|Environment and development
in coastal regions and in small islands
Coastal region and small island papers 3
The Data Management Centre and Data Summary
Dulcie M. Linton and Jeremy D. Woodley
Centre for Marine Sciences, University of the West Indies-Mona, Kingston 7, Jamaica
The CARICOMP Data Management Centre is responsible for the collation, storage, and redistribution of data. At each CARICOMP site, data are entered into spreadsheet templates prepared for each of the fourteen CARICOMP Level One methods. These are sent to the Centre by various means, increasingly by e-mail. Regional data summaries are distributed to the network each year. After a period of internal review and use, these will be published, both in hard copy and on the World Wide Web. Final storage will be in a relational database program. Meanwhile, the Centre has coordinated Caribbean-wide research and discussion within the network.
The Caribbean Coastal Marine Productivity Network is a cooperative effort to study and monitor three productive Caribbean coastal habitats, namely mangroves, seagrasses, and coral reefs. Twenty-one marine laboratories, parks, and reserves in nineteen countries are collecting data according to prescribed methods given in the CARICOMP Methods Manual, Level 1 (1994, unpublished). These data are sent to the Data Management Centre (DMC) for collation, storage, and redistribution. The DMC was established in 1992, with grant funds from UNESCO, at the Centre for Marine Sciences, University of the West Indies (Mona), Kingston, Jamaica. The Centre became fully operational when a full-time Data Manager was employed, thanks to grant support from the MacArthur Foundation, in November 1994. The agreed functions of the DMC are:
Receipt and Storage of CARICOMP Data
Data Entry. At each site, data are collected using fourteen methods, described in the CARICOMP manual and summarized in the first CARICOMP data report, which appears in this chapter. In 1992-93, raw data were sent to the DMC and entered into a specially written database program. In 1994, spreadsheet templates were designed (in Quattro Pro) for each of the above methods and distributed on disk to all participating sites for entry of their data.
The use of spreadsheets has several advantages (Ledgister et al., 1995, 1997):
However, if Site Directors are unable to enter their data promptly, they are encouraged to send their raw data sheets to the DMC for entry.
In addition to the basic CARICOMP methods, data are being collected and stored electronically by automatic recording equipment: weather stations and Hobo temperature recorders.
Data Transfer. Completed spreadsheets have been returned to the DMC by courier, by mail or, increasingly, by e-mail. By October 1996, 19 of the 21 active CARICOMP sites had access to the Internet. So far, six have successfully transferred data to the DMC by e-mail, while others have yet to master the appropriate software.
Data Storage. Datasets received by the DMC are examined to ensure that the correct template has been used; they are then checked for obvious data entry errors. For HoboTemp data, which are presented in hundreds of rows, the DMC has designed macros (commands written in a software program that automates repetitive procedures) to speed up processing. All data received to date are stored in Quattro Pro spreadsheets on a 1.0 GB hard disk at DMC. Periodic backups are made to 120 MB data cartridges and stored at both on-site and off-site locations. Also, the original diskettes sent to the DMC are retained.
The entire CARICOMP database is now stored in hundreds of Quattro Pro spreadsheet files. This arrangement is part of a two-phase data management strategy of the DMC: to use spreadsheets at the data entry phase, with subsequent storage in a database management system. This strategy allows flexibility in data handling at the entry level, where the data are readily summarized and transferred to the CARICOMP network. Then the spreadsheets are imported into a relational database for ultimate storage in a more robust and versatile programme (Ledgister et al., 1997).
The initial and most crucial stage of database development has been completed: sorting the data and designing data tables, in which the data are normalized (reduced to the simplest structure possible). The next phase, designing the data tables in an appropriate database, will take place after the pros and cons of various relational database programs have been fully explored. Meanwhile, a converter program has been written for the DMC, by which data will be extracted from the spreadsheets and entered into the database program.
Preparation and Distribution of Data Summaries. The schedule originally proposed was that the sites should send data every month to the DMC, which would return regional summaries to the network every quarter. This schedule has not been maintained, largely because of the irregular flow of data from the sites. All sites received copies of the 1993 dataset in January 1994, and annual summaries to date were mailed to all contributing sites in September 1995. Furthermore, copies were distributed at the Site Directors Meeting in the Dominican Republic in December 1995. Abstracts of papers presented by CARICOMP at the 8th International Coral Reef Symposium (ICRS), held in Panama in June 1996, were drafted at that meeting, using the summaries provided.
Annual Data Reports. Site Directors agreed that their data should not be published until they had the opportunity to work with the data themselves. Therefore, the report for 1992-1995 in the following pages is the first to appear. The data herein will be made available on the CARICOMP home page on the World Wide Web, which was established in November 1997.
Analyses and Other Summaries. Specific datasets were distributed to members of working groups assigned to prepare the CARICOMP papers for presentation at the 8th ICRS. Those papers are listed in the 1992-95 report.
Communication Between CARICOMP Sites. All 25 participating sites are linked to the DMC by the traditional means of communication: telephone, fax, and mail service; at least 20 now use electronic mail, as do all members of the Steering Committee. No system is completely reliable, but the use of e-mail has made network communication much quicker and easier than even by fax.
In 1995-1996, the DMC organized a Caribbean-wide survey of coral bleaching and subsequent mortality, which was reported at the 8th ICRS. The DMC also played a role in the Caribbean-wide survey of a seafan sickness, the subject of another paper at the 8th ICRS. Reports of other mortality events, notably at the CARICOMP coral reef site in Moroccoy, Venezuela, were also distributed. In addition the DMC has facilitated discussions on CARICOMP methods and on future CARICOMP programs.
It is expected that the DMC will also be a focal point for communication with other individuals and agencies. So far, most of the enquiries received, especially from institutions wishing to join CARICOMP, have been passed to the Steering Committee for decision. Additional information is available to all interested persons at the CARICOMP website, which is maintained at the University of the West IndiesMona, in Kingston, Jamaica: http://www.uwimona.edu.jm/centres/cms/caricomp
Cintrón, G., Y. Schaeffer-Novelli. 1984. Methods for studying mangrove structure. In: The Mangrove Ecosystem (edited by S. C. Snedaker and J. G. Snedaker), pp 91-113. UNESCO, 251 pp.
Golley, F., H. T. Odum, R. F. Wilson. 1962. The structure and metabolism of a Puerto Rican red mangrove forest in May. Ecology, 43(1):9-19.
Ledgister, R., D. M. Linton, J. D. Woodley. 1995. Data Collection and Management in CARICOMP, A Regional Programme of Long-Term Ecological Research. Presented at the European Symposium of the International Society for Reef Studies (Newcastle UK, September 1995).
Ledgister, R., D. M. Linton, J. D. Woodley. 1997. Use of spreadsheet templates for general data entry and their integration into the database system. In: Proceedings of the 8th International Coral Reef Symposium (Panama, June 1996) (edited by H. A. Lessios and I. G. Macintyre), Vol. II, pp 1565-1568. Smithsonian Tropical Research Institute, Balboa, Republic of Panama.