Geography Reference
In-Depth Information
5. Put data acquired into proper form (reformat, or encode it) for inclusion in the GIS database.
6. Check accuracy of each step of the collection process; also check the first form of the data against the
form in the GIS.
7. Repeat Steps 4, 5, and 6 for the complete set of data for the database.
8. Employ techniques for monitoring and updating of the database.
We now look at these steps one at a time.
1. Determine what types of data are needed for the information products(s).
This assumes that you have determined the needs of the decision-making apparatus and, further, that
you can identify the sort of information which will satisfy those needs.
The most important component in this step is to have people involved who understand (a) the information
required (and how it is to be ultimately used), (b) the characteristics of data that might be used to generate
the information, and (c) the manipulation of the data necessary to produce the information.
At this early stage, it is wise to consider alternative and innovative ways of getting to the same
information. For example, if the information required is a delineation of areas of potential high soil
erosion, then calculating the potential soil loss might be produced using the universal soil loss formula
and data, including rainfall, soil erodibility, slope length and gradient, and vegetative cover. However
such data might be produced by interpreting aerial photography for existing erosion conditions. Many
projects that get into trouble at this stage do so probably because communications with the analysis
and decision-making group ceased after the initial contact. Problems are constantly changing in both
importance and type. Thus, to be most responsive, the data support sector of a decision-making process
must have its roots not in the data collection area, but rather in the analysis and decision-making area.
2. Once the basic types of data required have been identified, more thought must be given to the
characteristics of the data to be acquired.
In determining the characteristics of the data needed, several fundamental questions should be asked.
What geographic area is involved? What geographic identifiers are necessary for use of the data? With
what accuracy must the coordinates be known? Are the values of the data “continuous” (like elevations
above sea-level) or “discrete” (classifications of land cover)? How frequently do data values change?
What causes these changes? Are the most basic data types in use, or can other data be derived from
more basic sources? If the latter, what are the advantages and costs in using the most basic information
available? What degree of detail is required? When the detail level desired is “multiplied” by the area
involved, how large are the datasets? How sensitive to errors in the data is the process that is used to get
information products from the data?
3. Make some preliminary studies to determine that the data and characteristics specified will
produce the information wanted.
How this task is done depends greatly on what is available. If a GIS exists and new datasets to support
a new product are being added, the best course might be to generate some typical subarea of data to
try out the process. If the GIS does not exist but is being installed to produce the product, the issue of
preliminary testing of the data-product relationship must be approached differently. Perhaps an analysis
Search WWH ::
Custom Search