4.4.4 — Analysis of point data

Geographic point data are defined simply by their geographic latitude and longitude. These co-ordinates can be stored in any database, and imported easily into a GIS for further analysis. Besides direct observations at certain sites, point data sets are generated by ringing recoveries, museum data (Mearns & Mearns 1998) and satellite tracking. Some of the major ringing databases contain more than a million data points, and only a small fraction of these data has been used to produce maps. In the future, these scattered data sets will hopefully be available for GIS analysis. Even extracts from these data sets, such as presence/absence data, would facilitate the production of maps and analysis in the way outlined above (cf. Riede 2001).

A new, rich source of high-quality data is provided by satellite tracking of migratory species (e.g. Øien & Aarvak 2001). These studies have produced surprising insights about migratory pathways and hitherto unknown staging areas of highly endangered species, such as the lesser white-fronted goose (Anser erythropus, Figure A2.28).⁴⁸ Satellite data can be considered a time-coded sequence of point data, and therefore analysed in the same way as other point data.

A wealth of information comes from species observations at well-defined sites. Such observation points include a wide variety of habitats where migratory species aggregate, such as bat caves, turtle nesting beaches or Important Bird Areas (IBAs), as identified by BirdLife International (http://www.birdlife.org). Wetlands International maintains an impressive database of regular waterbird counts at certain key sites worldwide (http://www.wetlands.agro.nl/), some of which have been added as point data to GROMS (see section 3.4). Once integrated, the species number at each point was calculated and mapped (Figure A2.39). In a second step, it was superimposed with a map of wetlands protected under the Ramsar Convention and regional threats, such as wind parks, oil spills and major industrial developments in the Wadden Sea area (Figure A2.40). The map shows the uneven distribution of protected wetlands, and illustrates the vicinity of human-induced disturbance to important staging areas. In fact, decision makers seem to prefer protected areas for their development plans (see Table 4.13).

Nesting beaches of marine turtles or bat caves are further examples of point data, which can be mapped and analysed in a similar way. The "Marine Turtle Nesting Database"⁴⁹ contains the present distribution of nesting beaches for the Indopacific at a high resolution, and is one of the few GIS data sets which are fully available on the World Wide Web, both for online map display and for download. Figures A2.77-A2.83 show composite maps of these nesting beaches with the general distribution as adapted from Iverson (1992). Figure A2.82 shows both layers at a higher scale, illustrating the difference between the highly precise point data and the general distribution, which was adapted from a low-scale map based on museum data (Iverson 1992). The obvious differences should be interpreted with care, but seem to indicate former — or overlooked — nesting sites on the eastern coast of India, around the southern tip of Malaysia (for details, see caption of Figure A2.82).

In summary, the GROMS geodatabase design even makes it possible to administer different map sources, because every combination of species and territory has its unique reference, referring to the map or country list used. Comparison of these different information sources will permit discovery of knowledge gaps, contradictions, or local extinctions. It is hoped that it will be a useful tool for member states preparing reports on the progress of conservation of "their" migratory species.

This document should be quoted as part of the publication "Riede, K. (2001): The Global Register of Migratory Species ­ Database, GIS Maps and Threat Analysis. Münster (Landwirtschaftsverlag), 400 pp." + CD

 by Klaus Riede