A studbook may only be started in North America after the species and its studbook keeper have received official approval by the AZA's Wildlife Conservation and Management Committee (WCMC). All studbook petitions must have an endorsement from the appropriate AZA Taxon Advisory Group (TAG). The TAG will assist with proscribing the taxonomic and geographic scope of the studbook. WCMC evaluates petitions for North American regional studbooks and international studbooks heid by North American institutions. After approval by WCMC, petitions for international studbooks are forwarded to the International Union of Directors of Zoological Gardens (IUDZG) for approval at the international level.
Instructions and forms for petitions to initiate a new studbook may be found in the AZA Conservation Resource Guide.
10
In North America, all studbooks must be approved by AZA's Wildlife Management and Conservation Committee (WCMC), and must be maintained in and produced by the Single Population Analysis & Record Keeping System (SPARKS).
There are three major steps to compile and complete an edition of a studbook: data acquisition and verification; data collation and compilation; and publication. The first step is the input of data into the data management system (SPARKS), the second is the processing of the data, and the third is the production of output. These steps are essentially the same for the first edition and each subsequent revision.
QUESTIONNAIRES SHOULD NOT BE
DISTRIBUTED UNTlL THE STUDBOOK
HAS BEEN APPROVED
Upon receiving approval for a studbook, the studbook keeper should contact ISIS to request a SPARKS dataset for the studbook taxon. SPARKS datasets are extracted from the ISIS database and distributed in the SPARKS backup format. Taxonomic classification is reported by an institution and a single specimen may change taxonomic status as it moves from one institution to another. Additional requests should be made for related taxa (e.g., subspecies, species, or genus) because it is common for specimens to be misidentified by institutions reporting to ISIS. Thus, in response to these requests, ISIS will provide a diskette containing all information for each taxon that has ever been sent to ISIS. This dataset will be in the SPARKS format; when accessed through SPARKS it will appear as if it is a studbook database (see page 30 on SPARKS datasets from ISIS). Because SPARKS datasets contain only a portion of the information in the ISIS database, studbook keepers may find it useful to obtain specimen reports directly from ISIS or from the lSIS database distributed on CD-ROM (compact disk).
3 "SPARKS dataset" and "ISIS SPARKS dataset" are used to denote a extract from the ISIS database. "SPARKS database" is used to denote a studbook database compiled and maintained by a studbook keeper.
11
Owners and locations listed within the SPARKS dataset are the first sources that a newly approved studbook keeper should contact for current and historical information on the studbook taxa. Many institutions do not participate in ISIS and many that do participate have submitted little, if any, historical data. Thus, it is important to consult every source possible in pursuit of the complete captive history of the studbook taxon. Other valuable sources include the "Census of Rare Animals in Captivity" and "Species of Wild Animals Bred in Captivity" sections of the International Zoo Yearbook. Reptiles and Amphibians in Captivity: Breeding, Longevity, and Inventory, published annually by Frank Slavens, is a good source for locating holders of reptiles and amphibians. Institutions that do not report to ISIS are often mentioned by other institutions as sources or recipients of specimens. These non-ISIS institutions should also be contacted for information on origin and disposition of specimens in the studbook,
MARVIN JONES, ALAN SHOEMAKER
AND ANDY ODUM ARE ALSO
EXCELLENT SOURCES OF
INFORMATION
After loading the information contained within the SPARKS dataset into the SPARKS software, the studbook keeper's first task in verifying information should be to print out and mail a questionnaire to each institution reported as ever possessing the species. This will provide a good start, but will usually not yield all of the necessary information to complete a studbook. With the questionnaire, there should be a letter specifically requesting Animal Record Keeping System (ARKS) taxon reports, ARKS specimen reports, other historical information about the specimens held at the institution not entered into ARKS, and leads on locations that may also have held, or hold specimens. It is recommended that, for AZA member institutions, the initial questionnaire be sent to institutional records keepers, not to curators or keepers. Questionnaires sent outside of North America might best be directed to curators. Follow-up questions about information not contained in the initial response to the questionnaire may require contact with staff other than the registrar. To encourage a prompt response, the studbook keeper should include a cover letter with the questionnaire stating that the studbook is approved by WCMC and therefore official in stature. This letter should also state whether the studbook is regional or international. A reasonable response time should be requested in the letter along with a statement that a historic report generated by the ARKS will suffice
12
in lieu of a completed questionnaire. Because many institutions, particularly dealers and non-AZA institutions are difficult to contact or elicit responses from, the studbook keeper should be prepared to contact reporting institutions for additional information pertaining to addresses or locations of a specimen’s ultimate disposition.
As data are acquired, collation and compilation can start. It is not necessary to have all possible data in hand before this phase is started; There are as many ways to sort and compile studbook data as there are studbook keepers. Each studbook keeper will develop a personal style for collating and compiling data. Some studbook keepers start by directly entering data into SPARKS, or editing an existing ISIS SPARKS dataset. Others use cards or paper to assemble and collate the data. Regardless of the approach used to organize and collate data as it is received, it is recommended that new studbooks be startd from scratch instead of editing or just cleaning the SPARKS dataset from ISIS. SPARKS datasets contain many subtle errors, ambiguities,and inaccuracies that are difficult to detect. An edited SPARKS dataset almost invariably retains some of these potential problems (see page 30 on SPARKS datasets from ISIS).
In the data management system, the studbook keeper could be considered a filter through which data passes. A good studbook keeper scrutinizes the information that has been submitted from the various sources for illogical events, obvious errors, ambiguity, and inconsistencies. If there are problems, the data are rejected until additional information is received for clarification. The Data Quaiity routine in SPARKS is an invaluable assistant in identifying potential errors and inconsistencies. The human perspective is essential to evaluate and interpret potential problems with the data; this is beyond the scope of current computer technology and is an important component of studbook keeping.
Publication is largely a printing and paper duplication process. The quality of the end product varies from printed, perfect bound volumes to photocopied materials. Studbook reports produced directly from SPARKS do not constitute a complete studbook: an introduction, husbandry section, and bibliography sections are also required (see the AZA Conservation Resource Guide). The initial studbook is often distributed to all institutions that are either currently holding the species or have had the species in the past.
13
For population management, the most important part of a studbook is not the printed volume, it is the SPARKS database (on floppy diskette). Direct analysis can be performed on the SPARKS database to determine the best possible breedings and demographic trends.
Although it is tempting to defer publication of the studbook until the database is historically complete, the AZA Conservation Resource Guide specifically states that all North American regional studbooks must be published within 18 months of authorizatien by WCMC. International studbooks must be published within two years of authorization.
14
The first two steps of the three-step process of producing a studbook (i.e., data acquisition, verification, and collation and compilation) are repeated yearly. The third step, publication, may or may not be repeated yearly. It is recommended that a full version of the studbook, including both historical population and current living population by location, be published every three years. In years the full studbook is not published, the studbook keeper may distribute a copy of the updated studbook on diskette, a list of changes to the studbook, and a list of living specimens by current location. Therefore, although the entire historical studbook may not be published every year, an updated version of the living specimens, deaths, and births, should be produced and be made available on a yearly basis.
It is up to the studbook keeper to decide how and at what time of year to update the studbook. Data acquisition can occur throughout the year as births, deaths, and transfers are reported directly to the studbook keeper, or indirectly through ISIS (but the studbook keeper. should verify data received from ISIS through direct contact with the reporting institutions) Data acquisition may also occur only once a year after questionnaires are sent to and have returned from holding institutions. Questionnaires and/or requests for ARKS taxon reports can be sent at any time of year and, while it is common practice to send questionnaires in the beginning of the year, this is not a requirement. For example, for species that produce young only once each year, data collection after the hatching or birthing season will facilitate the demographic and genetic analyses necessary for population management. For species that produce offspring in winter months, January may not be the best time of year to collect data. Studbook keepers for SSP© species should be prepared to complete a studbook update at least one month prior to any SSP© Master Plan meeting.
To be useful and used, studbooks must be up to date and readily available to animal management staff. Production and distribution of an up-to-date version of the studbook at the same time each year will allow people to begin relying on this information rather than resorting to using raw ISIS data or guesses.
15
A studbook tracks the movement of each specimen for its entire life, starting wifh the data of capture or date of birth (if born in captivity) and ending with the date of death or the date at which the specimen leaves the region or managed population (or is lost-to-follow-up). Ever event and location should be included in the database.
In addition to the date and location of each event, the studbook should also include information on origin, parentage, gender, cause of death (if known), house name, transponder number, band or tattoo number, and any other information that will assist in tracking individual specimens.
Although owners often report information that indicate specimens were transferred via loans, the actual ownership of stubook specimens is not the concern of the studbook keeper and should not be included in the database. Ownership information may be useful to the operation of cooperative management plans, but North American regional studbook keeper shoulds only record physical locations of specimens The terms of individual transfers should be tracked by the institutions themselves (see page 29 on ownership). Each institution is responsible for informing the owner of specimens on loan about management plan recommendations (e.g., transfers).
In general, the taxonomic scope of a studbook is a species, genus, or a group of closely related species or genera with relatively small population sizes. Interspecific hybrids are usually not listed in the studbooks for either species (but see page 20 on analytical studbooks). If the combined living and historical population size is large, separate databases can be created for each subspecies. However, if ali subspecies are included in a single database, each subspecies can be identified with a <USER DEFINED> field (UDF; see page 80 on how to use and create UDFs). Because a UDF allows sorting and selection of specimens for reports and analyses, in most cases it is best to create a single database for all subspecies. For species with uncertain taxonomy (i.e., they might change as more information is accumulated), a single database will accommodate addition or loss of taxa. Combining all subspecies in a single database also facilitates tracking inter-subspecific crosses. Subspecific crosses should be included in the studbook because many populations are managed at the species level; as genetic variation is lost through time, more populations may be managed at this level. If data for different subspecies are kept in separate databases, it will not be possible to manage at the species level without reentering all of the data into a combined
16
subspecies database.
AZA studbooks are approved at either the North American regional or international level. An international studbook should include all specimens of the species held in captivity anywhere in the world. Regional studbooks must include specimens that are being, or have been, held in the North American region and all ancestors of those specimens, regardless of whether they have ever been in the North American region. It is important that every regional and every international studbook trace and record each specimen's pedigree back to wildcaught specimens. This means that a regional studbook keeper does not stop at the regional boundaries when collecting data. The regional studbook keeper should enter all captive ancestors of specimens that have lived or are driving in the region regardless of where those ancestors now live or have ever lived. Complete pedigrees are essential to reliable genetic analyses; incomplete pedigrees hamper genetic- analyses and must be "completed" via hypothetical ancestors and/or assumptions that are often little more than good guesses (see page 19 on assumptions in studbooks).
Zoo records for some species go back into the last century. Although in the early stages of the population, many specimens do not reproduce successfully, their records may be valuable for demographic analyses (i.e., age-specific mortality). Thus, every effort should be made to include as many specimens as possible. However, publication of the studbook should not be delayed during the pursuit of old records for the purpose of historic completeness (see page 13 on when to publish a studbook). Historic records may be added gradually, with successive versions of the entire historical studbook.
Studbook keepers should not limit temporal scope because of perceived, or real, changes in husbandry or management. No specimen should be excluded from a studbook because it lived under "outmoded" management and husbandry regimes. SPARKS provides many ways to select subsets of the database; identification of the most appropriate subset for each demographic and genetic analysis is accomplished through discussion and trial analyses by the population manager and/or the SPMAG advisor (see page 26 on views in SPARKS).
There are a few cases in which entering all specimens within the scope given above may be unrealistic and represent an inordinate amount of work without a corresponding
17
benefit. These cases are rare and should be determined by the appropriate Taxon Advisory Group (TAG), in consultation with the SPMAG advisor and the studbook keeper.
Any and ail institutions that can be verified as holding or having held specimens should be included in the studbook. Inclusion of data from an institution should not be contingent on whether it is a member or affiliate of AZA; if an institution provides data to the studbook keeper, it should be included in the studbook, provided it can be verified. When a reliable source reports a specimen's transfer to an institution that does not respond to the studbook keepers requests for information, the transfer should be accepted and the specimen should be denoted as lost-to-follow-up at the receiving institution (see page 74 on how to enter lost-to-follow-up).
18
Although a studbook should be the best representation of the true history of the population, most studbooks will contain a few minor assumptions and/or conventions. Most minor assumptions are guided by information provided by reliable sources; they fill the gap between partial information, hints, innuendos, and suspicions on the part of individuals and institutions reporting to the studbook keeper. Assumptions are usually specific to a single specimen, or to group of specimens held at the same institution. For example, estimates of age to particular calendar years (e.g., 1993 or 1994) would usually be considered minor assumptions. It is difficult to generalize about what might be considered a minor assumption. In most cases, inclusion of minor assumptions in the studbook is dependent on the studbook keeper's good judgement. Any assumption should be documented in the Special Data for that specimen
Conventions are rules developed by the studbook keeper for consistent application of minor assumptions. Conventions are usually based on the biology or husbandry of the studbook species; they are usualiy applied to the population at large. Most conventions combine common sense, logic, and biology into a simple rule to be used when data are unknown or unclear. One common exernple is a species that breeds once per year and can only be imported from the wild as juveniles (e.g., other than 6 months, younger than 18 months). A convention might be developed to set the birth date for alt wild-caught specimens, with unknown birth dates, as January (estimated to month) of the previous year. Another example might deal with specimens of unknown origin (wild-caught or captive-born) that enter the population before the first reported captive birth. A convention could be established that all specimens entering the population before the first reported captive birth are wild-caught. Conventions should be outlined in the Species Notes, and noted in the Special Data whenever they are applied to a given specimen.
One of the most common types of conventions is the method for back dating births from the date young are first observed or heard. Marsupials, bears, cavity nesting birds, and other species whose young are not likely to be noticed until some time after birth, will have conventions for estimating date of birth. These should be carefully documented in the Species Notes and Special Data.
It is essential that all assumptions and conventions be documented in the SPARKS database; conventions should be discussed with the SPMAG advisor before they are applied to the studbook or analytical database.
19
A studbook is the best possible representation of the truth as it is currently known. Each studbook must contain an unambiguous representation of what is known and what is not known (see page 19 on assumptions and page 36 on what does "unknown" mean?) about every specimen in the population. Although studbooks may contain minor assumptions or estimates, such as birth dates estimated to the month or year, information that is truly unknown should be entered as {UNKNOWN}. Thus, while a studbook contains only the truth, an analytical studbook contains assumptions, half-truths, and necessary fiction.
Analytical studbooks are constructed to facilitate population management. They may contain significant assumptions about variables such as dates of birth, transfer, death (but see page 74 on how to enter lost-to-follow-up), parentage, and origin (e.g., captive-born or wild-caught). In some cases, analytical studbooks contain hypothetical specimens (ancestors) created to complete the pedigree and facilitate genetic analyses. In effect, an analytical studbook "fills in" what the studbook keeper doesn't know and can’t find out. This "filling" in is essential to some demographic and genetic analyses.
Formulation of an SSP© Master Plan or PMP often requires that assumptions be made about some or all of the unknown information in the studbook. Many studbooks have specimens for which some combination of sire, dam, key dates, and origin is unknown. Demographic and genetic analyses have limited ability to deal with specimens whose vital information is unknown or incomplete Typically, these analyses either exclude such specimens or make assumptions about all specimens in a category. For example, GENES can treat all specimens whose origin is unknown as wild-caught or it can exclude them from all analyses; there is no middle ground; In some populations, one or more specimens or lineages with unknown or incomplete information may be critical to the survival or management of the population. To incorporate those specimens and lineages into the management plan, assumptions must be made to substitute usable data for unknown information. These assumptions are recorded in analytical studbooks. All assumptions should be documented in a <SPECIAL DATA> field (see page 79 on special data).
Analytical studbooks should be constructed whenever management decisions will be based on assumptions about demographic or genetic information. For SSP© Master Plans, these assumptions are discussed among the management group, SPMAG advisor, and studbook keeper before being entered in an analytical studbook or used in any analyses. Some populations may require a new or revised analytical studbook for every SSP© Master Plan or PMP. Each plan should have its respective analytical studbook clearly identified so that anyone can check on the results or recommendations.
20
The easiest way to construct an analytical studbook is to copy the true studbook into another SPARKS directory, then edit the new studbook. Analytical studbooks must be named starting with "XX" (e.g., XXOKAPI, XXHIPPO) and all deviations from the true studbook must be well documented in Special Data.
Hypothetical ancestors are often created to deal with unknown or uncertain pedigrees. These hypothetical specimens are usually the antecedents of living specimens that originated at the same institution, but whose parents cannot be identified. For example, if several living specimens were born at the same institution over a short period of time, but no potential parents were known to be, at that institution, hypothetical parents, usually with {WILD} parents, might be contrived to complete the pedigree. Similar situations might occur with specimens with one or bath paints listed as {MULT}.
It is essential that hypothetical ancestors, or any "major fiction" about parentage, only be present in an analytical studbook. For example, hypothetical ancestors created to make pedigree links between specimens suspected of having some degree of relationship will mislead anyone (including the studbook:keeper) who looks at the studbook for real information about specimens in the population. Hypothetical specimens, altered parentage, sex changes, bringing back the dead, or the like, should only appear in an analytical studbook created for such testing of alternative realities.
21