The most common, and most important, use of studbook data is for demographic and genetic management of captive populations. However, there are a number of additional uses that can impact which data are entered. Three uses of studbook data are given and discussed below.
Multi-Institutional Genetic and Demographic Management: Studbooks are the basis for genetic and demographic analyses that guide SSP© Master Plans and PMPs. These plans are aimed at slowing loss of genetic diversity and maintaining or developing self-sustaining populations (i.e., not dependent on imports from other regions or the wild). Two kinds of analysis, genetic and demographic, are usually performed to assist population management. Genetic analyses assess ioss of genetic diversity among defined populations, within a population, end within induviduals (inbreeding). Demographic analyses include assessment of populations growth rates, the fecundity and mortality that determine population growth, and projection of future numbers under various management possibilities.
Animals have been kept and bred by humans for thousands of years. This history of breeding (domestication) has inveriably. entailed artificial selection for desirable characteristics. Humans have purposefully bred animals most likely to have, for examples, the most- or least-fatty meat, best dispositions, most fur, or greatest milk production. Through selective breeding of domesticated species, desirable characteristics are increased in frequency and less desirable characteristics ace decreased. This traditional, selective genetic management remains the basis for breeding strategies among most professionally managed domesticated species.
In contrast to domesticated species, breeding in zoos and aquariums strives to maintain populations that are as genetically similar as possible to the natural (wild) population. To maintain a captive population that is genetically similar to its wild ancestry, all wild-caught specimens should have the same number of offspring: their offspring should also all have the same number of offspring. Deviation from these ideals reflects loss of genetic information. For example, if only those specimens that are easy to breed are bred, in relatively few generations the population may become very good at breeding and surviving in captivity. However, that does not mean that it will be good at breeding or surviving in nature.
Unfortunately, it is not realistic to expect all specimens to have the same number
22
of offspring. Genetic management of many captive populations was not started when the first wild-caught specimens were brought into captivity: for these populations it is already too late to have all wild-caught specimens leave the same number of offspring. Even in intensively managed populations, chance events (e.g., deaths, differences in litter size) will cause some specimens to leave more offspring than others. The genetic aspects of SSP© Master Plans and PMPs use a variable called mean kinship to identify breedings that will restore, to the greatest extent possible, equality of offspring production among the founder lineages.
If is possible to minimize selection for captivity by making breeding recommendations based on mean kinship; this is the most common use of studbook data. This type of genetic management is not the same as inbreeding avoidance. Inbreeding avoidance is a common feature of many types of genetic management including selective breeding for desired traits and breeding to maintain a population of "genetically wild" specimens. Close inbreeding (a g., mating full siblings) is avoided to minimize juvenile mortality rates and other deletenous effects on individual specimens, but does not necessary affect the other genetic goals of population management.
Recommendations based on mean kinship have susbstantial impacts on participating institutions: specimens may be moved, bred, or not bred with regard to their individual mean kinships. Mean kinship is calculated directly from the pedigree information in SPARKS studbook databases: errors or omissions in those databases can result in recommendations that do not stern the loss of genetic diversity. Moreover, incorrect mean kinships may:result in unnecessary moves or recommendations not to breed.
Single Institution Collection Management: Individual institutions may use studbook data for purposes other than SSP© or PMP management. For example, studbook data can allow institutions to evaluate whether they are having unusual success or failure with a species. If animal management staff believes that their institution has had a large number of deaths in the past 10 years, their hypothesis could be evaluated using data in the studbook. Studbook data would reveal whether the number of deaths was unusual given the size and age structure of that institution’s collection. Age-specific mortalities calculated from the studbook data would reveal whether mortality rates were higher than average among ail institutional holders. Without a studbook as a reference, it would not be possible to determine what "normal mortality" rates were, and so it would not be possible to evaluate whether there was a significant mortality problem at any institution. Studbooks allow a multi-institutional perspective that would not otherwise be possible.
23
Animal rights groups have tried using legal means to stop animal moves among zoos (e.g., gorillas, elephants), and to have entire groups of animals listed as not suitable far captivity (e.g., cetaceans). One strategy has been to question the ability of zoos and aquariums to care for the specimens. A current example of this strategy is the Humane Society of the United States (HSUS) call to prohibit the keeping of marine mammals in captivity. Its representatives have testified before US Congressional committees and they have written articles about the longevity rates of wild and captive dolphins and killer whales. Their claims and the numbers they use are often unsupported by data, and even if the claims are supported by data it is important to evaluate the source of the data and the appropriateness of their analysis. Two recent examples:
"Calculations taken from the study showed that on average the expected life span of a bottlenose dolphin in captivitj could be as little as 14 years, while in the wild the dolphin could.live twenty to twenty-nine years." (HSUS News, Winter 1995)
"Their [captive cetaceans] mortality rates are aberrantly high and captive breeding is virtually not happening " (HSUS Report - Small Whale Species: The Case Against Captivity)
If records are not kept there is little that can be said in response to these claims. The article containing the first quote cited- no data source other than an ongoing study. In that case, there is essentially nothing to refute other than the illogic of the comparison of average lifespan in captivity to maximum age recorded in the wild. The data used in the HSUS Report data were from the National Marine Fisheries Service Marine Mammal Inventory. Although studbook data would be preferable, those data can be used to evaluate these claims. If the claims are false this can be shown. If the claims are in fact true or even partially true, the data can be used to document the scope of the problem (a first step toward addressing the problem). Without data the only response would be a different opinion. In the battle of words, the burden of proof is often on the accused rather than the accuser and therefore, it is best to respond armed with hard data.
Unique Value of Studbook data: Zoological institutions have an opportunity to contribute data on species biology which can almost never be obtained in the wild. For example, detailed pedigree information from zoos has shown that inbreeding depression is a widespread phenomenon in captive populations of non-domesticated mammals. For
24
another example, Indian rhino population viability analyses, performed for the indian government in support of conservation of the wild population, were based on demographic information from the studbook of the captive population. Accurate recording and assembly of parentage and birth and death date information is extremely difficult in the wild. Thus, information which is contained in studbooks is often the only data that can be used to assess pedigree related questions for non-domesticated species.
Applied Science: In general, it is expected that the captive populations of the future will descend from the captive populations of the present. Therefore, it is imperative that have good quality studbooks be maintained and appropriate scientific analyses guide management for the long term. The SSP© program itself can be seen as applied science. Development of genetic variation maintenance breeding strategies, methods of dealing with specimens of unknown ancestry, and methods of predicting the rates of loss of genetic variation from captive populations have all been developed in part from and for studbook data The AZA SSP© are both scientific research and scientifically-guided management.
Basic Science: Data from zoos and aquariums can be used to answer important questions in a variety of biological disciplines. For example, the cheetah has become the class example of how loss of genetic variation can affect the probability of survival of an entire species. Acceptance of this idea by much of the scientific comrnunity is evident by its appearance as an example in college freshman level text books However, there is currently a debate over whether the cheetah is really in danger of extinction because of an apparent lack of genetic variation. One of the lines of evidence used to support the idea that the cheetah was genetically imperiled was the Intemational Cheetah studbook. .Based on analysis of the studbook it was concluded that low fertility rates and high juvenile mortality rates were caused by a lack of genetic variation. The cheetah studbook, as well as other felid studbooks, has been used to evaluate the hypothesis that the cheetah has unusually low fertility and/or high juvenile mortality.
25
Most studbook databases are too large to search through manually. For this reason a studbook keeper can opt to examine a subset of the SPARKS database. Selection of subsets, called the "view" in SPARKS, may be based on almost any combination of data entry fields. For example, a studbook keeper could create a studbook report that only contained males born at a specific institution within a single year. Views are often necessary for studbook data analysis and can be very helpful when compiling and updating the studbook.
Views in SPARK$ are selected through the Retrieval Criteria (left arrow from the Reports menu) listed under the View Criteria menu. The View Criteria permits selection of data subsets based on dates when specimens were in the population, age, sex, geographic range, management plan (e.g., SSP©), event type (birth, capture, release, death, transfer), origin (captive-born, wild-caught, or unknown), organizational association (e.g. AZA), living or dead, and <USER DEFlNED> fields (UDFs, see page 80 on how to create and use UDFs).
When selecting a view it is important to be sure that ail necessary date field criteria are set to obtain the correct subset of the data. This may sound both easy and obvious, but it takes practice and an understanding of how SPARKS works For example, to determine the number of specimens in the studbook living in North America today, the View Criteria would seem to be "Living = Yes" and "Geographic = North America." Wowever, that view of the data would select all those currently living specimens that live or have ever lived in North America. Specimens that at some point in their lives were in North America, but are currently living outside of North America, will also be included. The necessary View Criteria for the number of specimens living in North America today are "Geographic = North America", "Living = Yes", "Date Span = yesterday to today", and "Date Span = during".
GENES is the software currently used to construct genetic analyses from data maintained in SPARKS. Prior to any genetic analyses, data must be exported from SPARKS to a format that contains the appropriate information for genetic analyses. When genetic analyses are done on studbooks, it is very important that the view be set to include the subset of the specimens in the studbook on which the genetic analysis is to be performed. For this to be possible, specimens must be correctly identified as to location,
26
and status (i.e., alive, dead, or lost-to-follow-up). If no view is explicitly requested, the SPARKS software will make an assumption about which specimens are to be included. The default (no view) View Criteria in SPARKS and GENES includes all specimens ever recorded in the studbook that have not been recorded as having died. Thus, specimens that can no longer be located or are presumed dead (i.e., lost-to-follow-up), have moved out of the geographical region, or are unavailable for breeding would be included in the genetic analysis.
Genetic analysis of a population usually starts with an assessment of how much of the genetic variation (measured in various ways) that was present in the original source population (usually the wild) has been retained to date in the studbook population. Genetic analysis then is used to provide a ranking of the top priority specimens for breeding and the best pairings from the studbook population. The analysis assumes that all specimens within the view are available for breeding (or will be after they grow up) and that no other specimens are or will be available for breeding in the population. The measures of genetic variation, and the priority specimens identified for breeding, will change when different subsets (views) of the studbook are analyzed A specimen,might not be considered a priority breeder because it has many offspring or siblings in the population, but that specimen could become the priority breeder if many of those sibhngs were to die. For analysis of a breeding population in one geographic area (e.g., North America, or an SSP© or one zoo), a specimen would become a higher priority breeder if and when some of its siblings and offspring were sent out of the regional population. thus, which specimens are included in an analysis will affect not only the summary statistics on the genetic variation of the population, but also the genetic assessment of each specimen as a potential breeder.
IT IS ESSENTIAL TO GENETIC
ANALYSES THAT ALL ANCESTORS -
EVEN THOSE THAT HAVE NEVER
LIVED IN THE NORTH AMERICAN
REGION - ARE ENTERED INTO THE STUDBOOK
The pedigree of all specimens in the view is needed for a genetic analysis. Thus, for a complete and accurate genetic analysis, parents, grand-parents, and all ancestors back to the original wild-caught founders must be recorded in the studbook. For example, if some specimens were imported from outside the regional population, then a genetic
27
analysis for the regional population will be incomplete if the ancestors outside of the region are not in the studbook. The analysis will give the wrong numbers if some of the imported specimens are related to each other but this is not shown in the parentage in the studbook. Thus, it is important to include in any studbook the ancestors (if known) -- all the way back to wild-caught founders -- of all specimens in the region. These ancestors must be in the studbook for genetic analyses but can be selected out of other analyses and reports by setting the appropriate view (e.g., North America).
DEMOG is the software currently used to construct demographic analyses from data maintained in SPARKS. Prior to any demographic analyses, data must be exported from SPARKS to a format that contains the appropriate information for demographic analyses. any of the considerations for setting views for demographic analyses are similar to those for genetic analyses (see above). However, demographic views may: differ from the view for genetic analyses of the same population! Demographic analyze often entail setting a date widow to exclude a period of time when husbandry or population management differed greatly from the present. For example, the history of many vertabrate populations in zoos begins with a period of growth fueled almost exclusively by imports reproduction is typically low during this period as managers work out the best management schemes to promote reproduction. Thus, demographic analyses intended to assess patterns of reproduction with an eye toward predicting future population size would need to exclude this period of non- or low reproduction.
DATA SHOULD NOT BE EXCLUDED
FROM THE STUDBOOK JUST
BECAUSE THEY SEEM TO HAVE NO
GENETIC VALUE
DEMOGRAPHIC AND GENETIC VIEWS
MAY BE VERY DIFFERENT AND WILL
OFTEN INCLUDE DIFFERENT
SPECIMENS
28
Specimen ownership can be a very involved issue, particularly with the advent of multi-party breeding loans. Although owners often report information that indicates specimens were transferred via loans, tracking the actual ownership of studbook specimens is not the responsibility of the studbook keeper. The terms of individual transfers should be tracked by the institutions themselves. In most cases, there is no reason for ownership to be included in the database. Ownership does not usually impact development of population management plans.
There are good reasons to omit ownership from studbooks. For example, there have been several instances of ownership records in studbooks being consulted for legal issues. Some state and governmental agencies may consider studbooks to be the official (legal) record of a specimen’s history. Under some interpretations therefore, a "transfer" could be construed as a sale, loan, donation, trade or exchange unless different types of transfers are clearly noted for every specimen in the studbook. Under extreme circumstances, this could also place the studbook keeper ligitation.
Ownership has, in the past, been confused with location. This is often a problem with births to specimens on loan, when the data for ownership and birih location are accidentally reversed during data entry. Data Validation may not catch these problems or it may only identify the consequences (e.g., a specimen not at its birth location). Confusion over ownership vs. physical location may be difficult to identify and thus difficult to correct. The risk of confusion is even greater if ownership is only tracked for some specimens or if all ownership transactions are not entered for a single specimen.
For some species, ownership information may be useful to the operation of cooperative management plans. When there is a special reason to track ownership, it is possible to do so in SPARKS. However, the studbook keeper should only track ownership when (1) it has been discussed and requested by a management group or population manager, (2) complete ownership data are available for every living specimen, and (3) ownership data are carefully verified upon entry into the database to avoid mistakes. Ownership issues concerning SSP© or other recommended moves require communication directly between the institutions involved; the Species Coordinator is not responsible for tracking ownership per se.
29