When starting a new monitoring program, it often is a good idea to collect four rounds of quarterly data, particularly if investigation data for your site are limited or obsolete.



When starting a new monitoring program, it often is a good idea to collect four rounds of quarterly data, particularly if investigation data for your site are limited (e.g., from one round of sampling, or from only one time of year) or obsolete (e.g., more than 3 years old). Four quarters of analytical and water-level data will help establish the presence of any temporal (such as seasonal) and spatial variability. In addition, four data points often are considered the minimum for conducting any sort of statistical evaluation.

It is essential that all monitoring data be collected using the same sampling and analytical methods to ensure comparability. Your ground water monitoring plan should be used to document these methods. If a recent, well-designed site investigation has been conducted, starting a monitoring program with semi-annual or even annual monitoring may be more appropriate.

Following the first year of quarterly data collection, monitoring frequency may be reduced as appropriate, following decision criteria built into the monitoring plan. Specific decision criteria should be included for determining when monitoring may be discontinued at the site. A review period – most likely annual – should be specified in the plan to periodically evaluate the potential for site closure based on monitoring data and closure decision criteria.

The purpose of a well should be taken into account when determining the sampling frequency. Down-gradient, plume-edge wells generally require more frequent sampling than an up-gradient or background well. Special purpose wells – such as sentinel wells – may need to be sampled often to safeguard human health. Likewise, point-of-compliance wells may need to be sampled more frequently than on-site wells to help maintain good faith between your installation and neighboring communities.

Following a year of quarterly sampling, you then could seek a decrease in monitoring frequency, tailoring frequency to the function of the well. Whereas point-of-compliance and corrective-action observation wells are recommended for semi-annual sampling, up-gradient, background and supplemental wells could be dropped to annual sampling. If approximately half the monitoring wells at the site were decreased to semi-annual sampling, while the other half were decreased to annual sampling, more than 60 percent of analytical costs could be saved in the second year of sampling. Field labor costs would decrease significantly, and eliminating two quarterly sampling rounds would cut mobilization and demobilization costs.

Flow Calculations

Calculations can be used to estimate the rate of ground water flow at a site. Although the rate of contaminant movement usually is not as fast as ground water movement, the use of simple flow equations can provide a conservative estimate of how long it will take contamination to reach a particular point, such as the installation boundary or a supply well. This information can then be used to determine an appropriate sampling frequency and duration.

The following example describes how basic flow rate information can be used to determine the frequency of monitoring at a given point: Benzene from a fuel spill site is estimated to travel at the same rate as ground water, which is modeled to move at 7 feet per year. The frequency of monitoring at a given point should be related to the rate of contaminant movement. For instance, it may be necessary to sample wells close to the contamination on a quarterly basis. However, it is not reasonable to sample a clean well that is located 350 feet down-gradient from known contamination with the same frequency, because it will take 50 years for the benzene to reach the well.

Additionally, if contamination isn’t detected in down-gradient wells within a reasonable timeframe based on flow calculation results, it may be determined that contamination will not reach the site boundary, and monitoring may be discontinued. An example exit criterion for such a case would be any well where contamination is not detected over the course of three travel times (i.e., the estimated or modeled time that it would take for the contaminant to travel from the source area to the well) would no longer be monitored.

Trend Analysis and Statistics

By identifying data trends at your site, you will be able to propose the most appropriate sampling frequency. If a simple concentration-vs.-time plot of the data indicates that concentration trends in target analytes are not changing rapidly, monitoring may be decreased to semiannually. Following a year of semi-annual data collection, a similar analysis can be made to see if a reduction to annual monitoring might be implemented. If the trends of concentration over time are not clear, it may be helpful to conduct temporal trend analyses. Trend analysis or statistics also may be used to support a decision to stop monitoring at a well or a site if contaminant concentrations are found to be stable over a long period of time. It may be possible to statistically show that there is not a significant difference between up-gradient and down-gradient concentrations of target analytes at a site. In this case, it also may be appropriate to stop monitoring at the site.

The Decision Criteria

After each sampling event, or at least annually, you should re-evaluate the objectives of the ground water monitoring program. Determine if any of the decision criteria for reducing the frequency or duration of monitoring at a site or individual monitoring point have been met. Decreasing the number of samples through reductions in sampling duration and/or frequency is an important aspect of optimizing an existing ground water monitoring program. Reducing monitoring frequency by 50 percent will decrease sampling labor, analysis, validation and reporting costs by a like percentage. The general approach to this type of optimization essentially is the same as presented for designing a new program. The important difference is that existing programs may not have pre-approved decision criteria for optimizing frequency and duration. 
ND



This article is provided through the courtesy of the Air Force Center for Engineering and the Environment. It is excerpted from the “Guide to Optimal Groundwater Monitoring,” prepared for the Naval Facilities Engineering Service Center by Radian International, White Rock, N.M.