Groundwater is relied upon as a major source of water for agriculture, industrial and domestic use. In particular, groundwater serves as a public water supply in much of the world, including Texas. Despite the potential damages due to groundwater contamination, Texas does not currently have laws protecting aquifers which are analogous to laws which protect surface water. According to the Texas Environmental Almanac, all nine major aquifers and 20 minor aquifers have experienced some form of contamination. Although groundwater can be contaminated through natural activities, human causes of groundwater contamination are of particular concern as these can be most easily prevented. Landfills for both hazardous and non-hazardous waste, improperly completed or abandoned water wells, oil- and gas-related wells, petroleum storage tanks, agricultural pesticides and leaking septic tanks are all potential sources for contaminants that can seep down to the aquifer. Deep petroleum, mining and injection wells can likewise directly pollute aquifers. The figure below summarizes typical routes of groundwater contamination.

Source: Adapted from Environmental Protection Agency, Office of Water Supply and Solid Waste Management Programs, Waste Disposal Practices and Their Effects on Groundwater (Washington, DC: U.S. Government Printing Office, 1977).
To develop a way to graphically represent groundwater vulnerability to contamination using GIS. Ultimately, the goal of this project is to protect human health by protecting public water supplies, in particular groundwater public water supplies. There are two main methods of protecting the human population from unsafe groundwater:
Groundwater is often treated significantly less than surface water, as it tends to be a cleaner source. This also means that if contamination occurs, it is more likely to have a negative affect on the population, as less steps are present to remove the contaminant before it reaches the public. A graphical representation of vulnerable aquifers, combined with graphical representations of potential sources of contamination and public water supplies would allow decision makers to evaluate current land use practices and make recommendations for changes in land use regulations which would better prevent the groundwater from contamination. For example, it may not be considered responsible to build a new chemical plant in the contributing area of a particularly vulnerable aquifer or area of an aquifer. Additionally, such a representation would provide a quick tool for determining possible responsible parties if contamination is found, thereby expediting the remediation process. Another potential benefit from mapping vulnerability is that it aids in the prioritization of remediation sites. Responsible parties must pay for the remediation of a polluted site, however in many cases the responsible party can not be found. In these cases the federal government pays for the remediation of the waste site. Unfortunately, there are so many such sites around the nation that they must be prioritized due to limited resources. A vulnerability map would aid in this prioritization.
Before one can begin discussing vulnerable aquifers and public groundwater supplies, it is useful to have an understanding for where these aquifers and water supplies are located. Using data from the TNRIS, I have created maps of the major and minor aquifers in Texas. As is shown in the map below, major and minor aquifers cover much of the state of Texas. The major aquifers have been labeled as polygons, whereas the minor aquifers appear in blue.

Groundwater public water supplies are located on all of the major aquifers. Minor aquifers are also relied upon for public water, but to a lesser degree. In a few cases, groundwater supplies do not seem to be connected to any aquifer. In these cases it must be assumed that some sort of groundwater exists, perhaps in the form of a spring or even smaller aquifer than the ones pictured above. The groundwater public water supplies were plotted using data from a USGS database which contained the latitude and longitude of public water supply wells.

The goal of this project is to protect aquifers from contamination, thereby protecting human health. If potentially contaminating sources are not located in the contributing area to an aquifer, then the aquifer cannot be contaminated due to human causes. Unfortunately, even the most pristine undeveloped regions are most likely not completely immune, as roads generally pass over them, allowing the potential of a chemical spill. When the potential sources of contamination (PSOC) in Texas are examined, one quickly notes that all of the aquifers are threatened by human activities. Aquifers can be polluted from both point and non-point sources. While non-point sources are particularly difficult to predict and control, point sources or potential point sources can be used as a first step in assessing vulnerability. Below I have mapped potential point sources based on their location. The first map contains sites where toxic and or hazardous chemicals are handled. Permitted Industrial and Hazardous Waste landfills are abbreviated as "pihw." Radioactive waste landfills are abbreviated as "radio." Superfund sites are abbreviated as "suprfund."

This map displays the potential sources of contamination which are in the map above, as well as municipal landfills as PSOCs. Anywhere that human populations reside, landfills must be built. Although regulations exist on the types of waste that can be placed in a municipal landfill (i.e. no hazardous waste), municipal landfill leachate can still be dangerous to human health if it reaches the water supply. Therefore, when all potential sources of contamination are considered the entire state is covered. The potential sources of contamination plots were created using data from TNRIS on water quality.

Many regulations are in place to prevent contamination from hazardous and municipal landfills, and from industrial sources. Nonetheless, toxic releases do occur due to unforeseeable occurrences and to negligence. The map below shows the Toxic Release Inventory for the state of Texas. One should note that the most releases occur near and in major urban areas, such as Dallas-Fort Worth in the northeast, Houston in the southeast, and El Paso-Ciudad Juarez in the west. The data for the Toxic Release Inventory was provided by the USGS.

For this component of the project, I relied primarily on internet searches and my prior understanding of contaminant fate and transport. Numerous approaches can be taken to model groundwater vulnerability. A comprehensive groundwater vulnerability model must include parameters to describe how likely it is that a site will contaminated, how the contaminant moves from the site of contamination to the aquifer, how the contaminant moves within the aquifer to public water supplies.
One approach would be to look at the distribution of potential sources of contamination and their relationship to public water supplies. A vulnerability rating would be determined based on the density of PSOCs and the distance from the PSOCs to the contributing areas to the public water supply. However, simply because a PSOC is located on the ground surface in the contributing area to an aquifer does not mean that the aquifer is vulnerable. Multiple hydro geological factors will also contribute to groundwater vulnerability. For example, many contaminants can sorb onto soil surfaces, but the degree to which sorption takes place is dependent on the soil characteristics. Additionally, different types of soil will allow the water to travel faster from the soil surface to the top of the aquifer. If the contaminant moves with the water, then this characteristic will affect the rate of transport and therefore the vulnerability. Many other characteristics similarly can affect the vulnerability of the groundwater to contamination. Additionally, the characteristics of the potential contaminants play an important factor in assessing groundwater vulnerability. An industrial site which uses or manufactures highly volatile compounds would appear in a list of PSOCs, but due to the high volatility of the compound it is unlikely that the compound would reach the aquifer. Other chemical characteristics such as the partitioning coefficient and solubility in water would also have an effect.
A complicated model to completely describe groundwater vulnerability could be developed. Unfortunately, such a model would not likely be useful for planning purposes. As is the case with any model, more parameters also mean that more data is necessary to calibrate and ultimately use the model. To gather data for all of the parameters which affect groundwater vulnerability on a state-wide or even county-wide level would prove to be very difficult. Many parameters have not been recorded. Additionally, uncertainties in the relative weights in the model would make it extremely difficult to calibrate. Hopefully few examples (groundwater contamination incidents) exist where the model could be adequately evaluated.
For these reasons, a simpler model is preferred to a complicated model for the practical application. One model that could provide guidance for comparing groundwater vulnerability on a state or smaller scale level is DRASTIC. DRASTIC is a method developed by the EPA to provide a systematic evaluation of the potential for groundwater contamination that is consistent on a national basis (Aller, L et. al. NWWA/EPA Series. 1987). The DRASTIC parameters are the hydro geologic parameters which affect water transport from the soil surface to the aquifer. The DRASTIC parameters are weighted and then summed to come up with a vulnerability rating or DRASTIC index. DRASTIC assumes that all contaminants move with the water and are introduced at the soil surface. Although it is easy to identify examples in which this assumption is false, DRASTIC provides a provides a tool for relative vulnerability assessment.
From these parameters a DRASTIC index or vulnerability rating can be obtained. The higher the value for the DRASTIC index, the greater the vulnerability of that location of an aquifer.
Depth to Water affects the time available for a contaminant to undergo chemical and biological reactions such as dispersion, oxidation, natural attenuation, sorption etc. A low depth to water parameter will lead to a higher vulnerability rating. Depth to water can be estimated based on well log data from the USGS or the Texas Water Development Board.
Net Recharge is the amount of water which enters the aquifer. This value can be calculated on an annual or monthly basis with data available. Although recharge will dilute the contaminant which enters the aquifer, recharge is also the largest pathway for contaminant transport. Therefore, the amount of recharge is positively correlated with the vulnerability rating.
Net Recharge can be calculated using climate data by applying a mass balance on the water.
Net Recharge = Precipitation – Evaporation – Runoff
Aquifer Media is used to produce a rating based on the permeability of each layer of media. High permeability allows more water and therefore more contaminants to enter the aquifer. Therefore a high permeability will yield a high vulnerability rating.
Some aquifer media data can be found in well logs from USGS and the TWDB. However, this data is not routinely collected and therefore is not reliable. A study needs to be conducted which compiles aquifer media data for this layer to be completed. If the model is to be used on a single aquifer, then literature could be consulted to determine whether it would be reasonable to assume that the media is uniform for the particular aquifer.
Soil media is affects the transport of the contaminant and water from the soil surface to the aquifer. Some of the interactions with soil have already been stated, but for review, the soil media can affect the types of reactions which can take place. Sorption phenomena, for example, can be affected by the structure of the soil surface. Additionally, different soils will provide better habitats for microorganisms which can potentially biodegrade the contaminant. The rating system that is proposed by Aller et. al. follows. This rating system seems to be based on the hydrological transport of the contaminant to the aquifer, rather than on other characteristics.
Range
|
Rating
|
|
Thin or absent |
10 |
|
Gravel |
10 |
|
Sand |
9 |
|
Peat |
8 |
|
Shrinking and/or
Aggregated Clay |
7 |
|
Sandy Loam |
4 |
|
Loam |
5 |
|
Silty Loam |
4 |
|
Clay Loam |
3 |
|
Muck |
2 |
|
Nonshrinking and
nonaggregated clay |
1 |
Soils data is available for STATSGO for entire states and SSURGO for particular counties. Unfortunately the available date is difficult to interpret such that the above ratings could be applied. With some assistance, the STATSGO data could be classified into the above categories.
The topography of the land affects groundwater vulnerability because the slope of the land is in important factor in determining whether the contaminant released will become run-off or infiltrate the aquifer. With a low slope, the contaminant is less likely to become run-off and therefore more likely to infiltrate the aquifer.
Digital Elevation Data (DEM) may be used to calculate and project the slope using GIS.
The vadose zone is the typical soil horizon above and below the water table, which is unsaturated or discontinuously saturated. If the vadose zone is highly permeable then this will lead to a high vulnerability rating.
Information regarding the vadose zone is not readily available.
The hydraulic conductivity relates the factures, bedding planes and intergranular voids in the aquifer. These components become pathways for fluid movement, and likewise pathways for contaminant movement once a contaminant enters the aquifer. The hydraulic conductivity is positively correlated with the vulnerability rating.
The USGS has grid data of the hydraulic conductivity available for select aquifers. For example I have access to the Ogallala aquifer.
Using a GIS, each of the DRASTIC parameters could be graphically represented. By converting all of the files to grid documents, the RASTER calculator could then be used to produce a vulnerability map. An example of a vulnerability map is below.
Stacking of Drastic Layers to Produce a Vulnerability Map

It was originally envisioned that I would produce a similar vulnerability map for the state of Texas. Unfortunately, given the time restrictions of a semester project the steps for creating a vulnerability map have been outlined, but an actual map has not been created.
| DRASTIC can be used to model groundwater vulnerability. | |
| Results of applying DRASTIC model must be used carefully. This applies a framework but does not account for all the particulars of the chemicals released. A detailed study of a particular spill must incorporate the chemical properties of the contaminant. | |
| GIS can help make the results of a complicated model more clear through visual representation, thus providing an applicable tool for decision makers. |