Environmental Sampling Design - Fundamentals of Environmental Sampling and Analysis - page 49

sampling design should consider other factors such as sampling convenience, site
accessibility, availability of sampling equipment, and political aspects. On the top of
the above considerations, regulation constraints in most cases always need to be
consulted first. A good example is the sampling related to remediation projects under
the Superfund Act. The U.S. EPA has specific guidelines for representative sampling
of soil, air, biological tissue, waste, surface water, and sediment in Superfund
investigations. Specific details can be found in the suggested readings. These
nontechnical factors have a significant impact on the final sampling design, but they
are beyond the scope of this text.

3.2 SAMPLING ENVIRONMENTAL POPULATION

3.2.1 Where (Space) and When (Time) to Sample

Environmental sampling can be viewed in a space-time domain (Gilbert, 1987). In
the time domain, there is only one dimension (1-D). Specific time can be designated
as t 1 , t 2 , t 3 ,

t N in a time period such as days, weeks, months, or years. In the space
domain, contaminant variations and hence sampling points can be designated by the
coordinates in 1-D (x), 2-D (x, y), or 3-D (x, y, z). An example of 1-D problem is an
outfall of an industrial wastewater discharge. We would only be concerned about the
concentration as a function of the downstream distance from the discharge point. An
example of 2-D problem is the radiochemical contents in surface soil due to
atmospheric deposition from a nuclear weapon testing site or the lead content in
surface soils downwind from a local smelter facility. In these two cases, the soil
depth may not be the primary interest, so soil samples may be designated by
longitude and latitude, or by measurements relative to an existing structure. A 3-D
sampling site is common in a large body of water, or a solid/hazardous waste landfill
site where depth is a variable for contaminant variations.
In many cases, contaminant variations in both space and time are of interest.
That is, we are interested in both spatial and temporal patterns of contamination. For
example, in a wetland area subjected to seasonal nonpoint source pollution from
agricultural activities, wetland soil samples can be designated by longitude and
latitude as well as by time. Note that in the following discussions on various
sampling design approaches (Section 3.3), space and time are interchangeable from
the design standpoint—meaning that the statistical design approach applicable to
1-D space will also be applicable to temporal variation in the 1-D time domain.

...

3.2.2 Obtain Representative Samples
from Various Matrices

''Representativeness'' is one of the five so-called ''data quality indicators (DQIs),''
or PARCC mentioned in the previous chapter (Section 2.1.3). Unlike precision and
accuracy, representativeness is a qualitative parameter in defining data quality. The
U.S. EPA defines representativeness as ''a measure of the degree to which data