Page 31 - Environment: The Science Behind the Stories

would want to test all possible explanations. For instance, our The scientific process continues
researcher might formulate an additional hypothesis, proposing beyond the scientific method
that algae increase in fertilized ponds because chemical fertiliz-
ers diminish the numbers of fish or invertebrate animals that Scientific work takes place within the context of a commu-
eat algae. It is possible, of course, that both hypotheses could nity of peers. To have impact, a researcher’s work must be
be correct and that each may explain some portion of the initial published and made accessible to this community. Thus, the
observation that local ponds were experiencing algal blooms. scientific method is embedded within a larger process involv-
ing the scientific community as a whole (FIGURE 1.11).
We test hypotheses in different ways
Peer review When a researcher’s work is done and the
An experiment in which the researcher actively chooses
and manipulates the independent variable is known as a results analyzed, he or she writes up the findings and sub-
manipulative experiment. A manipulative experiment provides mits them to a journal (a scholarly publication in which sci-
the strongest type of evidence a scientist can obtain, because entists share their work). The journal’s editor asks several
it can reveal causal relationships, showing that changes in an other scientists who specialize in the subject area to examine
independent variable cause changes in a dependent variable. In the manuscript, provide comments and criticism (generally
practice, however, we cannot run manipulative experiments for anonymously), and judge whether the work merits publication
all questions, especially for processes involving large spatial in the journal. This procedure, known as peer review, is an
scales or long time scales. For example, in studying the effects essential part of the scientific process.
of global climate change (Chapter 18), we cannot run a manip- Peer review is a valuable guard against faulty research
ulative experiment adding carbon dioxide to 10 treatment plan- contaminating the literature on which all scientists rely.
ets and 10 control planets and then compare the results! However, because scientists are human, personal biases and
Thus, in environmental science, it is common for research- politics can sometimes creep into the review process. Fortu-
ers to run natural experiments, which compare how dependent nately, just as individual scientists strive to remain objective
variables are expressed in naturally different contexts. In such in conducting their research, the scientific community does
experiments, the independent variable varies naturally, and its best to ensure fair review of all work. Winston Churchill
researchers test their hypotheses by searching for correlation, once called democracy the worst form of government, except
or statistical association among variables.
For instance, let’s suppose our sci- Scientific process (as practiced by scientific community)
entist studying algae surveys 50 ponds,
25 of which happen to be fed by ferti-
lizer runoff from nearby farm fields and
25 of which are not. Let’s say he or she Further
finds seven times more algal growth in research
the fertilized ponds than in the unferti- by scientific
lized ponds. The scientist would con- Scientific method (as practiced by community
individual researcher or research group)
clude that algal growth is correlated with
fertilizer input; that is, that one tends to Observations
increase along with the other. Publication
in
This type of evidence is not as scientific
strong as the causal demonstration that Questions journal
manipulative experiments can provide,
but often a natural experiment is the only
feasible approach for studying a subject Hypothesis Paper rejected Paper accepted
of immense scale, such as an ecosystem
or a planet. Because many questions Revise
in environmental science are complex Predictions paper
and exist on large scales, they must be Reject
addressed with correlative data. As such, Fail to hypothesis
environmental scientists cannot always Test reject Peer review
provide clear-cut, black-and-white hypothesis
answers to questions from policymak-
ers and the public. Nonetheless, good Results
correlative studies can make for very Scientific paper
strong science, and they preserve the
real-world complexity that manipulative
experiments often sacrifice. Whenever FIGURE 1.11 The scientific method followed by individual research teams exists within
possible, scientists try to integrate natu- the overall process of science at the level of the scientific community. This process
ral and manipulative experiments to gain includes peer review and publication of research, acquisition of funding, and the elaboration of
30 the advantages of each. theory through the cumulative work of many researchers.

M01_WITH7428_05_SE_C01.indd 30 12/12/14 9:32 AM