Continuation and proliferation of populations of species with a pelagic larval dispersal phase are reliant upon successful recruitment of larvae back into those populations. The source of any recruits into a population may stem directly from the larvae released by that population (a closed or retention system) or alternatively may originate from a population in a different locality (an open system). In terms of resource management it is crucial that the source of recruits into a population is known. For example, if a population based primarily on a closed system of recruitment is heavily over-exploited then as the number of sexually mature females is reduced through hunting, then so the total number of larvae release into the ocean by those females will also be reduced. This then will reduce the magnitude of any recruitment back into the population until the stage is reached where the number of recruits returning to the population, even under the most favourable of conditions, is insufficient to cover losses from the adult population stemming from natural mortality and hunting effort. The net result is recruitment failure resulting in a non-viable population which will eventually become wiped out. As such populations dependent on open system recruitment are less susceptible to, and can sustain higher levels of, exploitation in that depletion of numbers of sexually mature females in the population will have no direct effect on the subsequent magnitude of possible recruitment into that population.
Prevalence of open or closed system recruitment in maintaining a population is dependant upon the dominant pattern of larval dispersal in the pelagic environment. Within a geographical region the distribution and abundance of larvae is influenced by 1. the duration of the larval life-cycle, 2. the bathymetric distribution of larvae over time, 3. the response of larvae to environmental stimuli, and 4. the direction and speed of oceanic currents at various depths. Quantifying these factors to enable determination of the type of recruitment, open or closed, occurring in discrete populations is extremely difficult.
An alternative method of identifying the nature of recruitment into a population involves examining the genetic variation between coconut crab populations in adjacent geographical regions. Genetic homogeneity would suggest that there is a relatively free mixing of larval stock between geographically isolated populations and therefore that open system recruitment is the norm for such populations. Genetic heterogeneity would suggest that little or no mixing of larval stock occurs between geographically isolated populations and therefore that closed system recruitment is the norm for such populations.
In a genetic study of coconut crab populations in Vanuatu, The Solomon Islands, The Cook Islands and Niue, Lavery and Fielder (1991) found that the Vanuatu and Solomon Islands coconut crab populations appear to have originated from the same genetic stock, but that coconut crab populations on Niue and The Cook Islands are genetically separate and independent. On this evidence it would appear that the coconut crab population on Niue is maintained via a closed-system pattern of recruitment in which the source of any recruits is from those larvae released from the island.
Successful recruitment, be it open or closed type, of pelagic larvae into a discrete population is reliant upon a complex interaction between oceanic currents, suitable larval behaviour, plus a large element of chance. Although functional, this system is extremely unreliable in terms of constancy of annual recruitment levels. Fletcher et. al. (1991) suggests that in Vanuatu the stochastic pattern of coconut crab larval dispersal is such that successful recruitment is both infrequent and sporadic, to the extent that coconut crab populations in the Vanuatu archipelago probably experience large scale recruitment only every 5 – 10 years. The generality of such a temporal pattern of larval recruitment appears supported by size-frequency data for Christmas Island crabs (Schiller 1988a).
If Fletcher's lower estimate of 5 years is taken as the period between successful recruitments into coconut crab populations the implication is that for an individual female to be assured of replacing herself, by way of releasing larvae into the ocean and having them recruit back into the population, then she must release eggs for at least a period of 5 years (given that only one batch of eggs is produced by a female per year). Hence in terms of a minimum viable population (MVP) (defined here as the size and size frequency distribution of a population which has an acceptable probability of continuation in the long term) coconut crab populations must contain females at least 5 years older than the age at which sexual maturity is attained.
Available data on attainment of sexual maturity pertains only to minimum size, not age. As mentioned previously estimation of age from size in coconut crabs based on the inverse function of the Von Bertalanffy growth function (VBGF) is fraught with inaccuracy. However the unreliability of age from size estimations is a function of the actual size and for the relatively small TL sizes relevant to this study, estimates of age should incorporate only small errors. Given the usefulness of these estimates such inherent errors are considered acceptable. A VBGF curve for females was calculated using Von Bertalanffy parameters and moult increment data from Fletcher (1991) in conjunction with data collected from Niue and Christmas Island (refer figure 9). All estimates of age from thoracic length and of thoracic length from age given below are derived from the produced female growth curve.
The smallest recorded sexually mature female coconut crab (as indicated by the presence of eggs) was located on Niue in 1987 and had a TL of 19.7mm (Schiller et. al. 1991). On Niue the mean minimum size at which females became sexually mature was approximately 22mm (5–6 years old). Consequently the MVP for coconut crabs on Niue must contain, in ‘reasonable numbers’, females at least 10 – 11 years old, which corresponds to a TL of approximately 31mm. Therefore on Niue female crabs in the size range 21 – 31mm TL can be considered as the reproductive core of the coconut crab population, without which the population would become non-viable (given a 5 year pattern of recruitment).
As has been stated in previous sections, analysis of survey results clearly indicates that the Niue coconut crab population is suffering from prolonged over-exploitation. It was suggested in section 5.4 that the level of over-exploitation was/is extreme, resulting in a small population of relatively small individuals. The true magnitude of the extent of over-exploitation is apparent when the female size-frequency data (Table 4) is examined in light of the MVP requirement that females of at least 31mm TL must be present.
Approximately 90% of female coconut crabs have a thoracic length of 32mm or less, leaving just 10% of available females which can be ‘safely’ exploited without seriously jeopardising the reproductive viability of the population. This is a precariously small safety margin, particularly in light of the wide-ranging assumptions underlying calculation of the 31mm TL requirement for a MVP. Even if recruitment occurs more frequently than every 5 years as has been assumed, the situation on Niue is extremely serious-exploitation of female crabs is rapidly nearing the stage where the population's reproductive core will come under exploitative attack and breakdown of the recruitment process will ensue. While the rugged terrain of Niue makes it unlikely that hunters could completely eliminate coconut crabs from the island, the very small numbers of crabs remaining would result in males having great difficulty in locating females for the purpose of mating. Hence egg production would be extremely low as would be the magnitude of any recruitment stemming from these eggs. In terms of exploitable stock the coconut crab would in effect be rare to nonexistent.