«International Journal of Comparative Psychology, 2003, 16, 65-84. Copyright 2003 by the International Society for Comparative Psychology Application ...»
International Journal of Comparative Psychology, 2003, 16, 65-84.
Copyright 2003 by the International Society for Comparative Psychology
Application of Behavioral Knowledge to
Conservation in the Giant Panda
Ronald R. Swaisgood
Zoological Society of San Diego, U.S.A.
Xiaoping Zhou, Gwiquan Zhang,
Wolong Nature Reserve, China
Donald G. Lindburg,
Zoological Society of San Diego, U.S.A.
Wolong Nature Reserve, China
Over the past several years we have developed a research program to increase knowledge of giant panda behavioral biology to facilitate efforts to reproduce giant pandas in captivity, particularly at the Wolong Breeding Center in Sichuan, China. Studies of estrus and reproductive behavior have enabled us to better pinpoint the timing of the fertile period and evaluate problems when a pair's behavior diverges from the norm. Experiments with chemosignals demonstrate that pandas possess a sophisti- cated chemical communication system, extracting information about the signaler’s sex, reproductive status, age, social status, and individual identity, as well as the age of the chemosignal. The effects of scent on sexual motivation have important applications for captive breeding. Given the proper behav- ioral environment, most pandas now mate naturally at Wolong. Following observations documenting several behavioral problems, we developed an environmental enrichment program to reduce abnormal behaviors, encourage behavioral diversity, and promote well being. Females are monitored closely for signs of pregnancy and we are evaluating behavioral and morphological indices to distinguish preg- nancy from pseudopregnancy. Identification of pregnant females is important because they are man- aged differently than nonpregnant females. In cases of maternal abandonment of cubs, we have devel- oped a method to train the mother to accept her cubs. The result of such behavioral research, coupled with efforts by other disciplines and basic husbandry changes, is a dramatic increase in the number of natural matings, pregnancies, births and cub survivorship.
The highly endangered giant panda may be the most popular animal in the world. What other animal draws such huge crowds and the constant scrutiny of the global media? Given such popularity, it is surprising how slow scientists have been to take up the cause and devote their time and energies to understanding their biol- ogy and behavior. Despite some early efforts (e.g., Chaudhuri et al., 1988;
Kleiman, 1983; Pan & Lü, 1993; Schaller et al., 1985), many aspects of panda Many have contributed substantially to the development of this research and management program and/or data collection. Among the Wolong staff, major contributors include Wei Rongping, Li Dasheng, Hu Damin, Wu Dafu, Huang Yan, and Han Hongyin. Collaborators and assistants from the Zoological Society of San Diego include Angela White, Megan Owen, Suzanne Hall, Valerie Hare, Laura McGeehan and Staci Wong. Many others too numerous to mention provided invaluable support and assistance. Correspondence concerning this article should be addressed to R. R. Swaisgood, Center for Reproduction of Endangered Species, Zoological Society of San Diego, P. O. Box 120551, San Diego, CA 92112, U.S.A. (firstname.lastname@example.org).
-66biology remain a mystery. Before the late 1990s, most published accounts of captive pandas were based on one or two individuals. The success of captive breeding programs depends in part on the acquisition of such knowledge and its application to the species’ management in captivity.
In recent years there has been a surge of research on the giant panda—almost all in captivity—by several institutions, including Zoo Atlanta, Chengdu Giant Panda Breeding Base, National Zoological Park, Wolong Breeding Center, and the San Diego Zoo. These research projects cover a diverse array of topics, including nutrition, veterinary care, endocrinology, reproductive biology, genetics, and behavior. As a result, new insights into giant Panda biology, as well as conservation breakthroughs, are on the horizon. The purpose of this paper is to describe a part of this research effort, namely the behavioral research program at the Wolong Breeding Center in Sichuan, China.
Rigorous behavioral studies of captive animals in zoo settings have been notably rare, and for good reason. Zoo-based studies are notorious for their small sample sizes and consequent lack of statistical rigor, often because zoos generally maintain a single breeding pair or small group, following the "Noah's Ark" approach. Moreover, goals of husbandry and management, as well as public exhibition, almost inevitably run counter to research needs. These obstacles tend to be magnified for endangered species held at zoos. With endangered species comprised of a small number of founders, it is crucial to obtain breeding and genetic representation from all founders quickly so that genetic heterozygosity is maintained (Ralls & Ballou, 1986). At this stage experimental manipulation that might compromise well being or reproduction in some individuals, even if the ultimate goal is to better understand processes that promote well being and reproduction, is justly ruled out.
For monoestrous species such as the giant panda, where the opportunity to obtain reproduction comes but once a year, this problem is further exacerbated. Also, the effects of any scientifically-guided, long-term change is almost certain to be confounded by other changes, such as diet, veterinary care, and husbandry practices, rendering determination of cause and effect difficult.
Behavioral researchers are beginning to find a way over--or often around-these obstacles. In an exemplary study, although still with relatively small sample sizes, Wielebnowski and colleagues (Wielebnowski et al., 2002) were able to manipulate housing for female cheetahs, with predicted detrimental effects on stress and reproduction. Female cheetahs housed socially with other females showed behavioral signs of elevated stress and ovarian cyclicity was suppressed compared with those housed singly. This study was possible because (1) the cheetah population is not critically low and (2) social housing was already widespread, and thus research activities would not negatively impact current practices, and could improve them. Such "deprivation" studies are, however, frequently not feasible. Instead, a combination of educated guesswork and compromised scientific investigations proceed hand-in-hand toward an optimal management strategy. Consider the ideal study design to determine the environmental requirements for the development of normal behavior, absent stereotypies and other behavioral problems. It would involve the systematic manipulation of rearing environment, with some animals kept in small, barren cages, and other experimental groups reared in various enriched environments. Clearly, this design would not be ideal for animal well
-67being and reproduction in an endangered species. Add to that the almost certain need for animal transfer between institutions for genetic management, and opportunities for such lofty experimental goals vanish.
One way around these obstacles is for researchers to capitalize on "accidental" experiments, for example, studying animals that are housed in different environments. Although a reasonable alternative, often animals are moved around from one enclosure to another, so, developmentally speaking, their behavioral phenotype is the consequence of multiple enclosures and husbandry regimes. An example of accidental experiments from our own work comes from studies of stress. To determine if noise affected indices of stress in two giant pandas, we recorded amplitude levels daily to see if days characterized by loud noise were associated with behavioral and hormonal measures of stress (Owen et al., in press). While neither experimentally ideal nor statistically generalizable to the species as a whole, this study did provide insight into the behavioral biology of stress, and aided in management of these two animals. Another strategy is to manipulate some variable that will not harm the animal or reproduction, but may be correlated with reproductive success or otherwise provide insight into reproduction.
For example, with giant pandas an ideal experiment to determine the role of scent in reproduction would be to deprive some pairs of any olfactory exchange and evaluate whether aggression levels were higher and mating behavior less successful. Both ethical and conservation issues make such a manipulation impractical. Thus, our strategy has been to look at other effects of conspecific odors on sexual and aggressive motivation. For example, do pandas show increased sexual motivation (as evidenced, for example, by vocalizations) when exposed to certain odors? Such studies led to increased understanding of the effect of odors on motivational processes in support of reproduction, and suggested improved management strategies for captive breeding (Swaisgood et al., in press).
Despite these constraints, it is clear that sound science can be conducted with captive endangered species, and that such science, although sometimes falling short of ideal, can play an instrumental role in development of improved captive breeding techniques. Clearly, such constraints and opportunities arise in giant panda conservation. In almost direct proportion to its appeal, the giant panda is notorious for its reluctance to breed in captivity. In 1996 at the IUCN Captive Breeding Specialist Group meeting in Chengdu, China, Chinese researchers estimated that only 39% of captive female pandas had produced offspring and as many as 80% failed to display normal behavioral estrus. In 1996 only six living males in captivity had ever mated naturally (Lindburg, Huang, & Huang, 1998), and the population was not self-sustaining. In the past six years, however, there has been a steady increase in the number of births and surviving cubs, predominantly at Wolong and the Chengdu Breeding Base. At Wolong the captive population has tripled from about 25 to more than 70 individuals in the past seven years. Wolong’s success has hinged to a large degree on the implementation of a highly successful natural mating program. This program owes much of its success to behavioral management strategies based on knowledge or assumptions about what is important for pandas in the wild, coupled with detailed scientific studies to determine the behavioral needs and abilities of pandas, and how these map onto well being and reproduction.
-68Here we present for the first time a summary of these various research activities, providing a synthesis of previously published work and an update that includes unpublished efforts. These research activities have provided insight into many aspects of the behavioral biology of giant pandas, yielding knowledge we have applied to captive breeding efforts and hopefully—in the not-too-distant future—to conservation in the wild. While no specific applications to wild populations have been made to date, we have taken advantage of the presence of these captive animals to learn a great deal about panda biology, much of which should prove to be relevant to wild pandas as well.
Behavioral Ecology in the Wild
Relatively little is known about the giant panda's behavior in the wild.
What is known is based on two long-term field studies in the Wolong Nature Reserve (Schaller et al., 1985) and Qinling mountains (Lü, Pan, Zhu, Wang, & Wang, 2000; Pan & Lü, 1993). The giant panda lives a solitary existence, devoting most of its time to consumption of its primary food source, bamboo. Bamboo makes up about 99% of the panda's diet, yet the panda's digestive tract is inefficient at assimilating protein from this food source. As a result, pandas spend about 55% of their time eating bamboo, 43% resting, and only 2% of their time is devoted to remaining activities (Schaller et al., 1985). Both males and females occupy home ranges that overlap extensively with conspecifics of both sexes, with some evidence of core areas that are mutually exclusive intrasexually. Throughout most of the year pandas encounter one another infrequently, avoiding direct contact with or responding aggressively to conspecifics. Although animals in some populations appear somewhat more social, this asocial temperament is expressed in both intraand intersexual encounters. Outside the mating season, pandas vocalize infrequently. Long-distance acoustic communication is limited to male "song medleys" heard during the mating season. By contrast, pandas capitalize on opportunities for chemical communication (Schaller et al., 1985). In areas of range overlap, pandas create traditional sites used to deposit urine and scent from a specialized anogenital gland. These communal scent marking stations, consisting of several large conifers in a prominent location, are visited by a number of pandas from the neighboring community. Pandas also mark on isolated communal posts along the network of trails utilized by several pandas.
Females experience a single estrus each spring and males show seasonal rut-like behavioral and physiological changes (Kleiman, 1983; Platz et al., 1983).
The fertile period lasts 1-3 days. Several males may locate isolated females in estrus and vie for mating opportunities; the female is known to mate with more than one male on some occasions (Schaller et al., 1985). Following fertilization of the egg, the development of the embryo is arrested and uterine implantation delayed for a period of several weeks to months (Monfort et al., 1989). Accordingly, the gestation period ranges from less than three months to approximately six months (Zhu et al., 2001). At birth the infant panda is the most altricial of all eutherian mammals and the mother expends a great deal of effort on the care of her young.
The cub remains in a den for the first 15 weeks of life, and is weaned and separated from the mother at 18-24 months of age.
Overview of Methods