Date Published: June 12, 2018
Publisher: Public Library of Science
Author(s): Naoki Agetsuma, Emmanuel Serrano Ferron.
Hunting records have proven useful for examining the historical status of wildlife populations. The number of animals harvested can provide information on past population sizes that would have been required to support harvest yields. Therefore, when statistical data on annual harvests are available, a minimum estimate of past population sizes can be calculated. A very simple method for estimating the sizes of historic wildlife populations using only annual hunting records and the maximum annual population increase rate is presented in this study. This method was applied to estimate past population sizes for Japanese sika deer (Cervus nippon yesoensis) in Hokkaido Island, Japan, using hunting records from 1873 to 1882, and assuming 15% and 35% population increase rates. The annual number of deer harvested during 1873 to 1882 ranged from 15,000 to 129,000. The minimum population size in 1873 was estimated as 349,000–473,000. This method was validated by applying it to the eastern population of Hokkaido Island in 1993 when the population size was approximately 260,000, and population sizes estimated by this method were 0.50–1.17 times the nominal population size. Thus, the population estimates from this method were approximately equal to or less than the expected population sizes, and this method can be used to obtain minimum estimates of wildlife populations. Because shorter durations of hunting records result in population size underestimates, it would be better to use hunting record of 10 years or longer in this method. In addition, the degree of underestimation may change with hunting pressure intensity on the populations, other causes of mortality, and maximum annual increase rates of the species. The method can be applied to any wildlife species for which records of annual harvest and maximum annual population increase rates of the species are available. The estimates obtained can provide benchmarks for the population size required for ecosystem conservation, and can be useful for wildlife management as they indicate the lowest limit to maintain the population.
Wildlife population size has been estimated using many models that require information on various population parameters such as carrying capacity (for logistic growth models), age structure, age-specific survival and reproduction rates (for age structure models), demographic and environmental stochasticities (for logistic growth models and age structure models), catch-per-unit-effort (CPUE), and catch-effort (CE) (for Poisson catchability models) [1–3]. However, obtaining or estimating these parameters would be difficult or impossible for past populations. Therefore, the models requiring multiple population parameters as inputs can only be applied to the populations for which detailed ecological information is available. Models that use easily obtainable parameters, and require a small number of assumptions would have a much wider application for the estimation of past populations.
The minimum estimates of the population size of Ezo sika deer between 1873 and 1882 were calculated using Eq 2 with r-values of 15% and 35% (Fig 1, S2 Dataset). Back-calculation of the population size from 1882 to 1873 in a sequential manner showed that the estimated population of deer increased uniformly. The estimated numbers of deer in 1873 were 473,312 and 349,044 when the r-values were assumed to be 15% and 35%, respectively.
Similar to other models for the estimation of wildlife population sizes using hunting records, the model proposed in this study assumes that the number of harvested wildlife is correctly recorded [e.g. 2,20]. If there are many omissions of hunting reports and/or illegal hunting, the population will be underestimated by the method proposed. However, to obtain a minimum estimate of population size, amplification of hunting reports is a great concern that leads to overestimation, although it is unlikely to occur unless fraudulent rewards are claimed for killing wild animals. It should also be noted that the reported number of hunted animals is usually lower than the actual number of hunting-related mortalities because it does not include injured animals that escaped and died later. For white-tailed deer (Odocoileus virginianus), such cases may account for more than 30% of the total number of harvested .