Integrating behaviour into wildlife conservation: the multiple ways that behaviour can reduce N_e

Abstract

There has been a recent interest in integrating an understanding of behaviour into conservation biology. Unfortunately, there has been no paradigm for such a process. Without a clear framework for integration, conservation biologists may have difficulties recognising how behavioural knowledge can help solve real-world conservation problems. Effective population size (N_e) is a key demographic parameter used to understand population viability. A variety of behaviours and behavioural traits impact N_e, yet their importance for conservation is under-appreciated. We suggest that identifying behavioural traits that affect N_e provides a paradigm for integrating behavioural biology into conservation biology. Behaviour can affect N_e through at least three different mechanisms: reducing N — the population size; reducing r — the population growth rate, and/or by increasing reproductive skew. We discuss how nine common behavioural traits can reduce N_e, and suggest how an understanding of these traits may inform management of both free-living and captive animals.

Introduction

Conservation biology is a crisis discipline aimed at saving biodiversity (Soulé, 1986). A common and important approach to saving biodiversity has been to preserve patches of habitat in order to maintain ecosystem-level processes which, in turn, preserves populations of species (Franklin, 1993). With these broad objectives and methods, conservation biologists have paid relatively little attention to how an individual animal's behaviour can help save biodiversity. Recently, a number of behavioural biologists have written reviews and book chapters on the role of behaviour in conservation arguing that a fundamental understanding of behavioural processes can contribute to conservation biology (Caro & Durant, 1995, Hoglund, 1996, Lima & Zollner, 1996, Ulfstrand, 1996, Curio, 1996, Clemmons & Buchholtz, 1997, Strier, 1997, Caro, 1998a, Caro, 1998b, Sutherland, 1998). Despite this recognition that it may be important to apply knowledge of animal behaviour to conservation problems, there is no clear framework to help conservation biologists identify the specific cases when they should be concerned about behaviour, nor which behaviours they should be concerned about.

The number of individuals in a population, N, is a first approximation of endangerment. However, other factors influence the likelihood of a population going extinct over time. For instance, variation in the number of breeding individuals, variation in breeding success, the ratio of breeding males to females, as well as other factors influence the maintenance of genetic variation in a population (Falconer, 1989). Genetic variation influences long term sustainability because genetic variation is required to combat any negative effects of inbreeding, and to allow evolutionary adaptation to an ever-changing environment. To compensate for the inadequacy of N alone in predicting the likelihood of a population persisting over time, population geneticist's have developed the concept of the effective population size, N_e which better reflects the likelihood of a population persisting over time (Gilpin and Soulé, 1986). N_e is an estimate of the theoretical number of breeding individuals assuming they behave in an ideal way. N_e models an ideal population with the following properties: the population is split into sub-populations where there is no migration between sub-populations, generations do not overlap, the number of breeding individuals is the same for all generations and sub-populations, mating is at random and includes a random amount of self-fertilisation, there is no selection, and mutation is assumed to be unimportant (Falconer, 1989). N_e affects population viability by increasing homozygosity and decreasing the number of non-selected alleles. The loss of variation is compounded by an increase in linkage disequilibrium — nonindependent assortment of alleles — which reduces the frequency of novel gene combinations. N_e is influenced by factors that halt the passing of gametes to the next generation. Falconer (1989) identified six factors which influence N_e: (i) exclusion of closely related matings, (ii) skewed sex ratios, (iii) unequal generation size, (iv) unequal family size, (v) inbreeding, and (vi) overlapping generations.

A number of behavioural traits either directly or indirectly influence N_e by changing demographic parameters that contribute to N_e. We define behaviour broadly and recognise a hierarchy that begins with the neurobiological, genetic, and physiological processes that underlie observed motor patterns, includes the functional integration of those motor patterns into behaviours, as well as the integration of behaviours into behavioural traits. For instance, infanticide — a complex behavioural trait where adults kill young of their own species — directly influences N_e by reducing the population size, while another behavioural trait, reproductive suppression of adults, reduces the number of breeding individuals and may decrease a population's rate of increase — r — therefore reducing N_e. To integrate behaviour into wildlife conservation, we must understand how behaviour skews the operational sex ratio-the ratio of breeding males to females. Behavioural traits can influence the operational sex ratio in a number of ways. For instance, mature animals may be prevented from mating by dominant conspecifics or there may be active mate choice mechanisms that prevent certain animals from reproducing. Behaviour contributes to predation risk and skewed operational sex ratios emerge via differential mortality and survival — a ‘double whammy’ for species already in danger of extinction. Wright (1938) illustrated how skewed sex ratios influence N_e. Specifically,

$$\text{N}{}_{\mathrm{<mtext>e</mtext>}}\text{=}\text{1}\text{1}\text{4N}{}_{\mathrm{<mtext>m</mtext>}}\text{+}\text{1}\text{4N}{}_{\mathrm{<mtext>f</mtext>}}$$

where N_f is the number of breeding females and N_m is the number of breeding males.

From this we clearly see that N_e decreases by either decreasing the number of breeding individuals or by skewing breeding sex ratios.

Population viability analysis (PVA) is an important tool for managers because it provides an estimate of the viability and sustainability of a population. PVAs model the effect of certain biotic (fecundity, age of senescence) and abiotic factors (habitat availability) on N_e. It is through the calculation of N_e that a PVA predicts population persistence. N_e is therefore a central parameter determining population viability in PVAs.

As we will discuss in a number of following examples, individual behavioural strategies influence how individuals respond to habitat modification, hunting, fragmentation, corridor construction, reduced resource quality, and resource fluctuation. Understanding factors that influence behaviour over short time scales provides vital information for those developing more accurate population models as well as for those charged with managing populations. It is therefore surprising that most PVA models ignore behavioural variation (Derrickson et al., 1998).

N_e·s central importance in PVA models suggests that the best strategy for integrating behaviour into conservation will involve identifying behaviours and behavioural traits that impact N_e. N_e has already been recognised as one of a series of ways in which behavioural ecology can contribute to conservation biology (Parker & Waite, 1997, Caro, 1998a). Our approach differs in that we view that identifying and modeling the ways in which behaviour influences N_e as the perhaps the single most important way in which knowledge of animal behaviour can contribute to wildlife conservation.

When dealing with threatened populations, N_e is commonly quite low. Using knowledge of animal behaviour to design management regimes may only marginally increase N_e, yet this may be all that is needed to ensure population viability.

Behaviour can affect N_e through its effects on N,r, and reproductive skew. N,r, and reproductive skew influence N_e in at least five ways by their solitary and combined effects (Fig. 1). In this paper we discuss nine common behavioural traits that either directly or indirectly affect N_e. We focus on reproductive suppression, sexually-selected infanticide, mechanisms of mate-choice, mating systems, social plasticity, dispersal, migration, conspecific attraction, and reproductive behaviours which require special resources. These behavioural traits are found in many taxa. By discussing these, we hope to illustrate how behaviour can help inform conservation biology and how a wildlife manager might go about determining whether knowledge of behaviour should be applied to a particular conservation question.