There are numerous ways to define and represent an ecosystem, and the choice of approach used, when studying an ecosystem, depends on a complex of reasons. Foremost among these are the availability of (1) the input data required for the system to be studied, (2) other ecosystems previously studied with the same approach, and (3) the background and skills of the would-be modeler(s).
Given the data in its DIET COMPOSITION, PREDATORS, FOOD ITEMS and FOOD CONSUMPTION tables, FishBase has amongst its clients the users of the Ecopath approach and software (Christensen and Pauly 1992, with updates in www.ecopath.org). Indeed, these clients are the main reason why a routine was developed which, for any selected country and/or ecosystem (type), assembles and exports a file with information on the fish occurring in the country and/or ecosystem in question, culled from the above and other, related FishBase tables.
As it turns out, this output, judiciously linked, can be used to generate directly, without analysis by Ecopath, a simple representation of the trophic structure of an ecosystem, i.e., a ‘Lindeman pyramid’ (see Lindeman 1942), as commonly used to summarize food web information. Thus, once the ecosystem or part of a country to be modeled has been chosen, the routine in question does the following:
For each species, extract the trophic level (‘troph’) from the ECOLOGY table, or if none is available, from the ESTIMATES table (see Box 26 for a brief description of this new, very special table);
Group all species of fish in classes of ½ troph, from the first class (2.00-2.49) to 5.00+ (i.e., including all values higher than 5);
For all species with entries in the DIET COMPOSITION or FOOD ITEMS tables, group the food items by their troph, i.e., the same classes of ½ troph as in (2), using the default trophs for preys in the FOOD TROPHS table;
For each non-fish vertebrate species with an entry in the PREDATORS table, include the predator in the appropriate ½ troph class (note that the PREDATORS table now includes fields for the troph of predators and their size, with defaults for the former being provided in the FOOD TROPHS table);
For each ½ troph class, compute mean body size (
, with standard error, s.e.) of the groups included therein, from `
L »
1/3 Lmax, using Lmax values from the Species and/or the ESTIMATES table for fishes, and from the FOOD ITEMS table or the PREDATORS table for invertebrate preys, and non-fish vertebrates, respectively; Output a list of the fish, of the invertebrates, of the plants/detritus, or of the non-fish vertebrates, by troph class, along with their mean size and related statistics, when the corresponding element of the graphic representation of a Lindeman pyramid is double-clicked on.
This approach, which summarizes data from various FishBase tables, thus relies on a classic of the ecology literature for its metaphor, i.e., on a pyramid whose steps represent different trophic levels and the species (and/or functional groups) therein.
We intend to improve the routine underlying this pyramid along two lines: By identifying new ecological inferences that can be drawn from ecosystem-specific lists of taxa arranged by trophic levels and additional data extracted from FishBase, e.g., by considering the food consumption and transfer efficiencies of the species involved; and
By establishing further links, still to be developed, between Lindeman pyramids and full-fledged Ecopath models.
You get tot the TROPHIC PYRAMIDS routine from the FishBase Main Menu by selecting the Reports, Graphs, Trophic Ecology and Trophic Pyramids buttons.
On the Internet, you get to the TROPHIC PYRAMIDS routine by selecting the respective radio button in the ‘Information by Topic’ section of the ‘Search FishBase’ page.
Christensen, V. and D. Pauly. 1992. The ECOPATH II - a software for balancing steady-state ecosystem models and calculating network characteristics. Ecol. Modelling 61:169-185 [see
www.ecopath.org for updates].Lindeman, R.L. 1942. The trophic-dynamic aspect of ecology. Ecology 23(4):399-418.
Daniel Pauly, Rainer Froese and Ma. Josephine France D. Rius