by
Bjerne Ditlevsen
Danish Forest Service
Strandvejen 863
DK-2930 Klampenborg
Denmark
INTRODUCTION
The continuous accumulation of information in Forest Tree Improvement (FTI) programmes will by use of non-computerized information systems result in increasing difficulties in the utilization of the collected information. It is therefore desirable to set up computerized information storage and retrieval systems that can provide the required information quickly without laborious file searching.
At the Hørsholm Arboretum in Denmark where forest tree improvement on Norway spruce, sitka spruce, larch and contorta pine has been carried out for several years the need for an efficient information system has become evident, and in 1974 the development of a computerized FTI information system was initiated. In 1976–77 the system was implemented on an IBM 370/165 computer, and the major part of the “old”, already existing, FTI information from the Arboretum has been transferred to the system; the transfer of “old” data still continues, along with the loading of “new” information.
The need for an efficient information system is even more pronounced in the other Nordic countries, as the amount of accumulated information in these countries is far in excess of that in Denmark. In the light of this fact both Norway, Sweden and Finland have shown great interest in the development of the Danish system, and at the beginning of 1979 a joint Nordic research project on the development and coordination of Nordic FTI information systems was initiated. The Danish system, as described in the following, forms the basis for the work in this research project.
FOREST TREE IMPROVEMENT INFORMATION
The information system is designed to serve not only as a local system to process data from the Arboretum, but also as a system that can process data from a number of different research stations and even from different countries. The system is further designed to process information from different types of improvement programmes (e.g. provenance programmes, breeding programmes and programmes involving vegetative propagation).
Figure 1. Major activities that can be covered by the Danish/Nordic FTI system.
Activities which can be covered by the system as it appears today are given in Figure 1. The information from this network of activities can be grouped as follows:
Experimental information. The data from different experiments and tests are usually analysed by appropriate statistical analyses. At the Arboretum the statistical analyses are carried out in a separate “Statistical Analyses System”, and only the “key-data”, i.e. the final results from the analyses, will be put into the information system.
General descriptions of source material and selected material, e.g. description of provenances and plus trees.
General descriptions of breeding material, e.g. pollen, seed, cuttings and grafts.
General descriptions of the localities that are used in the improvement programmes, e.g. experimental sites, seed orchard sites, clonal archives and localities for seed or pollen storage.
General descriptions of the different methods or procedures that are used in the improvement programmes, e.g. procedure of seed storage; method of grafting, cutting or sowing.
Descriptions of the relatedness of different individuals or groups of individuals. The system must be able to describe the relationship from progeny to parents, grandparents, etc. and vice versa.
Descriptions of the inter-relations between the material (individuals or groups of individuals) and the localities. The system must be able, for any material, to provide a list of the localities where this material is found. Also it should be possible, for any locality, to provide a list of the material found in this specific locality.
Descriptions of the inter-relations between time and “activities” in the improvement programmes, e.g. the time of describing a plus tree, the time of grafting a plus tree, or the time of measuring an experiment.
A brief description of the computer system as the “tool” that handles the information will be given below. The main function of the system is illustrated by the flowchart in Figure 2. The system is more fully described, in Danish, by Ditlevsen (1977) 1.
DATA REGISTRATION
As the collecting of data will be done by different persons at different places and at different times, it is extremely important to have a guide that gives an unambiguous description of the information to be collected.
For this purpose a registration manual has been developed. The manual serves the following two main purposes:
It gives a complete definition and description of each data element that can be registered. Each description is composed of:
Data name;
Definition and description of the contents of the data name.
Description of the data type (numeric, alphameric or alphanumeric), and the length of the data field;
In case the information is coded, the possible codes are defined.
In this way the manual serves as a complete thesaurus for the system, and assures that each data element will be registered correctly and uniformly.
Figure 2. Flowchart for the FTI information system.
The registrations are carried out on 59 different registration forms, each identified by a form-code. The form-codes are preprinted in the first fields of each registration form. The form-code is always succeeded by an accession-code (material-, locality- or method-code), which the information on the remaining part of the registration form refers to. Figure 3 illustrates the principles of the structure of the registration forms.
| Form-code | Material-, | |
| Locality- | ||
| or | ||
| Method-code | ||
| 59 | V 1124 |
Figure 3. Illustration of the structure of the registration forms.
In each case the manual indicates which form or forms that should be used and further gives instructions on how to fill in each form, ensuring that the same registration forms are used in similar situations and that the forms are filled in correctly. The forms used for the registration have a fixed length of 80 characters, corresponding to a 80-column punched card, and as far as possible the data elements on each form constitute a logical unit that by normal practice will be registered at the same time. As the registration forms are identical to the records in the data bank (see below) the manual also gives a complete outline of the data structure in the data bank.
UPDATING THE DATA BANK
After registration, the data are transferred to punched cards or are entered directly on a disk via a computer terminal.
Before updating the data bank the data are checked for errors by a special programme. This error checking is done in two ways:
The programme checks all data fields on the records to find, for instance, illegally used codes, measurements that exceed certain limits, or missing data. This error checking is not complete, but it helps the researcher to detect severe errors such as illegal codes, missing values, etc.; a more detailed check has to be done manually.
The programme helps the researcher to check the new accession codes (material-, locality-or method-codes) by looking up these codes in the data bank to see if they already exist and if so, in which files they can be found. This extra check has proved to be very efficient in detecting erroneous accession codes.
Errors detected by the error programme or by a manual check are mostly corrected on the punched cards before the data bank is updated, but the development of a new programme for corrections of data in the data bank has made it very easy and quick to correct errors also after the updating.
New data are merged into the existing data bank by means of a special updating programme.
The error checking and the updating as described above can take place concurrently with registrations in the field, and thus the researchers can work independently of each other.
THE DATA BANK
All information in the system is stored in a data bank, from which any information needed can be retrieved.
The data bank, as it appears today, consists of the following 7 main-files 1:
UDV-file, which contains general descriptions of source material (stands) and selected material (stands and single trees). The material in question is identified by a material-code.
FORM-file, which contains general descriptions of breeding material (pollen, seed, cuttings, grafts). The file also contains information on the breeding methods that are used (seed handling, method of grafting or cutting, etc.). The material in question is identified by a material-code.
NGL-file, which contains “key-data” on the material in the system, i.e. the final results from the statistical analysis of experimental data (from laboratory tests, nursery tests, field experiments, etc.).
ADM-file, which keeps a list of all material-codes and contains information on relatedness and on propagation. From this file it is possible to list a complete pedigree for any material in the system.
LOK-file, which contains general descriptions of all localities used in the FTI programmes. For experimental sites information on design, measurement plans, etc. is included. Each locality is identified by a locality-code.
DIST-file, which contains information on the material-locality relations. Thus for any material the file will contain a list of the various localities where this material is found, and vice versa, for any locality the file will contain a list of material in this locality.
MET-file, which contains descriptions of the different methods of procedures that are used in the FTI programmes e.g. procedures for seed or pollen storage, methods used for inducing early flowering, etc. Each method or procedure is identified by a method-code.
To enable efficient and flexible utilization of the data bank, the data is stored in two ways:
On magnetic tapes, where the records are organized sequentially, i.e. the files can only be processed from one end to the other. This version of the data bank is used frequently to provide listings; concurrently with processing this data bank, information from the direct access data bank (see point B) can be picked out and incorporated into the information to be listed, which gives a great flexibility in the utilization of the information.
On disks with direct access to each record. As mentioned earlier the records are identical to the registration forms (see Figure 3). The material-, locality- and method-codes serve as accession keys to the different files.
The files in the data bank are accessed and cross-referenced as described in the following (see also Figure 4).
The UDV-file is accessed by a material-code. The file is cross-referenced to the ADM-, NGL- and LOK-files.
The FORM-file is accessed by a material-code. The file is cross-referenced to the ADM-, NGL-, DIST-, LOK- and MET-files.
The NGL-file is accessed by a material-code. The file is cross-referenced to the ADM-, FORM-, UDV- and DIST-files.
The ADM-file is accessed by a material-code. The file is cross-referenced to the FORM-, UDV-, NGL- and DIST-files. In addition the ADM-file keeps material-codes as pointers to related material. These pointers can be used to reaccess the ADM-file.
The LOKfile is accessed by a locality-code. The file is cross-referenced to the DIST-file.
The DIST-file can be accessed by either a material-code or a locality-code. The file is cross-referenced to the ADM-, FORM-, NGL- and LOK-files.
The MET-file is accessed by a method-code. This code cannot be used to access any other files.
Figure 4. Illustration of the linkage between the 7 files in the direct access data bank.
To illustrate the movements in the data bank the following example may be used:
If you want to have a description of a certain plus tree including a list of progeny, and in addition a list of the different localities where the progeny is found, one way of collecting the required information could be: Enter the UDV-file where you get a general description of the plus tree. From there you go to the ADM-file to get a list of progeny, and for each progeny found in the ADM-file you enter the DIST-file to have a list of the localities where the progeny is found.
As can be seen this data bank organization yields a flexible and powerful tool for the utilization of the stored information.
UTILIZATION OF THE SYSTEM
As the system has been designed to be flexible in regard to its utilization and as information from several FTI programmes can be stored in the same data bank, the possibilities of the utilization of the system are numerous.
Examples of the three main uses of the system are given below to illustrate some of its possibilities:
To provide listings for internal use
The most important function of the system is to provide different kinds of up-to-date listings. These lists usually contain detailed information used by the researchers in their daily work. As examples can be mentioned provenance lists, plus tree lists, lists of field experiments, seed orchards, etc. Other types of lists are produced mainly for administrative purposes, as for instance lists of pollen or seed in store, or summary lists of all tree improvement activities carried out in a certain year.
To provide listings for external use
One of the purposes for establishing a computerized information system is to facilitate exchange of information between researchers, thus facilitating coordination and cooperation. The type of listings under point (1) are not always suitable for external use, partly because of the language used and partly because of the very detailed information resulting in too voluminous lists. In addition, the lists for external use are usually requested to meet certain specific criteria or combinations of criteria. To give an example of external use, a Swedish tree breeder might request a list of Danish plus trees of Norway spruce, Istebna origin, that have been selected for high wood density. Listings might also be requested by state authorities or by international organizations such as FAO or IUFRO, and in these cases the lists should preferably be printed in English.
Interrogations
In addition to the possibilities of providing different kinds of listings, it is very important to have a powerful and flexible inquiry system.
The system can provide answers to specific enquiries in two ways:
By means of programmes providing listings of information which meet specific criteria. As mentioned earlier under point (2) this procedure is used frequently to meet external requests.
By means of an inter-active enquiry programme, through which the researcher can “communicate” directly with the data bank and thus ask questions, get answers, ask questions, etc.
In both cases very complex questions can be asked, but especially the inter-active enquiry programme is a powerful tool that gives the researcher almost unlimited possibilities of formulating questions to obtain exactly the information he wants.
CONCLUDING REMARKS
The system briefly described above is currently being developed to satisfy the demands of forest tree breeders and, as already mentioned, its future development will be carried out within the framework of a joint Nordic project.
Even though the system is primarily developed for forest tree breeders it can easily be used to process information from other related fields, such as conservation of gene resources, registration of trees in arboreta, parks, etc. Although the information from such related fields will be mixed with FTI information in the data bank, it can be processed separately thus enabling researchers using just parts of the full system to work independently of the forest tree breeders.
