Environmental Engineering Reference
In-Depth Information
control that variable. An example of such an analysis is shown in Table 2.1 in Chapter 2.
Alternatively, scientists often test whether some factor controls an ecosystem by compar-
ing ecosystems that differ in that factor and not (to the extent possible) in any other rele-
vant characteristic ( Cole et al. 1991 ). For instance, if we wanted to test whether
phosphorus inputs control primary production in lakes, we might try to measure primary
production in a series of 10 lakes of similar size, depth, and terrain that differ in their
phosphorus inputs. In practice, it often is difficult to find such a perfect series of well-
matched study sites.
Experiments
Experiments, whether conducted in the laboratory or in the field, are powerful ways to
reveal controls on ecosystem structure and function ( Likens 1985; Carpenter et al. 1995 ).
There are no rules about the size of experimental units: manipulations have been made across
hundreds of square kilometers (e.g., iron fertilization experiments conducted in the ocean)
and within square centimeters. Often, the goal of experiments is to measure an ecosystem's
response to a change in a single variable while holding all others as constant as possible. For
example, to understand whether phytoplankton in lakes were controlled by phosphorus or
by other nutrients such as nitrogen and carbon, scientists in the Experimental Lakes area of
Canada added phosphorus, nitrogen, and carbon to one-half of a lake (cut in two by a mas-
sive curtain) and just nitrogen and carbon to the other half. They then compared responses—
such as the amount of primary production—in each half of the lake to see what effects the
treatments had (see Chapter 8). This whole-lake experiment helped to demonstrated that
phosphorus was a major factor controlling algal productivity in lakes.
Theory and Conceptual Models
As in other sciences, ecosystem scientists routinely use theory and conceptual models.
Such theories and models are highly varied in structure and purpose ( Canham et al. 2003;
Pickett et al. 2007 ). Models may be as simple as a statistical regression (see Chapter 11) or
a box-and-arrow diagram drawn on a napkin, or as complex as a detailed simulation
model ( Figure 1.4 ). Models are highly flexible, can cover scales of time and space that are
difficult to study using other approaches, and often can provide quick answers at low
cost. They also are very useful as a way to organize facts and ideas; to generate, sharpen,
or narrow hypotheses; and to guide research activities. Scientists often make rapid prog-
ress by tightly coupling theory and models to other approaches.
What Do Ecosystem Scientists Measure?
Ecosystem scientists are inherently interested in the connections between structure and
function of ecosystems and how they develop over time. Thus, many of the examples of
measurements or values in this topic are related to structure and function, such as biomass
of a species, or rates of carbon cycling. They are what is often found on the x or y axes of
graphs, or are used as treatments or are measured as responses in experiments. Ecosystem
structure is sometimes measured by variables such as leaf area index or the number of
trophic levels in a lake (see Chapter 11). Productivity (Chapters 2 and 3), rates of
Search WWH ::




Custom Search