SamplingAs part of obtaining permit coverage, you must develop a sampling program to demonstrate that your facility will meet the compliance standards listed in WAC 173-308 Biosolids Management. A biosolids sampling program has three main components:
Developing a Sampling and Analysis PlanGeneral Sampling Considerations
All Sampling and Analysis Plans must include sampling protocol. A sampling protocol must take into consideration: the medium (biosolids, soil, ground water, surface water, plant tissue), purpose of the monitoring, and conditions at the site. Refer to Useful Links & Tools of our Regulatory Links & Resources for more information, and consult with experienced professionals and the biosolids coordinator in your region to make sure that all aspects of your sampling program will meet regulatory and management needs.
Sample Collection Responsibility
The operator has responsibility for collecting samples. You may manage the sampling program independently, or you may contract with an outside company. Regardless of the party that actually performs the sampling; the facility is ultimately responsible for the integrity of the sample collection and the resulting data.
Choosing a Lab
The laboratory providing the analytical service can provide guidance on container requirements, and may provide the containers. Below is a general guide for sample container materials. Consult the laboratory prior to sample collection.
Samples must be thoroughly documented. Set up a:
Identify which safety precautions are needed for the planned sampling method. Toxic fumes (ammonia or other gases from anaerobic digestion) can be hazardous when samples are collected in an enclosed area. To minimize pathogen exposure be sure to clean properly after contact with any potential pathogen containing material. For in-plant sampling, the sampling should be supervised by a certified operator.
Determine the sampling protocol:
Composite sampling is a good way to account for sample variability without increasing analytical costs. Grab samples are discrete samples collected at one location and time. Composite samples are a series of grab samples collected over several locations or times, and combined together.
To obtain high-quality composite samples, observe the following precautions:
Accuracy of Analytical Results
One way to check for consistency in the laboratory is to submit a reference sample with every batch of samples. This is especially important if several labs will perform the same analytical test. You can compare reference sample results over time and among laboratories as a check on results.
Because element concentrations are stable for many years in dried material, a dried, finely ground biosolids or soil will work well as a reference sample for trace elements, nitrogen, and plant nutrients. Store the dried reference sample in a sealed container. Because pathogens are not stable when stored, reference samples cannot be used for pathogen testing.
Biosolids Sampling Considerations:The sampling protocol varies depending on wastewater treatment and stabilization processes. Consult with the Department of Ecology and others using similar wastewater treatment and stabilization systems for guidelines on designing a sampling protocol.
Since trace elements are not altered during transport and short-term storage, these samples can be collected at the wastewater treatment plant. If the historical data shows that trace element levels are always well within Table 3 limits, then a single composite sample submitted at the required intervals should be sufficient. If levels sometimes exceed limits, then a statistical estimate for determining sample numbers similar to the one for nitrogen should be used. For stable constituents, such as trace elements, a series of daily composite samples can be combined into a longer-term (typically two week) composite.
Pathogens and Indicators
Seven discrete samples for fecal coliform analysis are required for Class B biosolids when Alternative 1 (WAC 173-308-170 Pathogen Reduction) is used. Microbial samples are often collected as grab samples rather than composite samples over time because the samples are perishable. The samples should be collected over a two week period. You should establish specific sampling points for pathogen sample collection. Pathogen contamination can occur even after pathogen treatment processes have been completed.
Salmonella testing to meet Class A biosolids requirements is more expensive and requires greater expertise than fecal coliform testing. The basis for using fecal coliform instead of salmonella testing is the correlation between fecal Coliform density and the presence of salmonella. Research with a variety of biosolids types and stabilization methods has shown that salmonella are usually not detectable when fecal coliform density is less than 1000 MPN per gram of dry solids.
For nitrogen, collect samples as close to the time of field application as possible. Ammonia loss occurs rapidly during the final processing steps (dewatering, lime stabilization, and bed drying). This loss can also occur when biosolids are stored at the application site.
The number of biosolids samples for total nitrogen analysis can be estimated from the standard deviation derived from previous nitrogen analyses and the desired accuracy of the estimate. The equation below will estimate the number of samples required to be within the desired limit of accuracy with a 95 percent level of confidence.
n is the number of samples required,
s is the standard deviation estimated from previous results, and
L is the desired limit of accuracy.
Previous total nitrogen analyses had values of 4.8, 4.2, 5.2, 5.9, and 3.9%
The sample mean and standard deviation calculated from these values are 4.8% and 0.80%.
If the desired limit of accuracy of the total nitrogen analysis is 1.0% (20lb/ton),
The number of samples rounded to the nearest whole number is 3. If the standard deviation is larger or the desired error limit smaller, the number of samples would increase.
Field Sampling Considerations:Specific sampling instructions depend on the site and medium sampled. Consult with the Department of Ecology to determine whether your site requires a sampling plan, and if so, what type of design is appropriate.
When possible, analyze domestic supply wells on or near the project site for nitrate and fecal coliform bacteria. Contaminants in groundwater reflect past activities, even activities from off-site, so groundwater monitoring may not tell you much about current activities. Biosolids applied at agronomic rates should not affect groundwater quality. When required, monitor nitrate-nitrogen and fecal coliform bacteria in domestic water supply wells. If a detailed investigation of groundwater quality is necessary, a professional hydrogeologist should evaluate the site and design the sampling program.
Nitrate monitoring in and beneath the crop root zone can demonstrate that nitrate leaching is not degrading water quality. Nitrate monitoring may be appropriate on sites with high biosolids application rates. Soil sampling can be used at most locations (see the next section for details).
Use suction lysimeters for water sampling beneath the root zone on sites where soils are too rocky for routine soil testing. Only qualified personnel should install lysimeters, collect samples, and interpret the data. The lysimeters must be located accurately to represent the site. Because of site variability, you may need many lysimeters to characterize water quality. Periodic sampling gives only snapshots of nitrate levels. The quantity of nitrate leached cannot be calculated without an estimate of the volume of water moving through the soil.
Surface water is usually monitored only after biosolids application. Comparing pre and post-application surface water samples is usually not valid because there are wide variations in water quality due to seasonal or other factors. A good monitoring system for surface water provides an estimate of how off-site factors affect water quality. The usual approach is to monitor points both upstream and downstream from the biosolids application site. Surface water contaminants can originate from domestic and wild animal waste and from failing on-site septic systems.
Basic measurements for surface water include ammonium-nitrate, temperature, pH, and fecal coliforms. Ammonia is the most toxic form of nitrogen in surface water. Ammonia levels are derived from measurements of ammonium, pH and temperature.
A soil test shows the amount of available nutrients. Soils tests are a standard part of agricultural production, and are important when making biosolids applications to agricultural land. The test should include pH and available phosphorus and potassium. Ask for a micronutrient test if you suspect micronutrient deficiencies.
Repeated biosolids applications can increase soil phosphorus to excessive levels. A soil test program is a tool to help avoid excessive phosphorus applications, and apply biosolids phosphorus to sites where it is most needed.
Prior to initial applications, analyze the surface soil (0 to 1 foot) for pollutants: total arsenic, cadmium, copper, lead, mercury, nickel, molybdenum, selenium and zinc.
It is not necessary to monitor soils during the active life of a site for the pollutants regulated under the WAC 173-308-160 (Biosolids Pollutant Limits). If needed, cumulative trace element amounts applied at a site can be calculated from records of biosolids application rate and biosolids trace element analyses. The small changes in soil concentration of pollutants are usually obscured by variability in sampling and analysis.
As an example consider zinc. Background soil zinc levels for most soils are about 50 mg/kg. A site where biosolids having 1000 mg/kg zinc is applied at 5 dry tons per acre and incorporated to a depth of six inches. This application adds 10 pounds of zinc per acre, and increases soil zinc concentration by about 5 mg/kg. It is almost impossible to detect a change of soil zinc from 50 to 55 mg/kg, because of sampling and analysis variability.
The samples you collect must be representative of the medium you are monitoring (for example, a stream at a specific location, or soil in a field unit). To obtain a representative soil sample you need to define the sample area, and then collect 15 or more subsamples on a grid or in a more random pattern, avoiding atypical areas. The samples are then combined and fully homogenized to represent the field unit. A representative ground water sample may come from a single well, but the well needs to be pumped for a specified period of time to ensure collecting water that represents the aquifer at that location. Surface water samples must be paired; up and downstream samples must be taken at the same time to provide a basis for comparison.
Sampling tools must be capable of collecting a representative sample without contamination. For example, a representative sample for a 0 to 12-inch soil depth has the same amount of soil from the soil surface (0 to 6 inches) as from the bottom of the sampling depth (6 to 12 inches). A shovel is not an acceptable sampling tool, because it will skew the sample toward the shallower portion of the depth. Use a cylindrical soil probe, which is designed to collect a soil sample that is uniform with depth.
The timing of a soil test is the key to its usefulness. Most release of available nitrate from biosolids occurs within a few weeks of application, unless soils are cold or dry. The available nitrate is then taken up by crops, with the most rapid uptake occurring in mid-season when the crops put on the most growth. As crops mature, they redistribute nitrate internally and the uptake of soil nitrogen declines rapidly. At this point both the release of nitrate by biosolids and the uptake of nitrate by plants is low. West of the Cascades, unused nitrate will remain in the soil until it is leached or denitrified by fall and winter precipitation. East of the Cascades, where precipitation is low, much of the soil nitrate will carry over to the next cropping season, although it will leach deeper into the profile.
It is important to collect samples during the window between uptake and leaching. Generally the best sampling time is from mid August to late September. Sample annual crops as soon as possible after harvest. For dryland sites harvested every two years, sample only in crop harvest years. Modify the biosolids application program based on test results.
In many cases sampling the 0 to 12-inch depth is enough, but you will need to sample deeper in some situations. Basing interpretations on a shallow (12-inch) sample assumes that most of the nitrate is in the 0 to 12-inch depth, or that the shallow sample is a good predictor of nitrate in the rest of the profile. The assumptions will not be valid if fall and winter rains leach nitrate deeper into the soil profile. If you delay collecting post-harvest samples until October in most areas west of the Cascades, you should sample in 12-inch increments at least to the 2-ft depth. Sampling later in the fall is not recommended because typical winter rainfall will leach nitrate out of the profile. Even if you collect samples before much leaching occurs, it may be worthwhile to sample to 2 or 3 feet, to see how the distribution of nitrate in the profile compares with assumptions. This is most useful when a field has not been sampled previously below the 0 to12-inch depth.
In dryland areas east of the Cascades, winter precipitation leaches nitrate into the soil profile, but often much of the nitrate remains in the root zone, available for the next crop. In these cases sample the entire root zone (typically 3 to 6 feet) in 12-inch increments to determine pre-plant nitrate.
Collecting SamplesPreparing Containers
Sampling equipment must be cleaned before it can be used. Specific procedures vary depending on which tests will be run. The water, detergents, and acids used during washing should be free of contaminants.
For nutrient and trace element analyses of liquid samples, add appropriate chemical preservatives to the sample containers before use. Chemical preservatives cannot be used with dewatered, composted, or dried biosolids because they will not mix well enough. For all types of biosolids, refrigeration at 4°C (40°F) is an acceptable preservation method.
Systematically label all sample containers before using them. An example of a label is shown above.
Before shipment, fill sample containers no more than three-quarters full. Check that the containers are completely labeled. To reduce biological activity and the risk of liquid samples exploding, use refrigeration or chemical preservatives.
Where existing data is inadequate to make sound land application decisions, more intensive initial sampling may be needed to develop a baseline of biosolids quality. This initial sampling can be used to estimate variability in biosolids quality, to improve the ability to identify outlier values, and to develop an efficient sampling program.
Example Baseline Sampling Program
A typical short-term baseline sampling program could include seven biosolids samples collected over a two week period, analyzed for nutrients and pathogen reduction. This approach is typical for sampling for pathogen indicators. Using this sampling method for nutrients allows for estimation of means, standard deviations, and developing a protocol for routine sampling. This is especially important for nitrogen, since confidence in biosolids application rates depends in part on confidence in nitrogen analysis values.
Analyze trace elements in a composite of the seven samples. If any of the trace element levels in the composite samples are close to Table 1 or Table 3 limits, analyze more individual samples to estimate means, standard deviations, and determine the appropriate number of samples to collect on a routine basis.
Repeating the two week sampling event several times during the year will help you gain an understanding of longer term variability, and help you choose appropriate sampling times. This sampling scheme allows rapid collection of baseline data, while keeping analytical costs at a reasonable level. If further cost savings are needed for small plants, pathogen indicator samples could be collected only at the times of year when pathogen reduction conditions were the poorest, to determine if Class A or B standards are met during the most adverse conditions.
Chill all samples at 4°C during compositing and holding. An ice-water bath chills samples more quickly than a refrigerator or ice. Collect the same weight or volume at each time or location of sampling unless you are using the weighted average composite sampling method.
Composite sampling can be done over time, such as collecting periodic samples from a liquid stream or belt press during a work shift or a day. When biosolids are sampled at the end of a flowing process (for example, after dewatering, or liquid from a pipe), it is usually best to take a composite sample over time.
Where sources of unequal size are being mixed into a composite (for example, piles of different size or tanks of different volume), use a weighted average composite sample demonstrated in the diagram below. For this type of sample, collect a weight (or volume) of sample from each source in proportion to the weight (or volume) of the pile or tank.
For biosolids in a drying bed or compost pile, the composite sample consists of grab samples collected at different locations. Collect at least five to ten grab samples per composite, using a grid system for locating sampling points. Sample the deeper parts of the pile more intensively, since they contain a larger volume of material per unit area. Collect the same size sample from each location in the grid. You may want to composite some, or all of the samples collected from the grid. Biosolids quality in lagoons can vary with depth, and composite sampling should be done over depth as well as area.
DocumentationRecord Sampling Event
Record the sampling activity on a chain of custody document (provide by you laboratory) and in a bound logbook. In the logbook, note any conditions at the time of sampling that may affect analysis results. The chain of custody document identifies the handling history from the time of collection until analysis, showing who handled the sample and whether holding time requirements were met.
Convert all analyses to dry weight basis. If your analyses are on an as-is (wet weight) basis, divide each analysis by the percent solids in the biosolids and multiply the result by 100. Enter the analytical values into a historical database for your treatment plant.
Identify any values that are questionable, based on historical results, and try to determine the cause of these results. By reviewing each of the steps in the sample collection and analysis process, you can check whether they were properly conducted. Samples can be contaminated by dirty containers, dirty sampling equipment or in the laboratory. If necessary, repeat the analysis.
Finally, you should perform statistical tests to verify the quality of the data. This should include mean, standard deviation, and coefficient of variation. The statistical methods should be consistent for each sample collection date. Statistical results are valuable for justifying revisions in sampling protocol and identifying outlier values.
Calculating Geometric Means
To demonstrate that biosolids meet Class B pathogen reduction standards, biosolids managers must document that fecal coliform densities are less than 2,000,000 most probable number (MPN) or (less than) 2,000,000 colony-forming units per gram of total solids.
This standard is met by calculating the geometric mean of seven representative samples. Fecal coliform levels can vary by an order of magnitude or more from one sample to the next. A geometric mean is the most reasonable representation of coliform levels when such variability occurs.
To calculate arithmetic means (averages) begin with a set of measurements (2, 4, and 8), add them (sum = 14), and divide by the number of values in the set (14 / 3), giving a mean value of 4.67.
A geometric mean is not much more difficult. It involves determining an arithmetic mean (just like above) for the logarithms of your sample values. For those who are familiar with them, logarithms make it easier to work with very large (or very small) numbers. Logarithms are simply exponents. If we express a number in exponential form (100 = 102), then 2 is the common (base 10) log of 100. Similarly the log of 1000 is 3 (1000 = 103), and the log of 10,000 is 4 (10000 = 104). The log of 66,271 is 4.8213 (66,271 = 104.8213). A scientific calculator or computer spreadsheet will convert a number to its common logarithm.
To calculate a geometric mean, convert each value to its common logarithm and add the logarithms together. Next, compute the mean by dividing the sum of the logarithms by the number of samples. The result is an average logarithmic value.
The last step is to convert the logarithmic value back to the original form. This is called the anti-logarithm and is the reverse of the conversion to a logarithm. Recall from above that the logarithm of 100 is 2. The anti-logarithm of 2 is 100. Scientific calculators or computer spreadsheets will determine anti-logarithms, using an anti-log, power, or exponent function.
Copyright © Washington State Department of Ecology|
Privacy Notice | Site Info | Accessibility | Contact the web team |