Geoscience Reference
In-Depth Information
Chapter 8 describes the traditional methods for analyzing data and develop-
ments that are more recent. In addition, methods for analyzing data based on
the recovery of dead animals are considered.
Chapter 9, “Occupancy Models,” covers situations where there are data on
the recorded presences and absences of species in different locations but some
of the absences might be because the species was present but not detected.
This then leads to the idea of having a model for the probability that a species
is present at a location based on its characteristics and a model for the prob-
ability of detecting a species if it is present, again based on the characteristics
of the location. These models then require more than one sample to be taken
from each location because if a location is sampled several times and the spe-
cies is seen at least once, then this provides information on the probability of
detecting a species when it is really present. Although site occupancy models
were originally just used when there were presences and absences, they have
now been extended to situations with more than two possibilities regarding
presence, such as absence, presence with breeding, and presence without
breeding. Chapter 9 therefore also considers these types of situations and
situations where the status of a location may change with time.
Chapter 10, “Sampling Designs Used for Environmental Monitoring,”
describes the various types of sampling schemes that are being used for
studies to track changes in environmental variables at local, national, and
international levels. It notes the difference between these types of studies
and research studies that are usually carried out over a shorter period of
time to examine possible changes related to an issue of concern (e.g., a study
to measure the adverse effects of an oil spill). The chapter considers the dif-
ferent types of spatial designs used with environmental monitoring and the
possible designs that allow for repeated sampling at individual locations and
changes with time of these locations.
Finally, Chapter 11, “Models for Trend Analysis,” considers in detail the
types of analyses that are appropriate when there are repeated observations
over time at a number of sampling sites and there is interest in both changes
at the individual sites and changes for the whole geographical area covered
by the sites. The chapter considers the use of various simple analyses and
graphical methods for exploring trends in the data and two approaches
for modeling. One approach involves using linear regression methods for
examining the trends at individual sampling sites and then combining these
results to examine the overall trend for all sites together. The other approach
does one analysis using the data for all of the sites together based on what
is called a mixed model, which essentially allows for individual sites to dis-
play random differences from an overall trend. This chapter uses a data set
on mercury concentrations in fish to illustrate the different methods consid-
ered, with the data collected from the yearly sampling of fish from the same
10 randomly chosen locations in a lake for 12 years.
