lohmann K.J, Witherington BE, Lohmann CMF, Salmon M (1997). In: Lutz PL, Musick JA
(ed) The biology of sea turtles. CRC Press, Boca Raton: 107-135.
27 McFarlane RW (1963) Disorientation of loggerhead hatchlings by artificial road lighting.
Copeia 1963:153
Mrosovsky N (1972) The water finding ability of sea turtles. Brain Behav and Evol. 5:202 –
225
Mrosovsky N and Kingsmill SF (1985) How turtles find the sea. Z Tierpsychol. 67:237-256
Pandav B (2000) Conservation and management of olive ridley turtles along the Orissa coast.
PhD Dissertation, Utkal University, Orissa, India.
Pandav B, Choudhury BC, Kar CS (1994) A status survey of olive ridley sea turtle
(Lepidochelys olivacea) and their nesting beaches along the Orissa coast, India.
Unpublished Report. Wildlife Institute of India, Dehradun.
R Development Core Team (2008) R: A language and environment for statistical computing.
R Foundation for Statistical Computing, Vienna, Austria.
Salmon M, Witherington BE (1995). Artificial lighting and seafinding by loggerhead
hatchlings: evidence for lunar modulation. Copeia 4: 931-938.
Shanker K, Pandav B, Choudhury BC (2003) An assessment of the olive ridley turtle
(Lepidochelys olivacea) nesting population in Orissa, India. Biol Conserv. 115: 149 –
160
28
Shanker K, Choudhury BC (2006) Marine turtles in the Indian subcontinent: a brief history.
In: Shanker & Choudhury (eds) Marine turtles of the Indian subcontinent.
Universities Press, Hyderabad, India.
Tuxbury and Salmon (2005) Competitive interactions between artificial lighting and natural
cues during seafinding by hatchling marine turtles. Biol Conserv 121:311–316
Tripathy B (2005a) A study on the ecology and conservation of olive ridley sea turtle
(Lepidochelys olivacea) at the Rushikulya rookery of Orissa coast, India. PhD
dissertation, Andhra University.
Tripathy B (2005b) Lighting and sea turtle hatchlings in Rushikulya: Letters to the editor.
Indian Ocean Turtle Newsletter 1:26-27
Tripathy B, Pandav B, Panigrahy RC (2003) Hatching success and orientation in
Lepidochelys olivacea (Eschscholtz, 1829) at Rushikulya rookery, Orissa, India.
Hamadryad 27: 213-20.
Verheijen FJ (1985) Photopollution: Artificial light optic spatial control systems fail to cope
with incidents, causations, remedies. Exp Biol. 44:1-18
Witherington BE (1990) Photopollution on sea turtle nesting beaches: problems and next best
solutions. In: Richardson TH, Richardson JI, Donnelly M (ed) Proceedings of the
29 Tenth Annual Workshop on Sea Turtle Biology and Conservation. NOAA Technical
Memorandum NMFS-SEFC-278
Witherington BE (1992) Sea-finding behavior and the use of photic orientation cues by
hatchling sea turtles. Unpublished PhD thesis, Univ Florida
Witherington BE and Bjorndal KA (1991) Influences of wavelength and intensity on
hatchling sea turtle phototaxis: implications for sea – finding behavior. Copeia
1991:1060-1069
30 TEMPERATURE DEPENDANT EFFECTS ON THE EGGS AND HATCHLINGS OF
OLIVE RIDLEY TURTLES ALONG THE EAST COAST OF INDIA
ABSTRACT
The impacts of global climate change on species biology have not been examined in the
Indian context. An ideal subject for such a study is the olive ridley sea turtle (Lepidochelys
olivacea) since it exhibits temperature dependant sexual differentiation and the eggs have an optimal range of incubation temperatures. Although data from previous years showed increasing mortality as the incubation period of eggs extended into summer, a relationship between temperature and mortality could not be discerned at both the study sites. However data from one of the sites suggests that a larger sample size might reveal a trend of increasing mortality with temperature. Incubation temperatures at both sites were close to the physiological thresholds indicating that future global warming could reduce hatchling production and have negative impacts on the population. Long term decline could also result from poor production of males, due to the predicted female bias in the sex ratio of hatchlings at both sites. Predicted sex ratios were based on mean and variance of incubation temperature, using a laboratory estimated pivotal temperature from a previous study.
31
INTRODUCTION
Evidence of global climate change exists as increasing temperatures of land, ocean surface and atmosphere (http://www.metoffice.gov.uk/research/hadleycentre.html, 2008). Global
climate change has affected species biology by forcing shifts in species ranges, flowering
phenology and changing the species composition of communities (Parmesan & Yohe 2003).
However the impacts on migratory species are hard to discern, as these species exhibit
frequent shifts in breeding site and timing (Walther et al. 2002). Migratory species like sea
turtles traverse large stretches of the ocean between feeding and breeding sites. Being cold
blooded and oviparous, their biology is highly influenced by environmental temperature.
Climate change could have a significant impact on those species of sea turtles that have very
limited ranges, low physiological tolerances or shallow nests. One of the more shallow
nesting, tropically distributed species is the olive ridley turtle (Lepidochelys olivacea), which
nests in the tropics including the Indian coast. This makes it well suited to studying the effect
of climate change on its biology.
The biology of L. olivacea could be affected by climate change in two ways. First, as sex in
sea turtles is determined by temperature, sex ratios of L. olivacea hatchlings could be
affected by changes in temperature regimes. Sea turtles show a MF pattern of sex
determination, where males are produced at low temperatures and females are produced at
higher temperatures, above a threshold called the ‘pivotal temperature’, where the sex ratio is
1:1 (Wibbels 2007). Skewing of sex ratio towards the production of females has been
demonstrated in a number of species including green turtles (Chelonia mydas), loggerhead
32 turtles (Caretta caretta) and leatherback turtles (Dermochelys coriacea) (Mrosovsky 1980,
Wibbels et al. 1991, Georges et al. 1994, Booth et al. 2004, Hulin et al. 2008). Although
some populations of L. olivacea show a high pivotal temperature, inter-population differences in these thresholds might have serious long term impacts on the adult sex ratio
(Wibbels et al. 1998). Second, egg and hatchling mortality is also temperature dependant.
Sea turtle eggs develop successfully within the range of 24 ºC - 35ºC (Matsuzawa et al. 2002,
Wibbels 2007). The critical thermal maximum for L. olivacea hatchlings in a laboratory study was 41.3 ºC while exposure to extended durations of temperatures of 36 - 38 ºC also caused mortality (Drake & Spotila 2001). Sustained exposure throughout the incubation period to temperatures beyond the critical incubation range leads to mortality of sea turtle eggs and hatchlings (Yntema & Mrosovsky 1980).
The relationship between temperature and egg or hatchling mortality has been inadequately studied in sea turtles (Matsuzawa 2002). There is a critical need for such studies since L. olivacea is categorized as Threatened (IUCN 2007) and the Indian Ocean population is evolutionarily important (Shanker et al. 2004). Some nesting sites of L. olivacea along the
Indian coast such as Chennai have shown a warming trend of about 0.04 ºC per year since
1951 (Quadir et al. 2004). There has also been an increase in mortality of eggs and hatchlings as the season during which incubation occurs progresses into summer (D.K, pers obs).
Furthermore, studies in other sites have reported the possible role of temperature in declining hatching success (Matsuzawa 2002).
An increase in ambient air temperature could translate into an increased incubation temperature. Under the scenario of global warming predicted by the regional climate models
33 (Kumar 2004) temperature along the east coast of India could rise to a level where nests of L.
olivacea experience temperatures approaching 35 ºC for sustained periods. As the east coast
of India harbours three important mass nesting sites of this species (Gahirmatha, Devi and
Rushikulya), such an increase in temperature could lead to a decline in the population.
Hence in this study I examine whether temperature is an important determinant of egg and
hatchling mortality of L. olivacea on the east coast of India, in the light of mortality patterns
from previous years. I also examine the relationship between various aspects of temperature
(mean, variance and maximum) and mortality of eggs and hatchlings at Rushikulya (a mass nesting site) and Chennai (a sporadic nesting site). Comparing predicted sex ratios across the two sites, I address the issue of sex ratio bias in the context of changing climatic profiles in
the region.
METHODOLOGY
Study area
The study was conducted at two sites on the east coast of India– the mass nesting beach at
Rushikulya (19º22’56”.30 N, 85º5’0”.23 E - 19º24’35”.74 N, 85º05’59”.48 E) and a sporadic nesting beach at Chennai (13º00’32.55”, 80º16’31.20”) (Figure 1).
The Rushikulya mass nesting site is at the mouth of the Rushikulya River in the State of
Orissa. Siltation from and flooding of the river, causes the beach profile to vary both seasonally and annually. I chose a 4 km stretch of beach, north of the river mouth, as our study site. Along the study area, plantations of Casuarina equisetifolia run parallel to the
34 coast at an average distance of 60 m from the High Tide Line (HTL). These plantations support predators of L. olivacea including the golden jackal (Canis aureus), striped hyena
(Hyaena hyaena), jungle cat (Felis chaus), jungle crow (Corvus splendens), black kite
(Milvus migrans) and white-bellied sea eagle (Haliaeetus leucogaster).
The sporadic nesting beach is located in Chennai and is highly disturbed by human activity.
The beach is flanked by beachfront houses and roads. The predators of turtle eggs and hatchlings are feral dogs (Canis familiaris), jungle crows, black kites and occasional human poachers.
Figure 1: Map showing position of study sites along the East coast of India
35 Field methods
At both sites, from February to May 2008, beach transects were used to locate nests or
nesting turtles and the nests were relocated to the hatching corrals (hatcheries) within 4 hours
of the eggs being laid. Clutch size and abnormalities of eggs were noted before relocation. In
the hatcheries, the spacing between the nests was approximately 0.5 – 1.0 m. Nest
temperature was monitored in eight nests at each site using digital temperature data loggers
(HOBO Pendant), placed at the bottom of the nest, at a depth of about 45 cm, the average
nest depth for L. olivacea (Lopez-Castro et al. 2004).
The nests were monitored for signs of hatchling emergence. Upon emergence, the number of
dead hatchlings and unhatched eggs were noted and percentage of mortality was calculated.
Data loggers were removed only after all the live hatchlings had emerged from the nest.
Sand temperatures were recorded in two ways; (1) By placing 8 data loggers at each site, at
depths of 45 cm, (2) By using a surface probe at depths of 20 – 25 cm. In Rushikulya, both
measurements were taken, whereas in Chennai only data loggers were used. The surface
probe data were recorded every 10 m between the HTL and the tree line, in 100 m segments
across the study area. Data logger temperatures were recorded every two hours for the
duration of the incubation period of each nest, while surface probe temperatures were
measured every alternate day for one and a half months.
36 Analysis
The daily, weekly and seasonal mean, variance and range of temperature were calculated for
each nest. Percentage of mortality was calculated as;
Percentage of mortality = No of dead eggs and hatchlings from the nest ҳ 100
Clutch size of the nest
Analysis was conducted using the R2.7.0 software platform (R Development Core Team
2008). The data from previous years were collated and compared. Linear regressions were
run to explore patterns and then a generalized linear model with binomial errors and a logit
link was performed using site, variance and maximum recorded temperature as predictors
and hatching success as the response variable. Two and three way interactions between the
predictors were included in the analysis. The complex predictor, site, was chosen to represent
all unaccounted variations between the two sites, including spacing of nests, sand grain size,
amount of rainfall received etc. The important predictors were mean, variance, site and
maximum temperature. However as the mean was correlated with site, it was removed from
the analysis. Hatching success data from previous years were collated to examine temporal
patterns. Residuals were checked for deviations from model assumptions. Nest temperatures were also compared to air temperatures obtained from meteorological sources.
The mean and variance of the middle third of incubation period, the established sex - determining period (Standora & Spotila 1985), were used to predict sex ratio. Sex ratio was assigned based on a mean pivotal temperature of 29ºC and thresholds of approximately 28ºC
37 and 30ºC reported from laboratory studies using hatchlings from the Gahirmatha mass nesting site (Dimond and Mohanty-Hejmadi 1983, Mohanty-Hejmadi and Dimond 1986).
Daily, weekly and seasonal means of temperature obtained using the data logger and sand probe were compared with nest temperatures using Pearson’s parametric correlations in order to determine the relationship between the two.
RESULTS
Mortality across years (2004 – 2007) showed an increasing trend with fortnight in which the nest hatched, taken as an index of temperature (Figure 2). In the present study nest temperatures at Chennai and Rushikulya did not rise above 35ºC for extended periods (Figure
4). The nests were exposed to temperatures of above 35ºC for an average of 6 (±0.09) days at
Chennai and 1.2 (± 0.01) days at Rushikulya, towards the end of their incubation periods.
The highest nest temperature experienced at Chennai was 38.9ºC for a period of 4 hours, while that at Rushikulya was 37.2ºC for a period of 2 hours throughout the incubation period.
Seasonal changes of nest temperatures at both sites reflected ambient air temperatures, although the range in temperatures experienced within the nest was lower than that of air temperatures (Figure 3a, 3b). Sand probe surface temperatures at Rushikulya also followed nest temperatures, but only over the interval of weeks (Pearson’s R = -0.7, n = 6, p = 0.01).
38 100 2005
80 2007
60 2006
40 2004 Perc ent m ortality
20
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 Fortnights (time)
Figure 2: Relationship of egg and hatchling mortality to fortnight of hatching from the beginning of March in all years. n (2007) = 65, n (2006) = 51, n (2005) = 65, n (2004) = 56. Error bars are showing standard error.
41 39 37 35 33 31 29 Average Temperature,ºC 27 25 0 2 4 6 8 10 12 Time (Weeks)
Figure 3 a: Monthly mean ambient temperatures (triangles) compared to weekly mean nest temperatures (line) with variance bars for nests across incubation period for the study site in Chennai in 2008. Error bars showing variance.
39 35 34 33 32 31 30 29 28 Tem ºC perature, 27 26 25 0 2 4 6 8 10 12 14 Time (weeks)
Figure 3 b: Monthly mean ambient temperatures (triangle) compared to weekly mean nest temperatures (diamonds) with variance bars for nests across incubation period for the study site in Rushikulya in 2008. Error bars showing variance.
37
35
33
31
29 Temperature,ºC
27
25 02468101214 Time (weeks)
Figure 4: Comparison of weekly average nest temperatures in Chennai (triangles) and Rushikulya (squares) throughout the incubation period across nests. Error bars show standard error.
Nest temperatures were significantly higher in Chennai than at Rushikulya (W = 64, n = 16,
p<0.001). But hatchling mortalities showed no significant difference between the two sites
(Figure 5; W = 45, n = 16, p = 0.19). In Chennai, egg and hatchling mortality showed an
increase with variance (Figure 6 b) over the incubation period (R2 = 0.47, p = 0.02), but no
relationship could be discerned with daily or weekly variances. At Rushikulya, mortality
40 increased with overall mean temperatures (R2 = 0.51, p = 0.01), although temperatures never
rose above the threshold level of 35ºC (Figure 6a). The generalized linear model was used to
resolve these differences. The interactions between all three predictors best explained the
data (Generalized linear model: 3 way interaction Deviance = 128.33, df = -3, p < 0.0001)
indicating that site specific differences governed the patterns in the data. Therefore, when
mean and variance are low, but maximum temperature is high, the embryo is exposed to high
temperatures for long periods without the buffering effect of low temperatures, and hence is
more likely to face mortality.
Figure 6: Comparison of overall egg and hatchling mortality (percentage) across the two sites of Chennai and Rushikulya, indicating the spread of data around the median
41 120
100 100
80 80
60 60
40 40
20 20
0 0 25 27 29 31 33 35 02468101214
M ean Nest Temperature Variance in Nest Temperature
Figure 7: Relationship of egg and hatchling mortality at both Chennai (triangles) and Rushikulya (squares); with (a) overall mean showing line fit at Rushikulya and (b) overall variance across the study period showing line fit in Chennai
The pivotal (29ºC) and threshold range of temperatures (28ºC and 30ºC) reported by Dimond
& Mohanty-Hejmadi (1983) were used to develop a sex ratio profile for each of the sites
(Table 1). The minimum temperatures for 6 nests and means for all nests were above both the pivotal and upper threshold temperatures for male production in Chennai. Six out of eight nests showed a mean temperature greater than the predicted pivotal temperature.
42 Table 1: Prediction of sex ratio in % female, in the two sites based on mean, variance and maximum recorded temperature from the middle third of incubation period
Temperature Predicted Incubation Site Mean Min Max sex ratio period Variance (% F) Rushikulya 14 Feb - 11 Apr 29.31 0.65 28.33 30.53 50 14 Feb - 08 Apr 29.79 0.83 28.7 31.24 50 24 Feb - 17 Apr 29.93 0.56 28.65 31.43 50 25 Feb - 19 Apr 30.51 0.8 28.81 32.33 >50 28 Feb - 19 Apr 30.33 0.62 28.52 31.93 >50 01 Mar - 21 Apr 30.58 0.77 29.01 32.39 >50 04 Mar - 26 Apr 30.15 0.51 28.38 30.95 >50 12 Mar - 26 Apr 30.4 1.02 27.78 31.7 >50 Chennai 14 Feb - 30 Mar 32.64 0.14 31.72 32.93 100 14 Feb - 30 Mar 32.31 0.17 31.32 32.66 100 16 Feb - 01 Apr 32.4 0.4 31.07 33.28 100 20 Feb - 05 Apr 32.04 0.7 30.82 33.6 100 23 Feb - 08 Apr 32.09 0.55 31 33.48 100 29 Feb - 15 Apr 33.68 3.61 30.32 35.76 100 29 Feb - 15 Apr 32.26 3.55 27.4 33.95 >50 10 Mar - 25 Apr 33.17 5.8 27.18 35.22 >50
DISCUSSION
In this study I aimed to examine the effect of temperature on the hatching success of L. olivacea nests at two sites across the range of nesting of this species on the East coast of
India. I also examined the expected sex ratios of the hatchlings, in the context of current differences in climate between the two sites. I argue that it is unlikely that current egg and hatchling mortality are linked to higher temperatures that cross the developmental thresholds of the eggs or hatchlings. However the data indicates that future changes in ambient temperature as predicted by regional climate change models (Kumar 2004) could have important impacts on the biology of the olive ridley turtles both in terms of sex ratio as well as egg and hatchling mortality.
43 There was no significant trend of mortality with temperature at both the study sites during the
study period. It seems unlikely that the trend of increased egg or hatchling mortality with time of incubation, over the past few years at Chennai, can be explained by increased temperature. Previous studies indicate a developmental threshold of 35ºC (Matsuzawa et al.
2002, Wibbels 2007), which was reached at both sites, but for very short durations. Further these temperatures were experienced only towards the end of the incubation period, where they might not have interfered with development to a great extent. Maximum temperature recorded was also unlikely to be a predictor of mortality, since those temperatures were exposed for very short durations and did not rise to a great extent above the thresholds (38ºC at Chennai and 37.2ºC at Rushikulya). This indicates the buffering ability of the nests, in contrast to the recorded surface temperatures. As much of the mortality in nests was accounted for by under-developed embryonic death (unpublished data), exposure to unsuitable temperature was not an appropriate explanation. In this study temperature was taken as an index of laying date and time of incubation. There are other factors associated with season or duration of incubation such as reduction in oxygen concentration of the nest environment (Ackerman 1977). Such factors are likely to explain mortality in L. olivacea nests better than temperature.
If temperatures above the threshold cause mortality even at short durations of exposure, I would expect a sharp rise in mortality for those nests, as well as mortality at that stage of development. A positive pattern emerged at Rushikulya but it was not supported by the data from Chennai which showed a slightly negative relationship between overall mean temperature and mortality. However Chennai faced an overall lower variance in temperatures than Rushikulya. The positive trend at Rushikulya and negative trend at Chennai is
44 suggestive of a parabolic relationship. But this cannot be justified physiologically. I also expected that regular variation in temperature over days or weeks would impact hatching success more than seasonal variation. Seasonal variation in temperature was unlikely to be of physiological significance to the development of an embryo, since the observed seasonal variation was not very drastic and the species is expected to have evolved to the local changes in nest environment over millennia. Further comparison of nest temperatures and surface temperatures (sand and air) reveal that that nest has some degree of buffering effect on temperature. An offshoot of this comparison is the fact that surface temperatures, as measured by the probe could be used in future studies as an index of nest temperature. This cheaper alternative will enable large scale monitoring of beach and nest temperatures.
Finally if temperatures did influence hatchling mortality, differential mortality should have been observed in the two sites. Since temperatures at Chennai were consistently higher than at Rushikulya, higher mortality rates would be expected at Chennai. Similarly variance was much greater at Rushikulya than Chennai. However the percentage of mortality observed at both sites was comparable. It is clear that hatchling mortality is being influenced by factors besides temperature. Ackerman (1977) reported various factors affecting hatching success including water retention by sand, un-seasonal rainfall, chemicals and limited gas exchange particularly when nests are crowded. Detailed studies on this suite of factors could help resolve the best predictors of hatching success in nests of L. olivacea along the East coast of
India.
Nevertheless temperature is a critical factor to a developing turtle embryo, since it is responsible for determining the sex of the embryo (Bull & Vogt 1979, Wibbels et al. 1991,
45 Georges et al. 1994, Booth et al. 2004, Hulin et al. 2008). Although adult sex ratios have not been established, a number of studies have shown that hatchling sex ratios are female biased
(Casale et al. 2000, Witzell et al. 2005). Predicted sex ratios from both sites in our study also indicate a female bias. The high minimum recorded temperatures in Chennai, during the sex determining period (middle third of incubation period) indicate the possibility that all nests from this site are producing females. The only two nests where temperatures below the pivotal temperature were recorded faced a spell of rain that might have resulted in cooling.
The overall temperature regime at Rushikulya was lower, allowing for the production of males. However the trend of mass nesting being delayed from January in the 1990s to March or April in the past few years (K. Shanker, unpublished data), could lead to skewed sex ratios even at this site.
Implications for the future
Current nest temperatures are very close to the physiological thresholds for development.
Predicted rises in global temperatures indicate that temperatures could increase by up to
3.5ºC by 2071 (Kumar 2004). Such a drastic increase in temperature could result in large scale mortality of olive ridley eggs and hatchlings. This could lead to a severe decline in the population. But even an increase of a lower magnitude could have long term impacts, resulting from skewing of sex ratios towards the production of females. The genetic implications for a skewed sex ratio in these turtles is something that needs to be studied in detail, before conservation measures to manipulate sex ratios are put into practice.
46 LITERATURE CITED
Ackerman RA (1977) The respiratory gas exchange of sea turtle nests (Chelonia, Caretta).
Resp Physiol 31:19 – 38
Booth DT, Burgess E, McCosker J, Lanyon JM (2004) The influence of incubation
temperature on post-hatching fitness characteristics of turtles. International Congress
Series 1275: 226 – 233
Bull JJ, Vogt RC (1979) Temperature-dependant sex determination in emydid turtles. J Exp
Zool. Science 206: 1186 – 1188
Bull JJ, Vogt RC, McCoy CJ (1982) Sex determining temperatures in turtles: a geographic
comparison. Evolution 36: 326 – 332
Casale P, Gerosa G, Yerli SV (2000) Female biased primary sex ratio of the Green turtle
Chelonia mydas, estimated through sand temperatures at Akyatan , Turkey. Zool
Middle East 20: 33 - 42
Dimond MT and Mohanty-Hejmadi P (1983) Incubation temperature and sex differentiation
in a sea turtle. Am Zool 23, 1017
Drake DL, Spotila JR (2002) Thermal tolerances and the timing of sea turtle hatchling
emergence. J Therm Biol 27 (1): 71 – 81
47 Georges A, Limpus C, Stoutjesdijk R (1994) Hatchling sex in the marine turtle Caretta
caretta is determined by the proportion of development at a temperature not daily
duration of exposure. J Exp Zool 40: 511 – 521
Glen F, Mrosovsky N (2004) Antigua revisited: the impact of climate change on sand and
nest temperatures at a hawksbill turtle (Eretmochelys imbricata) nesting beach.
Global Change Biol 10:2036–2045
Godfrey MH, Mrosovsky N (1994) Simple method of estimating mean incubation
temperatures on sea turtle beaches. Copeia 1994 (3): 808 - 811
Hawkes LA, Broderick AC, Coyne MS, Godfrey MH, Godley BJ (2007) Only some like it
hot: quantifying the environmental niche of the loggerhead sea turtle. Divers Distrib
13:447–457
Hulin V, Girondot M, Godfrey MH, Guillon JM (2008) Mixed and uniform brood sex ratio
strategy in turtles: the facts, the theory and their consequences. In: Wyneken J, Bels
V. and Godfrey M. H. (eds.), Biology of turtles: from structures to strategies of life.
Taylor and Francis, London
http://www.bbc.co.uk/weather. Accessed on 24th June 2008
http://www.metoffice.gov.uk/research/hadleycentre.html. Accessed on 25th June 2008
48
IUCN (2007) http://www.iucnredlist.org/search/details.php/11534/all. Accessed on 25th May
2008.
Kar CS (1982) Discovery of second mass nesting ground of the pacific olive ridley sea turtles
Lepidochelys olivacea in Orissa, India. Tiger Paper 9: 6–7
Kumar RK, Hingane LS, Borgaonkar HP, Sikder AB, Jadhav SK, Kothawale DR, Revadekar
JV, Ram S (2004) Regional aspects of global climate change and variability. In:
Annual Report 2003- 2004, Indian Institute of Tropical Meteorology, Pune, India.
Lang JW, Andrews HV (1994) Temperature-dependant sex determination in crocodilians. J
Exp Zool 270: 28 - 44
Lopez-Castro MC, Carmona R, Nichols WJ (2004) Nesting characteristics of the olive ridley
turtle (Lepidochelys olivacea) in Cabo Pulmo, southern Baja California. Mar Biol
145: 811 – 820
Matsuzawa Y, Sato K, Sakamoto W, Bjorndal KA (2002) Seasonal fluctuations in sand
temperature: effects on the incubation period and mortality of loggerhead sea turtle
(Caretta caretta) pre-emergent hatchlings in Minabe, Japan. Mar Biol 140:639–646
49 Mohanty-Hejmadi P, Dimond MT (1986) Temperature dependent sex determination in the
olive ridley turtle. In: Progress in Developmental Biology, Part A. Alan R. Liss, Inc.
159 – 162
Morreale SJ, Ruiz GJ, Spotila JR, Standora et al. (1982) Temperature dependant sex
determination: Current practices threaten conservation of sea turtles. Science 216:
1245 – 1247
Mrosovsky N (1980) Thermal biology of sea turtles. Am Zool. 20: 531-547
Mrosovsky N (2007) Translocating turtles: Trials, tribulations and triumphs. In: Instituto
Canario Ciencias Marinas; Marine turtles recovery of extinct populations. Monograph
Parmesan C, Yohe G (2003) A globally coherent fingerprint of climate change impacts
across natural systems. Nature 421: 37 – 42
Pina CI, Larriera A, Cabrera MR (2003) Effect of temperature on incubation period, sex
ratio, hatching success and survivorship in Caiman latirostris (Crocodilia,
Alligatoridae). J Herpetol 37: 199-202
Quadir DA, Shrestha ML , Khan TMA , Ferdousi N , Rahman M, Mannan A (2004)
Variations of surface air temperature over the land areas in and around the Bay of
Bengal. Natural Hazards. 31 (2): 561-584
50 R Development Core Team (2008) R: A language and environment for statistical computing.
R Foundation for Statistical Computing, Vienna, Austria.
Shanker K, Pandav B, Choudhury BC (2003). An assessment of the olive ridley turtle
(Lepidochelys olivacea) nesting population in Orissa, India. Biol Conserv 115: 149 –
160
Shanker K, Aggarwal RK, Rama Devi J, Choudhury BC & Singh L (2004) Phylogeography
of olive ridley turtles (Lepidochelys olivacea) on the east coast of India: implications
for conservation theory. Molecular Ecology 13, 1899 - 1909
VanDamme RD, Bauwens RD, Verheyen RE (1992) Thermoregulatory responses to
environmental seasonality by the lizard Lanicara vivipara. Herpetologica 43: 405 –
415
Wibbels T (2007) Sex determination and sex ratio in ridley turtles pp 167 – 189. In: Plotkin
PT (editor). Biology and Conservation of Ridley Sea Turtles. Johns Hopkins
University press, Baltimore, Maryland
Wibbels T, Bull JJ, Crews D (1991) Synergism between temperature and estradiol: A
common pathway in turtle sex determination? J Exp Zool 260 (1): 130 – 134
Wibbels T, Cowan J, LeBoeuf R (1998) Temperature –dependant sex determination in the
red-eared slider turtle, Trachemys scripta. J Exp Zool 281: 409 – 416
51 CONCLUSION
This study focused on the influence of lighting and temperature on the eggs and hatchlings of the olive ridley turtle (Lepidochelys olivacea). The short term impacts on the population were examined with respect to the impact of artificial lighting on hatchling orientation. Hatchlings reacted to both wavelength and intensity, and varied in response to different combinations of these two factors. Olive ridley hatchlings preferred lower wavelengths and higher intensities, which conformed to results from previous studies on other populations of this species. They were also indifferent to light in the red region of the spectrum at low intensities.
At Rushikulya, hatchlings were disoriented by intense point sources and glows arising from industrial and municipal lighting. The well spaced lights along the highway running close to the beach did not affect their orientation to a great extent. Disorientation was the least where
Casuarina equisetifolia plantations acted as a light barrier. However before such plantations are recommended as a suitable artificial light mitigation measure along the coast, their overall impact on the ecosystem must be considered.
Artificial lighting close to the Rushikulya river mouth threatens the mass nesting population.
If nesting shifts slightly south of its present location, a large proportion of hatchlings are expected to be affected by disorientation due to light. The site must be guarded against light pollution from potential development such as oil pipelines or ports. A combination of suitable wavelengths and intensities of artificial light, as well as the use of light barriers is recommended as a mitigation measure.
52 Egg and hatchling mortality were not found to be significantly affected by temperature at
either site. Although mortality increased with time, and temperature showed a similar trend,
the two were not correlated. Although there was a relationship between temperature and
mortality in Rushikulya, there was an opposite relationship at Chennai. More studies will be
required to resolve the impact of temperature on mortality.
The effect of temperature on sex ratio has been established, and on this basis, sex ratios of
hatchlings at both sites were predicted. According to the predicted sex ratios, hatchlings at
these and possibly other sites along the East coast of India were female biased. The impact of such a bias is unknown, although it is speculated to have genetic consequences, as olive ridleys are promiscuous and this is thought to give them a selective advantage.
By addressing the important factors of light and temperature influencing sea turtle eggs and hatchlings, this study has created a scientific base of knowledge that can be applied directly to the conservation of the species.
53