for pinpoint location mapping. The locations of the points may not be very
accurate. Complete metadata will include information on whom to contact
if there are questions about the collection, processing, and packaging of the
data.
10.2.2 Units
A clear definition of units used in data measurements and representation
is essential to proper data interpretation and use. Descriptive units may
include units of measurement including area, volume, temperature, etc., as
well as time-step definitions, such as minutes, hours, and days.
10.2.3 Assumptions
Assumptions underlying data dictate reasonable use. What has been done
to the data? What has not been done to the data? Were the data collected
under normal conditions? Are these data representative of a certain place
or phenomenon? An understanding of the limitations of data and models
is essential to good analysis.
10.2.4 Variability
Data variability, both inherent and that resulting from a sampling scheme,
should be clearly documented and quantified. It includes both spatial and
temporal variability. Knowledge of variability should guide the choice of
sampling scheme and direct the use and interpretation of existing data.
10.2.5 Consistency in Sampling
Changes in the data collection may compromise the use of the data or
change the underlying assumptions of the data. For example, if collection
of data on fish species was initially obtained by trapping or netting and later
by electrofishing, then assumptions about changes in population from the
initial to later times may be invalid (Cairns and Smith 1993).
10.2.6 Scale/Resolution
A complete description of spatial and temporal scales and resolution are
necessary to determine the appropriate use of data. Scale refers to a defined
dimensional relationship between reality and the representation of reality,
while resolution is a measure of the ability of a device to differentiate a
value (Robinson et al. 1984). For example, a map depicting the world would
be a small-scale map, while a similarly sized map depicting a county would
be a large-scale map. Resolution is positively related to differentiating fine
detail at a given scale. Once a data set has been formulated, changing the
10. Effective Ecological Modeling: Data Issues
185
accurate. Complete metadata will include information on whom to contact
if there are questions about the collection, processing, and packaging of the
data.
10.2.2 Units
A clear definition of units used in data measurements and representation
is essential to proper data interpretation and use. Descriptive units may
include units of measurement including area, volume, temperature, etc., as
well as time-step definitions, such as minutes, hours, and days.
10.2.3 Assumptions
Assumptions underlying data dictate reasonable use. What has been done
to the data? What has not been done to the data? Were the data collected
under normal conditions? Are these data representative of a certain place
or phenomenon? An understanding of the limitations of data and models
is essential to good analysis.
10.2.4 Variability
Data variability, both inherent and that resulting from a sampling scheme,
should be clearly documented and quantified. It includes both spatial and
temporal variability. Knowledge of variability should guide the choice of
sampling scheme and direct the use and interpretation of existing data.
10.2.5 Consistency in Sampling
Changes in the data collection may compromise the use of the data or
change the underlying assumptions of the data. For example, if collection
of data on fish species was initially obtained by trapping or netting and later
by electrofishing, then assumptions about changes in population from the
initial to later times may be invalid (Cairns and Smith 1993).
10.2.6 Scale/Resolution
A complete description of spatial and temporal scales and resolution are
necessary to determine the appropriate use of data. Scale refers to a defined
dimensional relationship between reality and the representation of reality,
while resolution is a measure of the ability of a device to differentiate a
value (Robinson et al. 1984). For example, a map depicting the world would
be a small-scale map, while a similarly sized map depicting a county would
be a large-scale map. Resolution is positively related to differentiating fine
detail at a given scale. Once a data set has been formulated, changing the
10. Effective Ecological Modeling: Data Issues
185