December 23, 1997
Submitted by J. Frederick Grassle, Institute of Marine and Coastal Sciences
Rutgers University, New Brunswick, NJ 08901-8521
On October 30, 1997, a "Census of the Benthos Workshop" was held to explore innovative approaches to estimation of the abundance and distribution of species in marine benthic habitats. This Workshop was part of a series "To explore the value, timeliness, and feasibility of stimulating, designing, and organizing a period of intense, comprehensive oceanic observation whose purpose would be to assess and explain the global distribution of marine life" (Ausubel 1997). Participants included:
Participants unanimously agreed on an initial phase of work to summarize major advances in marine systematics through a series of maps. There is particular interest in species ranges, species/area relationships, and maps of habitat. For many of the major taxa, species distribution data are being summarized into a number of electronic data bases. In the proposed workplan, the distribution of species richness and zoogeographic boundaries would be estimated to produce a series of on-line electronic maps of better-known abundant benthic taxa. In addition, overlays of environmental data such as: bottom topography, sediment type, kinetic energy, climatic oscillations, nutrient flux, and dissolved oxygen concentration would be developed. Surface productivity and benthic biomass would also be plotted. For each set of data, maps of sampling effort would also be needed. New analytical approaches would be explored using the GIS data bases to test ecological and zoogeographic hypotheses (similarities and differences in boundaries and ratios of diversity across taxa, species/area relationships, local vs. regional species richness, etc.). These analyses would provide a guide for the initial observational phase of the Benthos Census. In this program those most knowledgeable about marine species would be fully involved in the task of identifying sampling gaps, setting standards for future studies, and developing new hypotheses. Systematics and species-level information would play a central role in the design of future studies, thus revitalizing this area of marine science. Those present were willing to redirect their present efforts to the pilot program on the assumption that a significant breakthrough in marine biogeography could be achieved in approximately one year.
A parallel effort on design of ocean observing systems would use the existing information to design more efficient approaches to improving the overall data base.
For most major interdisciplinary marine science programs, marine life is commonly measured as density and/or biomass of major taxa. Reliable studies of species composition are less common. There is general agreement that the species is the basic unit for studying ecological and evolutionary relationships. Our ability to understand biological relationships, or measure change in biological characteristics with respect to space or time, is inhibited by inadequate attention to the accurate identification of species. Even at relatively well-studied sites, inaccurate species identifications have led to incorrect conclusions. Biomass and density are generally a function of carbon flux to the bottom but there are few reliable predictors of species composition or species diversity. Numerous theories have been proposed to explain species diversity patterns, but the lack of appropriate geographic frameworks to summarize data and guide sampling have slowed progress.
The work of describing new species has proceeded even though research funding for marine systematics was almost nonexistent in the 1970s and 1980s. The only map of species richness per unit area (Valentine 1974, Scientific American) recognizes 12 shallow water geographic areas and 6 undefined levels of relative species richness. Based on presentations at the Workshop, individual investigators are making remarkable contributions to increasing the numbers of known species in several taxa. Community ecologists, working on species-rich deep-sea and continental-shelf sedimentary environments, and coral reefs, and with the unique communities at hydrothermal vents, indicated that vast numbers of species remain to be described from these environments. The numbers of species in the most species-rich marine environments have been grossly underestimated.
Statisticians and quantitative ecologists at the Workshop developed several principles for a survey of species diversity. These principles draw heavily on the experience of terrestrial ecologists:
Log normal and log series distributions are useful in analyzing species richness data. Sampling should be driven by testable hypotheses rather than gaps in a sampling scheme.
Techniques are now available to provide the necessary map coverage needed to guide an intense, comprehensive benthic observation program. Maps, often taken for granted on land, are needed to divide the marine environment into regions providing greater resolution than latitude/longitude and water depth. Several relatively inexpensive methods are available to provide continuous coverage over large areas:
On much smaller scales, long-term ecosystem observatories such as LEO-15 (von Alt and Grassle 1992 and von Alt et al. 1997) can provide a much greater density of information on marine habitats and physical, chemical, and geological processes. In addition, they provide information on the properties and impacts of low-frequency events. For soft sediments, a high priority should be given to developing acoustic techniques to map the distribution of animals that produce structures such as molluscan shells and worm tubes. Autonomous Underwater Vehicles can be equipped with a high resolution, sub-bottom acoustic imaging system to count and classify tube-inhabiting worms and shell-bearing clams, snails, and crustaceans (T. Austin pers. comm.). This information might be used to map recurrent habitat, provide a better basis for within-habitat species-area relationships, and provide quantitative counts of some taxa.