Sample and samplerelationship

Support
1

I am not sure how to report in case of 2 or more grab samples done at the same site throughout the day and then pooled by the technician before sending to the laboratory.

Is it correct like this:

  • report each grab sample separately in sample table.
  • report the final pooled sample in sample table with collNum = nb of grab sample but since each grab sample comes from the same site the variable pooled should be false (if it had been from different sites then pooled would have been true, am I right?)
  • in table samplerelationship, the pooled sample is child towards each grab sample (three line entries then)
  • are the grab samples colocated between each others in table samplerelationship?

Or is it simply allowed to report only one sample corresponding to the pooled one in the sample table (and then none of the grab ones) with repType = colocated?

Thanks for the help.

2

Hi @wastewater_belgium ! Sorry for the delayed reply from our end. This is a great question, and definitely one we need to add as an illustrative example in our documentation.

The short answer is that I think the way you describe reporting 2 grab samples from the same site on the same day, along with a combined sample, is correct.

  1. You have 3 rows in the sample table, 1 row each for the grab samples, and then 1 row for the combined sample.
  2. The combined sample has a collNum = 2, pooled = FALSE for the reasons you mention.
  3. in the sampleRelationships table, you would have two rows to state that each grab is a parent to the combined sample/that the combined sample is a child of each grab sample. You could, however, use a third row to specify that that the two grab samples are colocated. This third row is more optional, however. But encouraged!

The longer answer is that while this is how we envision recording such an example, we also recognize that it can take extra time to record each sample’s details in this way. If you were comfortable sacrificing some of that more granular data and metadata for the sake of efficiency, then the second option you suggest (of only creating one samples table entry for the combined sample) would also be correct. You could also take the above enumerated approach and specify that each of the grab samples were reType = colocated as well.

We tried to be robust while also flexible while designing the ODM, so there are often situations like this where there are multiple “right” answers. Ultimately, the answer here is also influenced by what works best for your lab and your group.

3

This topic was automatically closed 24 hours after the last reply. New replies are no longer allowed.