5. Reporting, records, assessment and interpretation of results from an environmental monitoring programme
How to report on the results of your environmental monitoring to meet requirements, ensure best use of the data, and early indications of changes in the environment can be identified and acted upon.
Reports for verified environmental monitoring results should be produced in a timely manner (as defined in the Compilation of Environment Agency Requirements (CEAR) document for operators or work specification for regulator programmes) so that best use can be made of the data and early indications of changes in the environment can be identified and acted upon.
Reports should include or reference information on the methods used and the quality assurance process.
Results should be presented with information on units, uncertainties and detection limits. It should be clearly stated whether the results are decay corrected to the date of sampling.
Notification of unusual results
Procedures should be in place for the early notification of unusual results, particularly those that are unusually high. It may help if a table of ‘notification levels’ is maintained to trigger this consistently. There are various stages where this can occur:
- directly following sampling in the field, if dose rate monitoring is being undertaken – as would be good practice for samples coming from areas of potentially high activity
- following receipt of samples at the laboratory where dose rate readings should be taken
- directly following analysis where judgement should be used to determine whether they are significantly above normal environmental levels. This judgement should not only be based on reviewing the actual results, but also consider knowledge of other factors such as variation in sediment grain size and characteristics at a particular location
Appropriate pre-determined action, trigger or notification levels may be set up to aid in this process.
Unusual results should be investigated to rule out analytical issues.
Further context for results can be gained through considering information on the environment when samples were taken, for example, changes in farming practice, erosion or deposition leading to re-distribution of sediment. Or whether the radionuclides being detected could be the result of other releases (for example I-131 from nuclear medicine).
Record keeping
Records should be kept for a defined period (as specified by the regulator), the information should be traceable and retrievable, taking account of changing storage technology.
Results should be assessed in the context of the objectives for which the monitoring was designed.
Comparisons should be made with applicable standards and other available data.
For the representative person dose objective (A) the monitoring programme results should be compared to Generalised Derived Limits (GDLs), or a percentage of them (for example 10%). It may also be appropriate to undertake retrospective dose assessments with the data and compare with dose limits. Similarly for the total impact on wildlife objective (B) it may also be appropriate to undertake retrospective dose assessment and compare with appropriate dose rate criteria.
For other objectives, for example, providing assurance (Objective C) or assessing background (Objective E), comparison with previous results, action levels (for example 3 or 4 standard deviations) and natural backgrounds may be most appropriate.
Operators may set criteria for determining what might indicate an abnormal release and what subsequent actions should be taken.
Checks on the internal consistency of results can also be undertaken, where appropriate and taking account of uncertainties, for example:
- the sum of the alpha or beta or gamma emitting radionuclides can be compared with total alpha or total beta measurements
- consistency of results within decay series can be checked
Data can also be compared with other data sets such as those published in the Radioactivity in food and the environment (RIFE) report.
Proportional assessment
The assessed level of site impact should be taken into consideration for the degree of interpretation and assessment required.
For higher impact sites it may be appropriate to utilise the various statistical approaches available for further assessing results, for example:
- looking at averages and standard deviations
- undertaking trend analysis (looking for rising and falling trends) - this could be visual or with software
- assessing whether the data are censored (considering whether limit of detection data are truncating the data set, so the true distribution is unknown)
- using a box and whisker approach (that is, graphically depicting the data through their five-number summaries: sample minimum, lower quartile (Q1), median (Q2), upper quartile (Q3) and sample maximum)
- looking for the first arrival of a finite value (rather than limit of detection) in a data set
For lower impact sites, simple inspection of tabular or graphical presentation of monitoring results may be sufficient.