Topic: Simple metric?

Hi Sophie - a few of us have been thinking about useful metrics for seeing how well the NBNA is progressing, instead of the oft-quoted (but unhelpful) "number of records". We're interested in map completeness, and are looking at this in various ways.

However, it occurs to us that a very simple metric, and one that could be easy to produce, could be the count of distinct combinations of 10km square and species. This simple figure could then be assessed over time to see how we're doing.

Would it be possible to run a quick query on the back-end database please, along the lines of

select count (distinct Taxon Concept ID, OSGR 10km) from TABLE;

(or whatever the equivalent is). Be interesting to jot this down, then revisit it in a year's time.



Re: Simple metric?

Hi Andy,
Yes, of course. You would like the number of distinct taxon concepts in each 10 km square? It should be possible to do this through the web services that the Atlas has (https://api.nbnatlas.org). I use R to access and query the data. I can do this for you and give you the R code (in case you use R).

Thanks, Sophie


Re: Simple metric?

Thanks Sophie. That would be good. I was actually just thinking of returning a single number (i.e. the number of 10km x species combinations for the whole UK*), but a number per square would also be nice.

[*UK - might be easier to start with GB, as the 10km grids might not come out so easily for NI/Ireland. GB would be fine for starters]

I don't use R myself but I have many colleagues who could get me started. So yes, the code would be great thanks.



Re: Simple metric?

I read it properly this time. No problem, I will write the script to do both.

thanks, Sophie


Re: Simple metric?

The R code would be great. Is it possible to group the results by the taxon group?
insect - beetle (Coleoptera): 700,000
bird: 300,000
flowering plant: 100,000
all groups: 2,500,000

I thought taxon groups were stored in the Atlas but now I can't find them, just order etc. in the classification tab and records pages, which although obviously taxonomically important information are slightly less useful for things like this because the informal groups (imperfectly but more usefully) rather reflect the way biological recording is organised. The groups are in the UKSI  - see dropdown at http://www.nhm.ac.uk/our-science/data/u … ndex.html.


Teresa Frost | Wetland Bird Survey National Organiser | BTO
Other hat  | National Forum for Biological Recording Council
(Old hats  | NBN Board, ALERC Board, CBDC, KMBRC)


Re: Simple metric?

We have those groups in the Atlas but I am not sure that they are available in the web services. I'll see what's possible and let you know,

cheers, Sophie


Re: Simple metric?

Great, thanks very much.

On the number of taxa per hectad - hopefully mapping this in the Atlas itself for datasets (and the whole atlas) is on a feature development list somewhere along with number of records per hectad, as a Gateway feature much-missed (missed, anyway!)

https://forums.nbn.org.uk/viewtopic.php … 965#p26965

Teresa Frost | Wetland Bird Survey National Organiser | BTO
Other hat  | National Forum for Biological Recording Council
(Old hats  | NBN Board, ALERC Board, CBDC, KMBRC)


Re: Simple metric?

Going back to Teresa's question on taxon group, can this field be included in downloads from NBNA? (Either is there a way to do it that I haven't worked out yet or if not can it be added to the standard record downloads?

Gordon Barker
Biological Survey Data Manager
National Trust


Re: Simple metric?

Hi Gordon,
At the moment the taxon groups that Teresa mentions are not linked to occurrence records in the Atlas in a way that means that they can be added to the downloads. It is only possible to filter species using the groups.

We have a project to implement the taxon groups from the UKSI properly within the Atlas, rather than use the species groups that came with the ALA. This is a priority for us and I will let you know as soon as I have a timescale for it,

Best wishes, Sophie