Limits...
The Use of Surrogate Data in Demographic Population Viability Analysis: A Case Study of California Sea Lions.

Hernández-Camacho CJ, Bakker VJ, Aurioles-Gamboa D, Laake J, Gerber LR - PLoS ONE (2015)

Bottom Line: The projected population trajectories approximated the observed trends when surrogate data were used for one colony but failed to match for a second colony.Our results indicate that species-specific and even region-specific surrogate data may lead to erroneous conservation decisions.These results highlight the importance of using population-specific demographic data in assessing extinction risk.

View Article: PubMed Central - PubMed

Affiliation: School of Life Sciences, Arizona State University, Tempe, Arizona, United States of America.

ABSTRACT
Reliable data necessary to parameterize population models are seldom available for imperiled species. As an alternative, data from populations of the same species or from ecologically similar species have been used to construct models. In this study, we evaluated the use of demographic data collected at one California sea lion colony (Los Islotes) to predict the population dynamics of the same species from two other colonies (San Jorge and Granito) in the Gulf of California, Mexico, for which demographic data are lacking. To do so, we developed a stochastic demographic age-structured matrix model and conducted a population viability analysis for each colony. For the Los Islotes colony we used site-specific pup, juvenile, and adult survival probabilities, as well as birth rates for older females. For the other colonies, we used site-specific pup and juvenile survival probabilities, but used surrogate data from Los Islotes for adult survival probabilities and birth rates. We assessed these models by comparing simulated retrospective population trajectories to observed population trends based on count data. The projected population trajectories approximated the observed trends when surrogate data were used for one colony but failed to match for a second colony. Our results indicate that species-specific and even region-specific surrogate data may lead to erroneous conservation decisions. These results highlight the importance of using population-specific demographic data in assessing extinction risk. When vital rates are not available and immediate management actions must be taken, in particular for imperiled species, we recommend the use of surrogate data only when the populations appear to have similar population trends.

No MeSH data available.


Mean absolute error (MAE) between the count data and the simulated population values based on a range of survival values (0.72–0.82) for prime-age adult survival at G (a). Survival values that minimized the MAE resulted in simulated trajectories that fit the observed population trend at G (b). Spa = prime-age adult survival, and Soa = old adult survival.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4587556&req=5

pone.0139158.g004: Mean absolute error (MAE) between the count data and the simulated population values based on a range of survival values (0.72–0.82) for prime-age adult survival at G (a). Survival values that minimized the MAE resulted in simulated trajectories that fit the observed population trend at G (b). Spa = prime-age adult survival, and Soa = old adult survival.

Mentions: The survival rate that resulted in simulated trajectories that best fit the observed population trend for G was 0.78 for prime-age adult females and 0.73 for old adult females (Fig 4). These values are notably lower than the surrogate survival rates (0.97 and 0.91, respectively) from LI used in the original model.


The Use of Surrogate Data in Demographic Population Viability Analysis: A Case Study of California Sea Lions.

Hernández-Camacho CJ, Bakker VJ, Aurioles-Gamboa D, Laake J, Gerber LR - PLoS ONE (2015)

Mean absolute error (MAE) between the count data and the simulated population values based on a range of survival values (0.72–0.82) for prime-age adult survival at G (a). Survival values that minimized the MAE resulted in simulated trajectories that fit the observed population trend at G (b). Spa = prime-age adult survival, and Soa = old adult survival.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4587556&req=5

pone.0139158.g004: Mean absolute error (MAE) between the count data and the simulated population values based on a range of survival values (0.72–0.82) for prime-age adult survival at G (a). Survival values that minimized the MAE resulted in simulated trajectories that fit the observed population trend at G (b). Spa = prime-age adult survival, and Soa = old adult survival.
Mentions: The survival rate that resulted in simulated trajectories that best fit the observed population trend for G was 0.78 for prime-age adult females and 0.73 for old adult females (Fig 4). These values are notably lower than the surrogate survival rates (0.97 and 0.91, respectively) from LI used in the original model.

Bottom Line: The projected population trajectories approximated the observed trends when surrogate data were used for one colony but failed to match for a second colony.Our results indicate that species-specific and even region-specific surrogate data may lead to erroneous conservation decisions.These results highlight the importance of using population-specific demographic data in assessing extinction risk.

View Article: PubMed Central - PubMed

Affiliation: School of Life Sciences, Arizona State University, Tempe, Arizona, United States of America.

ABSTRACT
Reliable data necessary to parameterize population models are seldom available for imperiled species. As an alternative, data from populations of the same species or from ecologically similar species have been used to construct models. In this study, we evaluated the use of demographic data collected at one California sea lion colony (Los Islotes) to predict the population dynamics of the same species from two other colonies (San Jorge and Granito) in the Gulf of California, Mexico, for which demographic data are lacking. To do so, we developed a stochastic demographic age-structured matrix model and conducted a population viability analysis for each colony. For the Los Islotes colony we used site-specific pup, juvenile, and adult survival probabilities, as well as birth rates for older females. For the other colonies, we used site-specific pup and juvenile survival probabilities, but used surrogate data from Los Islotes for adult survival probabilities and birth rates. We assessed these models by comparing simulated retrospective population trajectories to observed population trends based on count data. The projected population trajectories approximated the observed trends when surrogate data were used for one colony but failed to match for a second colony. Our results indicate that species-specific and even region-specific surrogate data may lead to erroneous conservation decisions. These results highlight the importance of using population-specific demographic data in assessing extinction risk. When vital rates are not available and immediate management actions must be taken, in particular for imperiled species, we recommend the use of surrogate data only when the populations appear to have similar population trends.

No MeSH data available.