Focus On Water Division
Nebraska Department of Environmental Quality
Standard Operating Procedure

Water Quality Division
Water Quality Assessment Section - GW

Standard Operating Procedure Number GW-020
Title: Sample Location Selection for Regional Groundwater Sampling
Written Date: September 1, 2000
Approved Date:


Purpose: To select regional ground water study sampling locations in a uniform, efficient, and defensible manner.

Equipment/Materials Needed:

· Listing of registered wells for the area of interest
· Well construction and geologic logs for as many wells as possible
· Any available previous water quality analysis results for the area of interest
· Regional hydrogeologic information/maps (water table, base/thickness of aquifer, flow system, rock units, etc.)

Procedures:

1. Determine the number of wells in the study area, using the most current available Nebraska Department of Natural Resources (NDNR) well registration list (available on NDNR website).


2. Determine the target number of wells which will be sampled for the study.
2.1. The default target is 10% of registered wells within the study area. If 10% is not possible, a smaller percentage is allowable. In determining the target number, the project manager should consider the following factors:
2.1.1. The overall number of wells in the study area; e.g. 10% of a few hundred wells is obtainable, whereas 10% of many thousand wells may not be feasible.
2.1.2. Spatial coverage; e.g. if the study area has many wells concentrated in limited areas, careful selection of a smaller number of “representative” wells may ensure adequate coverage.
2.1.3. Background information; e.g. if an area has been extensively studied and the water quality documented by previous investigations, a smaller number of samples may be necessary for area delineation, contaminant level concentrations, etc.
2.1.4. Variability of water quality; similarly to #2.1.3., if an area has significant existing water quality data and the levels of contamination have been documented to be relatively stable at a certain level, fewer samples may be necessary.
2.1.5. If fewer than 10% of the wells are sampled, the project manager will document the reasons for the smaller percentage.
2.2. In general, the higher the percentage of wells sampled, the better (e.g. Luzzadder-Beach (1995) suggests about 15% is adequate). Especially for smaller study areas, it may be possible to sample a higher percentage (e.g. 20, 30, 40% etc.) of registered wells; where possible, this is encouraged.
2.3. Statistically-based methodologies for determining sample populations are also a possibility. For instance, the number of samples may be tied to variability in nitrate levels in existing ground water sample results. If such a methodology is used, document the reasons and/or references for doing so.
2.3.1. Wells with previous water quality analyses and/or water level data should be included wherever adequate background information is available to allow for comparison with historic data, evaluation of monitoring schemes, etc.
2.4. After arriving at the target number for actual samples, increase that number by an overselection factor to account for problems in performing the sampling (e.g. wells not running, false well locations, idle land, denial of permission, etc.). The default minimum overselection factor is 100% (e.g., if your target number for actual samples is 10%, select 20% of the registered wells). As always, higher percentages are acceptable if necessary and feasible.
3. Select sampling points. Use any one of the following methods (the actual method selected should be documented in the final report):
3.1. Random selection--Select the required number of samples randomly from the entire list of registered wells. Any random selection tool is acceptable; the most commonly available tool is a random number generator on a hand-held calculator.
3.2. Stratified random selection--designate specific strata (e.g. counties, townships, sections, well construction, landforms, aquifer types, etc.), and randomly select the required number of samples from within each stratum.
3.3. Regular spacing--select a well (if available) for a specific distance or area (e.g. one per section, one per four sections, etc.).
3.4. The above statistical methods of selecting sampling points may be augmented by selecting additional sampling points along ground water flow lines or up/downgradient of the area.
4. It is advisable to consider the quality of available information regarding wells when selecting them for sampling. For instance, depending upon the objectives of the study, it may be best to limit wells under consideration to those having a certain amount of registration information, those completed at a certain depth, certain well types, etc. In this sense, the selection process may not be entirely random or may have a certain amount of bias. This is acceptable in that this bias is toward wells that will yield more information.
5. Once an adequate number of wells have been selected, notify the cooperating NRD(s) with copies of the list of selected wells, including as much owner/operator data as possible. This should be done well ahead of sampling, usually at least a few months beforehand.
5.1. Determine who will be responsible for contacting landowners. In most cases, it is desirable to have the cooperating NRD(s) perform this task as they are more well-known in the study area. There are several methods of contacting landowners; all of them are acceptable, depending on the NRD’s preference:
5.1.1. Contacting individual landowners by phone.
5.1.2. Contacting individual landowners by mail.
5.1.2.1. Some NRDs prefer a positive response to permission requests by mail, i.e. requesting that the landowner respond to the request by calling the NRD or mailing in a response card specifically giving permission to sample. In other cases, NRDs assume that permission is given unless the landowner responds specifically denying the request. Specific permission is always preferred, but either method is acceptable.
5.2. Ensure that landowners are contacted well in advance of sampling, and that enough of them have given permission to obtain a target number of samples.
5.2.1. If not enough landowners have given permission to obtain an adequate number of samples, select additional sampling points using the above procedures.
5.3. In addition to the above methodologies, permission for sampling can be obtained in the field if necessary.
5.3.1. In many cases, an owner/operator of one well will ask you to sample others which he/she controls. This is very much encouraged as long as resources allow.
5.3.2. In some cases, field observations may reveal that samples from a certain well would be valuable because of the well’s location, characteristics, position relative to geologic features, etc. In this case, use the well lists, ownership maps, phone books, NRD contacts, etc. to find the owner/operator and contact her/him for permission.

REFERENCE
Luzzadder-Beach, S. 1995. Evaluating the effects of spatial monitoring policy on groundwater quality portrayal. Environmental Management, v. 19, no. 3, p.383-392.