Sea Turtle Population Ecology
287
large juveniles and/or adults have been identified as the stages with the highest
proportional sensitivity (Heppell, 1998). This means, for example, that the longterm average growth rate of a sea turtle population is more sensitive to a 10%
change in the survival rate of large juvenile turtles than it is to a 10% change in
the hatchling or adult survival rate, in large part because most of the population
consists of large juveniles when the population is at a stable stage distribution
(Heppell et al., 2000a).
The egg and hatchling stage is the most easily accessible stage of sea turtles.
Therefore, this stage has received the greatest amount of study and conservation
effort. However, models demonstrate the low contribution of egg survival to mean
population growth rate. Starting with a loggerhead population declining at about 5%
per year, Crowder et al. (1994) showed that a 90% decrease in egg/hatchling mortality was not enough to prevent the population decline, whereas a 50% decrease in
the benthic juvenile mortality alone resulted in positive population growth. This is
not to say that the egg and hatchling stages are not important. Increased egg or
hatchling survival cannot compensate for decreases in subadult and adult survival
rates; however, inputs from these stages are critical in maintaining recruitment to
the older stages. Egg harvest, coupled with fishing mortality, is thought to be the
primary cause of population crashes in Kemp’s ridley and Pacific leatherback turtles,
and fox predation of loggerhead eggs may be the cause of recent recruitment failure
in Australian loggerheads (Chaloupka and Limpus, 2001). Efforts to conserve nesting
beaches and protect eggs from harvesting are important and need to be continued;
however, conservation efforts that focus resources on this stage alone are not providing optimal benefits to population recovery.
Adult females exhibit a strong degree of nest site fidelity, allowing for the
possibility of recapturing the same turtle when she returns to nest. Therefore, most
estimates of sea turtle survival rates are for adult females (Table 11.2). An assumption
of survival rates estimated from nesting females is that females who nest at the
monitored beach will return to that beach to nest. It is known that nest site fidelity
in sea turtles is not perfect (Miller, 1997) and that females may try out beaches
before selecting a nesting beach. In analyses of nesting data, females that are tagged
and never seen again are assumed dead, when they may have moved to another
beach to nest. In most cases, then, survival rates estimated from nesting data would
underestimate actual survival rates.
Studies to estimate juvenile survival rates are complicated by their multiple
habitat use and highly migratory nature. Catch curves (Seber, 1982) have been used
to calculate survival rates for the elusive juvenile stages. Frazer (1987) applied the
technique to two cross-sectional data sets of loggerheads, one of dead strandings
and one of trawl-caught turtles. Survival rates for U.S. loggerheads have recently
been updated using new growth curves based on skeletochronology and mark–recapture (National Marine Fisheries Service, 2001). Catch-curve analysis assumes that
populations are stable through time and that the data set used represents a cross
section of the true population. Cohort analysis avoids the assumption of a stable
population, and has been applied to Kemp’s ridleys, where cohort strength is known
(Turtle Expert Working Group, 2000). To prepare a catch curve or cohort analysis,
1123 book.book Page 287 Tuesday, November 12, 2002 7:43 AM
287
large juveniles and/or adults have been identified as the stages with the highest
proportional sensitivity (Heppell, 1998). This means, for example, that the longterm average growth rate of a sea turtle population is more sensitive to a 10%
change in the survival rate of large juvenile turtles than it is to a 10% change in
the hatchling or adult survival rate, in large part because most of the population
consists of large juveniles when the population is at a stable stage distribution
(Heppell et al., 2000a).
The egg and hatchling stage is the most easily accessible stage of sea turtles.
Therefore, this stage has received the greatest amount of study and conservation
effort. However, models demonstrate the low contribution of egg survival to mean
population growth rate. Starting with a loggerhead population declining at about 5%
per year, Crowder et al. (1994) showed that a 90% decrease in egg/hatchling mortality was not enough to prevent the population decline, whereas a 50% decrease in
the benthic juvenile mortality alone resulted in positive population growth. This is
not to say that the egg and hatchling stages are not important. Increased egg or
hatchling survival cannot compensate for decreases in subadult and adult survival
rates; however, inputs from these stages are critical in maintaining recruitment to
the older stages. Egg harvest, coupled with fishing mortality, is thought to be the
primary cause of population crashes in Kemp’s ridley and Pacific leatherback turtles,
and fox predation of loggerhead eggs may be the cause of recent recruitment failure
in Australian loggerheads (Chaloupka and Limpus, 2001). Efforts to conserve nesting
beaches and protect eggs from harvesting are important and need to be continued;
however, conservation efforts that focus resources on this stage alone are not providing optimal benefits to population recovery.
Adult females exhibit a strong degree of nest site fidelity, allowing for the
possibility of recapturing the same turtle when she returns to nest. Therefore, most
estimates of sea turtle survival rates are for adult females (Table 11.2). An assumption
of survival rates estimated from nesting females is that females who nest at the
monitored beach will return to that beach to nest. It is known that nest site fidelity
in sea turtles is not perfect (Miller, 1997) and that females may try out beaches
before selecting a nesting beach. In analyses of nesting data, females that are tagged
and never seen again are assumed dead, when they may have moved to another
beach to nest. In most cases, then, survival rates estimated from nesting data would
underestimate actual survival rates.
Studies to estimate juvenile survival rates are complicated by their multiple
habitat use and highly migratory nature. Catch curves (Seber, 1982) have been used
to calculate survival rates for the elusive juvenile stages. Frazer (1987) applied the
technique to two cross-sectional data sets of loggerheads, one of dead strandings
and one of trawl-caught turtles. Survival rates for U.S. loggerheads have recently
been updated using new growth curves based on skeletochronology and mark–recapture (National Marine Fisheries Service, 2001). Catch-curve analysis assumes that
populations are stable through time and that the data set used represents a cross
section of the true population. Cohort analysis avoids the assumption of a stable
population, and has been applied to Kemp’s ridleys, where cohort strength is known
(Turtle Expert Working Group, 2000). To prepare a catch curve or cohort analysis,
1123 book.book Page 287 Tuesday, November 12, 2002 7:43 AM