Environmental Engineering Reference
In-Depth Information
certainly be consistent with general knowledge about the phosphorus cycle (see Chapter 8).
On the other hand, we just observed that human activities often control ecosystems, and
a graph that relates primary production in the lake to human population along the lake
shore over time might well show that primary production is well correlated with human
population density. So we might answer that primary production is controlled by human
population density. But an astute observer might notice that most of the lake frontage
has green lawns. This observer might reasonably conclude that primary production in
the lake is controlled by local economic growth or by social norms about having a green
lawn. This example illustrates the familiar problem of proximate (P-loading) versus
ultimate (social norms, technological capability, or economic growth) causes of a phe-
nomenon, as well as the common situation where more than one “controlling” factor
is important—all of which complicates our ability to identify “the” control over some
aspect of ecosystem function.
A somewhat related issue is that “control” may imply different levels of mechanistic
understanding of the problem. Sometimes, scientists and managers will be satisfied with a
“black box” approach that does not explicitly consider any of the mechanistic pathways
that link the controlling variable and the ecosystem (see Figure 1.3 in Chapter 1). For other
purposes, a much more detailed understanding of mechanisms may be needed. Although
much ink has been spilled over which of these interpretations of control is “correct,” both
approaches clearly can be useful, depending on the scientific or management issue at
hand.
Finally, the answer to “What controls ecosystem function?” often depends on the
scales of time and space over which the question is posed. Many ecosystem scientists
are interested in knowing what controls denitrification (the conversion of nitrate to
dinitrogen gas by certain bacteria), because this process leads to important losses of
fixed nitrogen from ecosystems and can generate greenhouse gases (see Chapter 7).
Denitrification occurs primarily under anaerobic conditions, and requires nitrate and
labile organic carbon. Not surprisingly, if we are looking for controls on denitrification
rates over the year at a single site, we often find that rates are controlled by soil moisture
or flooding (which controls oxygen concentrations) or pulses of nitrate. However, if
we are interested in what controls denitrification rates across a region, variables such as
local topography, soil texture, and temperature may be better predictors. If we were
interested in global denitrification rates at geologic timescales, we might find that
atmospheric oxygen concentrations would be the best predictor. Consequently, two
scientists studying the controls on the same ecosystem attribute may come up with
different answers, and both may be right. Thus, “answers” should be given along with
their spatio-temporal contexts.
Considering these complications, the great complexity of ecosystems, and the wide
range of possible controls, it is perhaps surprising that we can say much at all about what
controls ecosystems. Yet as the material in this chapter and throughout the rest of the
topic shows, scientists do understand a lot about what controls various aspects of ecosys-
tem structure and function. This understanding underlies modern ecosystem management,
and so is of great practical value. Ecosystem scientists of the future will be challenged to
refine and extend this understanding, as well as to determine whether broad generaliza-
tions about ecosystem controls can be developed (see Chapter 17).
