Design of microcosms depends on the nature of the experiment to be conducted and requires a number of straightforward decisions about materials, size and shape of container, energy inputs, and biota. The combination of these elements into a useable configuration is the design challenge. Although there are good reasons to standardize design for some purposes, the literature is filled with unique and ingenious microcosms that demonstrate a wide creativity for this subdiscipline of ecological engineering. General design principles for microcosms are covered by Adey and Love-land (1998) and Beyers (1964). Design of aquatic microcosms historically derives in part from the commercial aquarium hobby trade (Rehbock, 1980) and aquarium magazines can be a source of inspiration about possible microcosm designs. Terrestrial microcosms, on the other hand, seem less related to terrariums in terms of design. As with all constructed systems, cost is an important constraint on microcosm design. Cost is often proportional to size and number of replicates, and must include both construction (capital) and operation figures.
The primary challenge of microcosm design is physical scaling, in terms of both time and space (Adey and Loveland, 1998; Dudzik et al., 1979; Perez, 1995; Petersen et al., 1999). Scaling of hydraulic models in civil engineering is well developed (Hughes, 1993) and may be a guide to designers of ecological microcosms. The appreciation of scale as a fundamental consideration in ecology has been recognized only in the past 20 years (Gardner et al., 2001; Levin, 1992; H.T. Odum, 1996; O'Neill and King, 1998; Peterson and Parker, 1998; Schneider, 2001), though Hutchinson (1971) mentioned the subject much earlier. The basic way to portray scale is with a "Stommel diagram" where different systems are plotted on a graph with axes of time and space (Stommel, 1963). Figure 4.1 is this type of diagram, showing the relative scale of microcosms and mesocosms in relation to natural ecosystems. Figure 2.14 is another variation of a scale diagram, in this case for biota (see also the related early graph given by Smith, 1954). Scale is a somewhat abstract concept that is still being explored theoretically and empirically. As noted by O'Neill (1989):
Scale refers to physical dimensions of observed entities and phenomena. Scale is recorded as a quantity and involves (or at least implies) measurement and measurement units. Things, objects, processes, and events can be characterized and distinguished from others by their scale, such as the size of an object or the frequency of a process ... Scale is not a thing. Scale is the physical dimensions of a thing.
Scale also refers to the scale of observation, the temporal and spatial dimensions at which and over which phenomena are observed . The scale of observation is a fundamental determinant of our descriptions and explanations of the natural world.
Scale is an important concept because ecosystems contain components and processes that exist at different scales and because the ability to understand and predict environmental systems depends on recognizing the appropriate scalar context. For example, a forest may adapt to disturbances such as fire or hurricane winds, and to understand the ecosystem it must be recognized that the fire or hurricane is as much a part of the system as are the trees or the soil, even though the disturbance may occur only briefly once every quarter century. Obviously, microcosms often (though not always) are smaller scale than real ecosystems. This is an intentional sacrifice to provide for the benefits or conveniences of experimentation: ease of manipulation, control over variables, replication, etc. However, the reduction in scale affects the kind of ecosystem that develops in the microcosm and, according to some, limits the ability to extrapolate results (Carpenter, 1996).
Microcosm scaling issues fall into two broad categories that can be difficult to separate: fundamental scaling effects and artifacts of enclosure (Petersen et al., 1997, 1999). Fundamental scaling effects are those that apply in natural ecosystems as well as microcosms. These are primarily issues of sizing and temporal detail. In terms of sizing, perhaps the most often cited example is the work of Perez et al. (1977) in designing small-scale microcosms to model the open water ecosystem of Narragansett Bay, RI. Their design consisted of replicate plastic containers with 150 l of seawater from the bay. Paddles driven by an electric motor provided turbulence and fluorescent lamps provided light, timed to a diurnal cycle. A plastic box of bottom sediment from the bay was suspended in the containers to represent the benthic component of the system. Scaling was done to match Nararagansett Bay for surface-to-volume ratio and water volume to sediment surface area, along with underwater light profiles and turbulent mixing. Comparisons were made for plankton systems between the bay and microcosm. Microcosm zooplankton densities matched the bay, but phytoplankton densities were higher, perhaps due to the absence of large grazing macrofauna (fish, large bivalves, and ctenophores). The authors maintained that detailed attention to scaling was necessary for the microcosm to simulate conditions in the bay, and Perez (1995) has elaborated on this philosophy for ecotoxicology applications.
Other examples of scaling tests have compared different sizes of the same microcosm type (Ahn and Mitsch, 2002; Flemer et al., 1993; Giddings and Eddlemon, 1977; Heimbach et al., 1994; Johnson et al., 1994; Perez et al., 1991; Ruth et al., 1994; Solomon et al., 1989; Stephenson et al., 1984). There seems to be a tendency in these studies for plankton-based microcosms to have gradients with size, but benthic-based systems seem less affected by changes in size alone. These studies have the practical application of identifying the smallest sizes of microcosms that can be extrapolated to natural systems while minimizing cost. The most elaborate scaling test of this sort was done at the MEERC project of the University of Maryland's Horn Point Laboratory. This study examined plankton-based systems from the Choptank River estuary for three sizes of microcosms along both constant depth and constant shape (as expressed by constant radius divided by depth of tanks) gradients (Figure 4.11). Petersen et al. (1997) found that gross primary productivity scaled proportional to surface area under light-limited conditions and to volume under nutrient-limited conditions. These results represent a first step towards developing a set of "scaling rules that can be used to quantitatively compare the behavior of different natural ecosystems as well as to relate results from small-scale experimental ecosystems to nature" (Petersen et al., 1997).
Time scaling has received much less attention than spatial scaling of microcosms though both time and space are coupled. A sensitivity to time is often demonstrated in microcosm work in such aspects as diurnal lighting regimes and by the need to conduct experiments during different seasons. However, the central issue of time scaling is the duration of experiments. Most microcosm experiments are run only on the order of weeks or months in order to focus on special treatments such as the effect of a nutrient pulse or a toxin. Longer durations result in successional changes that can complicate the interpretation of these experiments. While the need for short-term studies is necessary for certain types of experiments, there does seem to be a bias in the literature against long-term studies of microcosms. This situation is unfortunate because long-term studies are necessary in ecology to understand many kinds of phenomena (Callahan, 1984; Likens, 1989). In fact, as a rule of thumb, most field ecological studies should be conducted for a minimum of 3 years so that
Was this article helpful?