Laboratory analysis
3.1 Laboratory quality management
Laboratories participating in the analysis of soil and land applied materials for the Nutrient Management Act (NMA) are required to have a sound quality management program. Quality Management (QM) is that aspect of the over-all management function that determines and implements the quality policy. International Standard ISO/IEC 17025 outlines management and technical requirements for implementing a laboratory quality management system.
3.1.1 Laboratory accreditation requirement
Laboratories analyzing soil for available nutrients, as required by the Nutrient Management Act (NMA), must be accredited for the applicable nutrient tests by the Ontario Ministry of Agriculture, Food and Rural Affairs under the OMAFRA Agronomic Test Accreditation Program.
Laboratories analyzing land applied materials for nutrients as required by the NMA must be accredited by the Ontario Ministry of Agriculture, Food and Rural Affairs under the OMAFRA Agronomic Accreditation Program, or by a body which accredits methods to ISO/IEC 17025 standards for analytical laboratories (such as the Standards Council of Canada, and Canadian Association for Laboratory Accreditation Inc.)
Laboratories analyzing soil or land applied materials for metals and the pathogens, as required by the NMA, must be accredited by a body which accredits methods to ISO/IEC 17025 standards for analytical laboratories (such as the Standards Council of Canada, or Canadian Association for Laboratory Accreditation Inc.)
3.2 Laboratory method
All laboratories participating in the analysis of soil and land applied materials for NMA activities must have a formal written method used for the analysis. Bench procedures must be documented in sufficient detail to ensure uniform application and must be readily available to technical staff.
3.2.1 Method summary
A summary of the method used for such analysis may be required by the Ministry to review data. It will assist the Ministry in evaluating if the laboratory method/performance data is in compliance with the data quality requirements of this program (Section 4).
A method summary should contain the following information as a minimum:
- Method Used, e.g., EPA 5030, MOE/LaSB E3394 or your laboratory Reference Method.
- Method Principle - Brief description of sample preparation and instrumentation.
- Sample preservation if required.
- Sample storage temperature.
- Accreditation (laboratory/method) - Type of accreditation and name of accrediting body.
- Method performance characteristics - Provide such information in a tabular form. The example is given in table 3-1.
| Analyte | Method Detection Limit (MDL) | Bias(%) | Precision Repeatability % RSD |
Precision Reproducibility % RSD |
Method Working Range |
|---|---|---|---|---|---|
| - | - | - | - | - | - |
| - | - | - | - | - | - |
| - | - | - | - | - | - |
| - | - | - | - | - | - |
| - | - | - | - | - | - |
Provide the name of material(s), e.g., in-house spiked matrix blank, CRM, other, and number of determinations used for this study.
3.3 Method Detection Limit (MDL)
The method detection limit is a statistically defined method attribute. Measured results falling at or above this point are interpreted to indicate the presence of an analyte in the sample with a specified probability—usually greater than 99%—and assumes that sources of error in identification or biases in measurement are known and controlled.
3.3.1 Procedure for MDL Determination
Take a minimum of eight aliquots of the sample to be used to calculate the method detection limit and process each through the entire analytical method.
If a blank measurement is required to calculate the measured level of analyte, obtain a separate blank measurement for each sample aliquot analyzed.
Calculate a result (x) for each sample/blank pair.
Calculate the standard deviation (S) of the replicate measurements as follows:
S = √[∑(xi - x̄)2] ÷ (n-1)
where:
xi = the analytical results in the final method reporting units for the n replicate aliquots (i = 1 to 8)
x̄ = the average of the n replicate measurements
An alternative is to use historic within run duplicate analysis data and calculate the standard deviation (S) of the duplicate measurements as follows. This is suggested for soil samples.
S = √[∑(x1 - x2)i2] ÷ (2n)
where:
x1, x2 = the two duplicate results for each of the n replicate pairs (minimum n = 40)
Compute the MDL as follows:
MDL = t(n-1, a = 0.01 ) S
where: t(n-1, a = 0.01 ) is the Student's value appropriate for a 99% confidence level given the degrees of freedom n-1.
S = the standard deviation as determined above.
| Number of Replicates | Degree of Freedom (n-1) |
t (n-1) |
|---|---|---|
| 7 | 6 | 3.143 |
| 8 | 7 | 2.998 |
| 9 | 8 | 2.897 |
| 10 | 9 | 2.821 |
| 11 | 10 | 2.764 |
| 16 | 15 | 2.603 |
| 21 | 20 | 2.528 |
| 26 | 25 | 2.485 |
| 31 | 30 | 2.457 |
| 8 | 8 | 2.369 |
3.4 Reporting Detection Limit (RDL)
The RDL is typically set at a value equal to 10% of the maximum permissible contaminant concentration (MPCC). However, in a few instances, due to limitation of available analytical technology, RDL may be set a value higher than 10% of MPCC but not exceeding MPCC.
Laboratories must achieve MDL equal or less than the RDL value.
Laboratories must report all results above MDL.
3.5 Precision
Precision is the degree of agreement among independent measurements of the same quantity under specified conditions.
Precision under repeatability conditions (ISO 3534-1, 3.15 and 3.16), and under reproducibility conditions (ISO 3534-1, 3.20 and 3.21) must be established. Control limits for these should be established and maintained as part of the analytical performance criteria.
It is desirable to determine precision at ˜ 10MDL.
The precision requirement for each test is given in Section 4.
3.6 Bias
Bias is the difference between the average value obtained from a large series of test result, and an accepted reference value. Certified reference materials (CRMs), if and when available, should be used to assess bias. If a CRM of exactly the same type of material as the sample is unavailable, a similar CRM may be used. For example, a CRM of plant tissue or biosolids may be used for manure analysis.
CRM is defined as a reference material, accompanied by an authenticated certificate, having for each specified quantity a value, measurement uncertainty and stated metrological traceability chain. Reference material is a material that is sufficiently homogeneous and stable with respect to one or more specified quantities, used for the calibration of a measuring system, or for the assessment of a measurement procedure, or for assigning values and measurement uncertainties to quantities of the same kind for other materials.
For this program, certified reference material(s) is identified under each test (Section 4). Other CRMs may be used, provided they produce data within the allowable range when subjected to the same method principle.
The bias for each test is given in Section 4.
Participation in one or more proficiency testing (PT) programs also demonstrates acceptable method performance.
3.7 Method Working Range
The working range of the method for each analyte must be established and documented in the method. Working range is the range over which the analytical system exhibits a linear or other well established relationship between the amounts of material introduced into the analytical system and the instrument's response.
Measurement should be performed within the working range. If the result is suspected to be outside of the working range, appropriate dilution should be performed.
3.8 Recommended Laboratory QC/QA Procedures
The following are recommended laboratory quality procedures:
Pre-service QC:
- lab ware and reagent blanks
- calibration standards
- alternate source standards to validate in-house standards
- CRM to validate recovery/bias
- instrument detection limits (IDLs) and detector linearity/working curves (minimum of 3 point calibration)
In-service QC:
- baseline drift blanks
- standards
- instrument checks
Run quality QC and QA:
- method blanks (matrix)
- method spiked blanks
- in-house matrix check material
- replicate sample (minimum of one set per run of 30 samples)
- spiked samples (standard addition), if applicable.
Laboratories should maintain records of data to show that the analytical systems were in control at the time of analysis. The results of these quality control and performance-monitoring checks should be control charted, and summaries readily available for inspection.
3.9 Data Acceptance Criteria
The basis for determining the acceptability of laboratory data should include the following:
- Method should be consistent with the principle as given under specific test (Section 4) has been applied.
- The performance characteristics (RDL, Bias, and precision/reproducibility) of a method used for NMA analysis should be within specifications as given under each test (Section 4).
- The results of all applicable quality control samples should be within the acceptable range. See specific tests (Section 4).
- The analytical system should be in a state of statistical control at the time of analysis as demonstrated by appropriate control charts.
3.10 Data Reporting
A laboratory's data-management system should establish and maintain direct links between sample information (such as source, field sample number or code, date and time sampled, tests required), and laboratory information (such as laboratory sample number or code, date and time analyzed, tests performed and identification of the analyst who did the work).
A properly recorded result shall include the test or analyte name or code, the units of measure, the method used for analysis and any qualifying remarks.
Analytical results may be corrected to take into account any positive results of associated method blank for some specific analysis. A method blank result above the method detection limit is normally considered a positive result. The criteria or control limits for blank corrections should be determined by laboratories on the basis of historical data, and these should be documented. Otherwise, data should be reported without correction. If a correction is made, it should be clearly identified and described.
All data should be reported. Data below RDL should have remark as < RDL. Data below MDL should have remark as < MDL.
All data for soil and land applied materials (= 1% solid) should be reported on a dry weight basis. Dry matter content should also be reported.
All data for dilute liquid land applied materials (< 1 % solid) should be reported on a volume basis.