Monitoring Program Design
no difference can be attributed to the management action. Testing a hypothesis is
based on refuting the null hypothesis in order to infer the alternative hypothesis
(Ha:).Thealtemative hypothesisemulates the monitoring objective. Forexample:
Ho: The trend in mean annual dissolved oxygen concentration in Hope Creek has
not increased significantly due to the upgrade at the municipal treatment plant.
Ha: The trend in mean annual dissolved oxygen concentration in Hope Creek has
increased significantly due to the upgrade at the municipal treatment plant.
Ho: The number of algal blooms per growing season in Green Lake has not
declined significantly due to manure management in the watershed.
Ha: The number of algal blooms per growing season in Green Lake has declined
significantly due to manure management in the watershed.
Ho: No significant reductions of nutrients and sediment loading to Grand Lake
have resulted from the implementation of
in the Grand River watershed.
Ha: Significant reductions of nutrients and sediment loading to Grand Lake have
in the Grand River watershed.
resulted from the implementation of
Monitoring Design
Existing data may be used for problem definition, or for a pre-implementation
baseline data set if the collection protocol matches the monitoring objective,
and the Use of
design, and quality assurance/quality control required for the post-implementa-
Existing Data
tion data collection.
Existing data may also be used for assessing concentration/load/biological mea-
surement variability and estimating the number of samples or the time period for
the monitoring survey, based on the desired level of significance and error.
To determine the required sampling frequency and evaluate monitoring feasibil-
Minimum Detectable
ity, the minimum detectable change (MDC) should be estimated from historical
Change
variable over time that is considered statistically significant. The larger the MDC,
the more change in water quality is needed
that it was not just a random
fluctuation. It may be reduced by accounting for explanatory variables, increasing
the number of samples per year, and increasing the number of years of monitoring.
Achieving a high level of statistical significance and power when background
variability is high requires a large number of samples and a sophisticated
monitoring design.
The type of change must be defined in relation to the pollutant constituent and the
water quality problem in order to specify the monitoring objective. For
that
are directed toward reducing acute impacts, monitoring extreme events may
provide evidence of change. However, tracking chronic impacts (e.g., toxins or
nutrients) may require a long-term monitoring program.
4.7
Previous Page
