Introduction to Field Sampling - Field Sampling: Principles and Practices in Environmental Analysis - page 12

whether different soils will be sampled differently or only be noted with a unique label
will be answered. Over what length of time sampling is to continue and what criteria are
to be used in determining when sampling is terminated will also need to be determined.
All of these considerations, along with all other available information, are put together to
produce the final detailed sampling plan [5].
1.7. STATISTICS
Statistics and statistical analysis are essential tools in any sampling plan. First, statistical
analysis of preliminary sampling data can help determine the number of samples that
need to be taken to obtain an accurate picture of the amount and extent of contamination.
The same statistic is applied when remediation has been effected to be certain that the
field is indeed clean [6].
In any sampling some of the samples will give analytical results that the researcher
cannot accept as being valid. A nonbias estimation of the validity of these analytical
results needs to be made, however. There are statistical tools that can be applied to
determine if the analytical results of such a sample are indeed valid or can be discarded.
These tools should be applied to both sample analytical results that are too high and those
that are too low [7].
There are also statistical methods called geostatistics, which can be used to estimate
the extent of contamination. These tools allow one to map the concentration or level of
contaminants over an area. Geostatistical methods can be used in conjunction with global
positioning system (GPS) and GIS in developing an effective sampling plan [8].
1.8. MODELING
Sampling, statistics, GPS, and GIS (see below) are all ways of obtaining information
useful in making a model of a portion of the Earth. In this case a model that will lead to
the most effective sampling plan is produced. Here the word effective includes labor,
containment of the spread of contamination, safety, resources, and money.
Models can be physical models of the environment, simple mathematical models of
single components, or dynamic, complex models of environmental systems. In Chapter 2
some physical nonmathematical methods for modeling the environment will be
described. Although they are nonmathematical, they can lead to an improved and
increased under-standing of what happens in certain environmental conditions. Models
may not be exact, and many refinements may be suggested; however, they do provide
valuable and insightful information about the portion of the environment modeled. For
instance, a model of the way water moves through porous media of different textures and
pore sizes can be informative in many sampling situations [9, 10].
There are a number of computer software programs and complex mathematical
equations that are useful in modeling water movement through the environment. They
can also be used to estimate the fate of contaminants or pollutants. Many of these
programs are specifically aimed at predicting the movement of water through porous