The American Zoo and Aquarium Association (AZA) has approved over 2251 regional studbooks that form the basis for genetic and demographic management of populations in North American zoos and aquariums. These managed species are the most important to the educational, research, conservation, and recreational missions of AZA and its member institutions. The AZA currently supports over 80` SSP© that manage all specimens of a species or subspecies as a single population. In September of 1995, the AZA endorsed a plan to implement more rigorous genetic and demographic management of non-SSP© species. These Population Management Plans (PMPs) will also be based on data maintained in studbooks,
Studbooks were initiated to. provide accurate, up to date information in a standard format that could easily be used for genetic and demographic analyses of a single population. Quantitative analyses of data in a studbook are used to develop a management plan that is carried out through specific recommendations for each specimen in the population. Without studbooks, it would be virtually impossible to manage populations scientifically. Genetic and demographic recommendations for population management are only all good as the data upon which they are based. Thus, the quality and consistency of data in AZA studbooks are critical to the effectiveness of AZA's cooperative programs for species management; inconsistencies and errors can compromise the desired effects of even the best management plans.
Genetic and demographic analyses are remarkably sensitive to minor errors or inconsistencies in data. A misidentified parent, a poor or undocumented assumption, or an error during data entry can significantly alter recommendations resulting from quantitative analyses. Recommendations based on incorrect or misleading data may, for example, result in loss of genetic diversity or a false sense of security about a population's genetic or demographic status. These recommendations have the potential for adverse effects on individual institutions, individual specimens, and the population as a whole.
Accurate, unambiguous data are essential to assessments of a population's future: such data can best be ensured though standardization of data entry procedures for the
1,2 As of 1 August. 1995.
1
electronic databases that are the basis for all North American regional studbooks.
A studbook is intended to be a true record of a species's history in captivity. Each studbook should stand alone as a document and electronic database. A reader unfamiliar with the studbook should be able to examine and analyze the data without additional interpretation from the studbook keeper. When genetic and demographic analyses are conducted by someone other than the studbook keeper, inconsistencies, ambiguities, inaccuracies, and true errors impede interpretation and may require extensive verification. Much of the time presently devoted to genetic and demographic analyses is associated with data clean-up: verification, correction, and modification into a standard format employed by the AZA Small Population Management Advisory Group (SPMAG).
The SPMAG is a scientific advisory group of the AZA. Its members conduct research on scientific management of small populations, develop methods of implementing the best scientific approaches to population management, and perform or assist with demographic and genetic analyses for SSP© Master Plans and PMPs Standardized data, entry procedures will greatly facilitate; genetic and demographic analyses, this will decrease the time necessary for these analyses; and permit SPMAG members to assist with the growing number of managed populations
At present, there is much ambiguity with respect to (1) how data are entered into studbooks, (2) what data are entered in studbooks, (3) how data are maintained or altered, and (4) how data, .particularly explanations or assumptions, are documented. When studbooks are passed from one studbook keeper to another, the new studbook keeper must reconstruct those ambiguities (if possible). The SPMAG advisor, or population manager, must also interpret ambiguous data before quantitative analyses can be completed. Standardization and documentation of assumptions will prevent loss of information and/or redundant verification of data when studbooks are transferred from one studbook keeper to another and reduce the clean-up needed to prepare for quantitative analyses.
THE USES FOR STUDBOOK
DATABASES ARE LIMITED ONLY BY THE QUALITY AND QUANTITY
OF THE DATA
2
This manual is separated into four sections. The first section presents an overview of studbooks and background on how studbook data are used in population management. The second section focuses on common problems faced during data entry. This section addresses generalized solutions and, like the first and third section, emphasizes the consequences of many data entry decisions. The third section offers step-by-step instructions on how to enter data using SPARKS, the Single Population Analysis and Record Keeping System software designed and distributed by the International Species Information System (ISIS). These instructions are presented in the context of how certain decisions may have significant consequences for population management. Although this section contains important information about SPARKS, it should not be considered a substitute for the SPARKS manual. The fourth section provides information on how to obtain additional assistance with studbook data entry, and a description of data quality problems.
Each section is rife with subsections and subheadings for specific topics on data entry. These subsections are heavily cross-referenced and the reader,is. encouraged to follow those. Cross-references.
This manual uses the following conventions: SPARKS field names (the name or
type of data entered) will be presented as <BOLD CAPS INSIDE ARROWS>. lnformation that should be entered into a field (i.e., what is typed or selected during data entry) will be presented as {CAPITALS INSIDE BRACKETS}. Items seleated from menus or windows in SPARKS will be shown in Italics With Initial Letters In Capitals, Important concepts or actions will be in bold italics or in boxes.
Throughout this manual, the reader will be guided by Speedy, the small population biologist. Speedy will emphasize key points that link data entry to the genetic and demographic analyses that are the basis for population management.
3
SPARKS, the Single Population Analysis and Records Keeping System is the software used by all AZA regional and most international studbook keepers; it is an AZA requirement that ail official studbooks be maintained in SPARKS (see The AZA Conservation Resource Guide, available from the AZA Conservation Office). These guidelines wee written for SPARKS 1.4 and consider features and capabilities unique to that version of the software. Although this document contains much useful information QA SPARKS, it is not a substitute for the SPARKS manual (produced in 1989 and distributed by ISIS).
SPARKS was developed and is maintained by the International Species Information System (ISIS). Using SPARKS, it is possible to enter data, edit data that have been entered, enter additional data, produce a variety of reports, undertake several types of descriptive analyses, and export data for demographic and genetic analyses produced by other software. Thus, SPARKS is a data management and an analyticaI tool; it is particularly powerful in its abilities to select subsets of the database for reports, analyses, and to export data for genetic and demographic analyses.
It is imporant to note, However, that SPARKS does not always make literal use of data as they are entered. In many instances, SPARKS interprets data in ways that are not readily apparent either to the studbook keeper or to the SPMAG advisor. Date estimates, sex, and data that are "unknown" are often interpreted by SPARKS. Standardization of data entry and maintenance procedures will ensure that the bases for reports and analyses are consistent across studbooks.
A section of these guidelines provides detailed instructions on how to enter data into SPARKS. Among these step-by-step instructions are explanations of decisions that are often necessary during data entry. These explanations provide essential information about the consequences of various options. It is important for every studbook keeper to understand how SPARKS uses data: creativity or inconsistencies in data entry can have dire consequences for population management.
4
DECISIONS AND ASSUMPTIONS
MADE DURING DATA ENTRY AFFECT
THE VALIDITY OF DEMOGRAPHIC
AND GENETIC ANALYSES
GENES is the software used for genetic analyses of AZA populations. It was written and, is maintained by Dr. Robert Lacy of the Brookfield Zoo (near Chicago, IL). GENES calculates the basic genetic parameters (i.e., gene diversity, founder genome equivalents, kinship value; and mean kinship) used to construct breeding recommendations for AZA management plans. Data from a studbook maintained in SPARKS must be exported to GENES before analyses can take place. Although GENES can interpret some data it receives from SPARKS, most of the data are used as exported. Thus, GENES is very dependent on the quality of data in SPARKS, how SPARKS interprets the data it contains, and what data are exported for use by GENES.
DEMOG is software used for demographic analyses and modeling. it was written by Dr. Jon Ballou and Laurie Bingaman Lackey of the National Zoological Park, Washington, D.C. The SPARKS life table (Mx/Qx) is exported for use by DEMOG. DEMOG provides a thorough demographic analysis of the life table and options for manipulation of the life table to devise strategies for altering population growth rates. In addition, DEMOG calculates and exports reproductive value (V,), a variable used by GENES to calculate kinship value. DEMOG cannot interpret the data it receives from SPARKS: the data are used as exported. Thus, DEMOG is totally dependent on the quality of data in SPARKS, how SPARKS interprets the data it contains, and what data are exported for use by DEMOG.
5
SPARKS offers two routines that identify potential errors and inconsistencies in a studbook database. These routines are found under the Data Quality menu. Clean-up Files compares locations listed in the studbook with the ISIS institution list, reindexes all files, and deletes records marked for deletion during editing. Clean-up Files should be run before and atter each editing session. Data Validation is a new option (it replaces Batch Validate from SPARKS 1.3) that checks the studbook database for logic errors, possible file damage, and inappropriate, invalid or missing information. This new option provides two levels, coarse and fine, for checking a SPARKS database; it also reports changes made since the last check. Data Validation should be run, and its output examined and checked, before each publication of the studbook and before any demographic or genetic analyses are conducted.
The options available under the SPARKS Data Quality menu r present a powerful set of tools to evaluate data entered in the database. They n ident! a ran e of 'possible data problems, any of which may affect genetic or demographic anaiyses used to develop breedin-g recommendations that are the basis for PMPs and SSP©s. The potential probiems flagged by Data Quality may arise during any editing session; they represent data entry errors, iogical errors in re aborts from institutions (e.g., mother and offspring not ail same institution on birth date), ambiguities, and omissions.
Data Validation systematically looks at all the major fields for each specimen record in the studbook. It checks that all required information has been entered correctly, that essential information is not missing, and that there are not logic errors with conflicting pieces of information. Data Validation also counts missing or unknown data, verifies field justification and contents, notes specimens marked as lost-to-follow-up (see page 74 on lost-to-follow-up), checks parentage, and looks for inappropriate characters that may be an indication of file damage.
It is imperative that Data Validation be used regularly to verify the quality and usefulness of data in the studbook. If the potential problems identifiecf in a Data Validation report cannot be reconciled, corrected, or explained, the database is not of sufficient quality to be used for planning and management purposes. Regular use of Data Validation will check data acquisition and entry methods, provide warning of file damage, and identify a variety of other potential problems with the database.
6
A studbook is the pedigree and demographic history of a specificaily defined taxon such as a genus, species, subspecies, or other specific population. The studbook may cover a single species that contains no subspecies (e.g., hooded crane, king cobra, Indian rhinoceros), a single species and its various subspecies but without specimens of hybrid ar unknown origin (e.g., tiger), a single species and its various subspecies and hybrids (e.g., orang-utan), or a genus and as many full species as are deemed appropriate by the Taxon Advisory Group (TAG) and the AZA Wildlife Conservation and Management Committee (WCMC). A studbook is delimited by geographic region: either regional (e.g., North American regional studbook) or international.
A studbook is intended to be a true record of a species's history in captivity. It should include information on all specimens that have ever lived within the taxonomic and geographical scope of the studbook. It must also include the pc atprs of all animals in the studbook. When. portions of the species's history have been lost, the true record should acknowlege these missing portions (see page 19 on assumptions in studbooks). Each specimen is assigned e unique numerical identifier (i.e., studbook number) that allows construction of the pedigree (for genetic analyces) and of age specific schedules of birth and death (for demographic analyses). The studbook traces the entire history of each specimen; these collective histories comprise the population's genetic and demographic identity.
From the point when a specimen enters captivity (i.e., through capture, captive birth or unknown origin), the studbook tricks that specimen for its entire life, recording all locations where it lives and the time spent at each location. Any specimen that ever enters captivity, regardless of whether it is captured and dies without offspring, is born and lives a natural lifespan, is present only in the form of an aborted fetus or stillbirth, or is represented as a captive-born yeang which dies shortly after birth, should be entered into the studbook. In the end, the studbook records each specimen's date of death or transfer from the managed population.
During each specimen's life, supplementary information may also be recorded, as appropriate for that species, to help present and future studbook keepers track a specimen and better understand its history. By recording as much information as possible about each specimen, studbooks will assist animal managers to improve management of subsequent generations through well conceived breedings, PMPs and SSP©s, and scientific analyses of studbook databases.
7
By definition, a studbook keeper is an individual that compiles and maintains a pedigree and demographic database on a group of animals (usually a specific taxon) in captivity. However, studbook keeping is not a secretarial effort. The studbook keeper is a resource: a knowledgeable individual with a vast amount of information about a group of animals in captivity. Moreover, the studbook keeper is a researcher, who systematically and relentlessly pursues those individual facts that together portray the true history of the population. Studbook keeping requires keen insight and the ability to link disparate and obscure pieces of information into complete and accurate histories for individual specimens. Tracking down specimens of unknown origin, linking specimens that have disappeared from one location and seem to have reappeared at another, and resolving conflicting information are among the more im jortant tasks done by the studbook keeper. The studbook keeper is the expert on, and the chronicler of, the history of a specific captive population of animals. However, the studbook keeper may, or may not, be the foremost expert about the species's biology, its status in the wild, or its ecology.
The studbook keeper has been described as a detective. Data roust be tracked down by following leads and evaluating clues about the history of specimens in the captive population. These clues may contradict each other and the truth may be difficult to discover. Testimony in the form of correspondence, documents, telephone conversations, and institutional records must be closely scrutinized for accuracy and consistency. The studbook keeper must hone his or her detective skills to produce an accurate record. Finally, the studbook keeper must make every effort to verify information from the best source possible: the institution that held the specimen. In cases of conflicting information, the studbook keeper must come to a decision, much as a jury in a trial, to conclude the truth "beyond a reasonable doubt" (but see page 19 on assumptions in studbooks).
From the preceding description, it should be evident that studbook keeping is an educational process. As the keeper compiles the data, he or she acquires a unique familiarity with the captive population. This familiarity is imperative to good studbook keeping. It allows the processing and evaluation of data on a level that is not possible by computers. A studbook keeper with a complete knowledge of the captive population will use logic and intuition to catch inconsistencies and potential errors in data reported by the participating institutions.
The studbook keeper is directly responsible for acquiring and recording the life history for each specimen contained within the studbook. This individual is also responsible for entering this data into the studbook using SPARKS software. The
8
studbook keeper accomplishes this by maintaining contact with the zoos, aquariums, and individuals who currently have the studbook taxon or have had that taxon in the past; these communications may be mediated by SPARKS-generated questionnaires or direct correspondence. The studbook keeper is also responsible for publishing the studbook and annual updates.
As a resource to the zoological community, the studbook keeper maintains pedigree, demographic, and other data about a captive population. This information may then be used to manage the captive population on several levels. The most intensive management leveis ere the ex situ conservation efforts, such as the AZA SSP© programs. In these cases, the studbook keeper is an active participant in the development of a masterplan to aid the conservation of the studbook taxon. In circumstances where an SSP© exists and is directed by another individual, the studbook keeper is responsible for sending the species coordinator changes in the database (e, g., births, deaths, transfers, etc.) so that management decisions can be made which are based on the most current information available.
When no SSP© exists, the studbook keeper is encouraged to use SPARKS technology to develop a Population Management Plan (PMP). A PMP will come to many of the same conclusions as the SSP© about breeding recommendations but is not as intensive, with goals that may or may not have conservation as a central focus. A common goal for this level of management is to maintain a self-sustaining captive population. ln these cases, the studbook keeper acts as a species manager, making recommendations for breedings and monitoring population trends. Support for the studbook keeper is available from the AZA Small Population Management Advisory Group (SPMAG): no PMP should be distributed without review and approval by a member of the SPMAG.
Studbook keepers are frequently contacted by owners, potential owners, and governmental agencies for information about breeding recommendations, specimen origins or the worth of individual specimens to the captive population. In general, the studbook keeper should not give advice regarding population management unless he or she is the population manager. If a PMP or SSP© Master Plan exists, the studbook keeper who is not a population manager should refer all enquiries to the written plan.
9