Problems

GIS is not the truth about the world!: There can be many problems with data by the time it reaches the computer of an SFU student. Sometimes these problems are mentioned in the metadata, however, for this project none of the metadata for the shapefiles addressed any issues. Therefore, an interpolation of possible problems include the fact no data represents the truth about the world. Data is always an abstraction of what is really present on the ground. The county boudaries, the well location data, the precipitation data, the surface water data and the governance data have all been filtered through several processes including data collection, input and rendering before reaching SFU. At each of these stages it is possible for error to creep in and perpetuate in ways that are not readily apparent. For example, someone collecting Texas river data using a GPS might enter that data incorrectly into their GIS data framework. As well, it is important to consider the level of abstraction that has taken place in the data rendering. Precipitation data for this project was available in the form of contour lines, but contours can misrepresent what might be very different local precipitation values, especially when they are rendered on a statewide scale as was used for this project. A preferable data set for precipitation would be actual precipitation measurements (rather than ranges) from sample sites on the Ogallala. It is also important to consider the age of the data. The Ogallala is changing at such a rapid rate that temporal considerations are important before appropriate decisions can be made. It is possible that the amount of wells on the Ogallala have increased significantly since 1990-1991 when the shapefiles used as source data for this project were created. Another problem was the lack of well withdrawal information. Without this data the project was forced to rely on well density instead of actual withdrawal levels. This could be a misrepresentation of actual stress placed on the Ogallala if water users in areas with less well density pump significantly more water per well than users in areas with high well density. Without withdrawal data available, one way to deal with this uncertainty would be to use a per-capita well density. Gauging the amount of people per well could provide an indication of the level of withdrawal. However, this too is somewhat misleading as it is not individuals but farms that are the major source of groundwater depletion in the region. As many of the farms distribute their products outside the Ogallala region, a per-capita representation would be flawed.

The main weakness of the analysis was the fact that several assumptions had to be made about the data due to lack of available research information. For example, in preparing the Governance/Surface water ranking, values from 0-255 were assigned rather arbitrarily based on common sense rather than scientific study. The following is a list of the project's assumptions and the problems associated with them:

1. Precipitation was assumed as the primary limiting factor due to the fact that without recharge, the aquifer will eventually disappear regardless of withdrawals. Problem: The Ogallala is a huge aquifer that extends across most of the Great Plains, rainfall over the Texan counties is not the Ogallala's only source of recharge.

2. Well density was the second most important limiting factor as this data was treated as an indicator of withdrawals from the Ogallala (ie. the reason for depletion). Problem: This was already discussed in the section above: high well density does not necessarily indicate high withdrawals.

3. The presence of various governance factors and of surface water were lumped together into one ranked index because they could all be rendered in a boolean format that indicated 'present' or 'not present'. It was assumed that the presence of each of these factors contributed to lower water withdrawals with varying degrees of importance placed on each combination of factors. Problem: This part of the project was the most problematic. Broad assumptions were made about the effectiveness of governance and how that might compare to surface water coverage in the Ogallala region. Without actual data on the effectiveness of the Groundwater Conservation Districts and the Priority Groundwater Management Areas it not fair to state that these factors are of less importance than the presence of surface water. In this part of the project, it was also assumed that all surface waters were accessible for irrigation and domestic use. It is more likely that this is not the case. An analysis that used only accessible surface water would probably produce a much smaller coverage and might alter the results significantly.