Coral Reefs

, Volume 32, Issue 1, pp 281–284

Decline in sea snake abundance on a protected coral reef system in the New Caledonian Lagoon


  • C. Goiran
    • Labex CORAIL, Laboratoire LIVEUniversité de la Nouvelle-Calédonie
    • School of Biological Sciences A08University of Sydney

DOI: 10.1007/s00338-012-0977-x

Cite this article as:
Goiran, C. & Shine, R. Coral Reefs (2013) 32: 281. doi:10.1007/s00338-012-0977-x


Monitoring results from a small reef (Ile aux Canards) near Noumea in the New Caledonian Lagoon reveal that numbers of turtle-headed sea snakes (Emydocephalus annulatus) have been in consistent decline over a 9-year period, with average daily counts of snakes decreasing from >6 to <2 over this period. Causal factors for the decline are unclear, because the site is a protected area used only for tourism. Our results suggest that wildlife management authorities should carefully monitor sea snake populations to check whether the declines now documented for New Caledonia and in nearby Australian waters also occur around the islands of the Indo-Pacific.


ConservationHydrophiineSurveysSea snake


There is growing evidence of widespread population declines in species from several vertebrate lineages, including mammals (Whitehead et al. 1997; McLoughlin et al. 2003), birds (LaDeau et al. 2007; King et al. 2008), reptiles (Gibbons et al. 2000; Winne et al. 2007; Reading et al. 2010), amphibians (Wake 1991; Alford et al. 2001; Keisecker et al. 2001; Pounds et al. 2006) and fishes (Feyrer et al. 2007; Light and Marchetti 2007). In some cases, the causes of those declines are readily identifiable, often involving habitat degradation or overexploitation of wild populations. In other cases, however, declines have occurred even in near-pristine protected areas (e.g., Winne et al. 2007; Reading et al. 2010). Such cases are particularly concerning, because they suggest some underlying threatening process that is not being dealt with effectively by current conservation and management strategies (Wake 1991; Kiesecker et al. 2001; Collins and Storfer 2003).

There are strong logistical challenges to obtaining accurate counts of the abundance of large cryptic predators such as snakes, making it difficult to detect any long-term changes in abundance (Gibbons et al. 2000). Nonetheless, a recent analysis based on 17 snake populations (of a total of 12 species) reported consistent declines over the last 20 years in taxa from Europe and Africa (Reading et al. 2010). All of these taxa were terrestrial species, reflecting the scarcity of long-term ecological data on sea snakes (Heatwole 1999). However, Lukoschek et al. (2007) documented the disappearance of sea snakes from reefs in the southern Great Barrier Reef over a 30-year period, and Guinea (2012) recently reported the virtual disappearance, between 1998 and 2008, of a previously abundant hydrophiine (sea snake) fauna at Ashmore Reef on Australia’s Sahul Shelf.

Materials and methods

Since 2003, volunteers under the auspices of the Centre d’Initiation à l’Environnement (CIE) have conducted standardized counts of sea snakes as well as other marine animals at a small island 0.95 km off the coast of Noumea, in the southern part of New Caledonia (22°19′S, 166°26′E). Ile aux Canards (“Duck Island”) is 10,830 m2 in area (Catala 1950) and has been protected from fishing or other faunal harvesting since 1989. This islet was the subject of the first marine ecology study in New Caledonia, published by Catala (1950). It is a popular destination for tourists, with hundreds of holidaymakers taking the 5 min taxi–boat ride from Anse Vata to the islet every day in suitable weather. The CIE has established a 400-m-long underwater route, marked by five buoys, along which snorkelers can view the coral reef ecosystem. The depth along this snorkel route ranges from 2 to 7 m, along the edge of a fringing reef with about 90 % coral cover ( The high latitude and large lagoon area of New Caledonia protect it from coral bleaching events, but there was minor bleaching in 1998.

In addition to policing tourist activities and providing information to tourists during the weekends, the volunteers regularly conduct standardized snorkel trips to count marine animals on the 400-m-long snorkel route. On those trips (which take about 30 min), the volunteers record all snakes that they see. By far, the most abundant sea snake in this area, and (as far as we can determine) the only one represented in the counts, is the turtle-headed sea snake Emydocephalus annulatus. This small (to <1 m total length) hydrophiine species feeds exclusively on the eggs of demersal-spawning fishes (gobies, blennies, damselfish: Voris 1966) and is abundant on inshore reefs near Noumea (Ineich and Laboute 2002; Shine et al. 2003, 2010). Although we have also seen two sea krait species (Laticauda laticaudata and L. saintgironsi) at this site, these species are uncommon in the area and were not recorded during the standardized surveys.

Results and discussion

In total, we have data for surveys on 913 days over a 9-year period (2003–2011) but have information on sea snake numbers for only 453 of those days (N = 10–71 days per year). Our analysis of the data from counts by volunteers reveals two major patterns. First, more snakes were seen per day in some months than in others (ANOVA with month as the factor, F11,452 = 6.42, p < 0.0001; Fig. 1a). Second, the number of snakes seen has declined consistently over the survey period, from an average of >6 per day in 2003 down to <2 per day in 2011 (ANOVA with year as factor, F8,358 = 4.72, p < 0.0001; Fig. 1b). Two-factor ANOVA (with year and month as factors) shows that the monthly variation in snake counts has remained consistent over the years (interaction NS; both main effects remain significant at p < 0.001).
Fig. 1

Temporal variation in counts of turtle-headed sea snakes on coral reefs around a small island (Ile aux Canards) in the Lagoon of New Caledonia, over a 9-year period. Snakes were seen more often at some times of year than at others (a), but overall sightings declined consistently over the survey period (b). Graphs show mean values and associated standard errors, based on a total of 453 days’ data

The monthly variation in snake counts (Fig. 1a) likely is due to seasonal shifts in snake activity. The period when snake numbers are lowest (February–May) is the time of year when females are heavily gravid, and hence cease feeding (C. Goiran pers. obs.). Because they are inactive at this time, the females are difficult to see. High counts in July–August fall within the mating period, when adult male Emydocephalus move about actively in search of mates (and hence are easily seen: Shine 2005).

In contrast, the consistent decline in snake numbers through the survey period from 2003 to 2011 (Fig. 1b) is difficult to explain. Very high philopatry of this species in two adjacent areas (Lukoschek and Shine 2012) suggests that snakes are unlikely to have dispersed away from the Ile aux Canards, so the reduced rates of encounter probably reflect actual declines in snake abundance. The only factor likely to have changed over that period involves the direct and indirect effects of protection and of increasing tourist numbers. Snorkelers and motor boats may frighten the snakes, and increased turbidity due to people and boats may have reduced the abundance and reproductive success of the pomacentrid (damselfish) taxa (Wenger et al. 2012) whose eggs provide much of the food for this sea snake population. Alternatively, protection of the system may have caused changes in the composition of the faunal assemblage in ways that have disadvantaged sea snakes: for example, reinvasion of large predators (e.g., sharks) that are known to consume sea snakes (Heatwole 1999; Ineich and Laboute 2002). These possibilities remain speculative, because we have no data on fish abundance. In his 1950 study, the only sea snake species cited by Catala was Laticauda colubrina. Apparently, no Emydocephalus lived around Ile aux Canards in 1950.

In summary, standardized counts of sea snakes by volunteers over a 9-year period have documented a strong decline in abundance of turtle-headed sea snakes on a protected coral reef system just off the city of Noumea. There has been relatively little anthropogenic modification to the system, so the decline might be the result of some process happening at a broader scale (e.g., variation in water quality), or the result of protection per se, or of habitat degradation due to tourism. Regardless of the causal mechanism, our results suggest that wildlife management authorities should carefully monitor sea snake populations in case the massive declines recently documented in nearby Australian waters (Guinea 2012) also occur around the islands of the Indo-Pacific. Sea snakes play a vital ecological role in coral reef ecosystems, so their conservation warrants a high priority (Heatwole 1999; Ineich et al. 2007).


We thank the Centre d’Initiation à l’Environnement and the many volunteers who gathered the data. Manuscript preparation was funded by the Australian Research Council.

Copyright information

© Springer-Verlag Berlin Heidelberg 2012