1 Topic by stuart.warrington

  • stuart.warrington
  • Member
  • Offline
  • Registered: 15-12-2013
  • Posts: 22

Topic: Quedius (Quedius) balticus - Huge number of data entry errors

Quedius (Quedius) balticus NHMSYS0001719922 is a very rare rove beetle (Coleoptera) - found in about 3 places in the UK & all high quality fen sites - hence its RDB1 status.

BUT the NBN Gateway shows hundreds of records. This is due to a data entry error made when uploading spreadsheets into Recorder - unless care is taken - the records of the common Quedius (Quedius) molochinus NHMSYS0001719926 will match to Q.balticus.
These are then sent to the NBN with the error.

How do we get these hundreds of errors removed from the Gateway?

2 Reply by charliebarnes (edited by charliebarnes 29-07-2014 11:54:52)

  • charliebarnes
  • Veteran
  • Offline
  • From: Greater Lincolnshire
  • Registered: 07-09-2008
  • Posts: 1,396

Re: Quedius (Quedius) balticus - Huge number of data entry errors

By getting each of the data providers to correct it - the NBN may be able to bulk-contact-and-correct, but that would still require the original data providers to correct it otherwise the error would be re-introduced.

I think the cause of the mismatch has been corrected (automatching not taking into account subgeneric names), so this type of problem shouldn't crop up as much now.

Charlie Barnes
Information Officer
Greater Lincolnshire Nature Partnership

charliebarnes's Website

3 Reply by Syrphus

  • Syrphus
  • Veteran
  • Offline
  • From: Strathpeffer, Highland
  • Registered: 16-12-2008
  • Posts: 535

Re: Quedius (Quedius) balticus - Huge number of data entry errors

I discover I am custodian of three of these, Stuart, from Highland and Orkney.

Can I assume that they are unlikely to be balticus?

I will be updating the set soon, so will deal with any corrections/quarantining before then.

Murdo

Syrphus's Website

4 Reply by TeresaF

  • TeresaF
  • Veteran
  • Offline
  • From: BTO
  • Registered: 12-05-2008
  • Posts: 172

Re: Quedius (Quedius) balticus - Huge number of data entry errors

Thank you for pointing this out. Cumbria Biodiversity Data Centre had four records and looking at them we are sure they were because of this error, so I've just updated them and they should be gone from the Gateway on our next data refresh, which I hope will be soon. As Charlies says each data provider will need to be made aware, both so the problem is not reintroduced and so the records are not used incorrectly at source.

(The records looked very odd in Recorder - Q. molochinus as the common name and Q. balticus as the scientific name for some, Q molochinus for scientific name and common name for some and yet being pulled out as Q balticus - anyway, re-selecting Q (Q) molochinus from the BEETLES list seems to have fixed it.)

The provider list is:
Biological Records Centre
Centre for Environmental Data and Recording
Cumbria Biodiversity Data Centre
Environmental Records Information Centre North East
Greenspace Information for Greater London
Highland Biological Recording Group
Kent & Medway Biological Records Centre
Merseyside BioBank
National Trust
Natural England
Norfolk Biodiversity Information Service
North East Scotland Biological Records Centre
RSPB
Scottish Wildlife Trust
Sheffield Biological Records Centre
South East Wales Biodiversity Records Centre
Suffolk Biological Records Centre
Sussex Biodiversity Record Centre
Warwickshire Biological Records Centre
Wiltshire and Swindon Biological Records Centre
Worcestershire Biological Records Centre

Stuart, you could email each directly or you could contact the ALERC national coordinator tom.hunt@alerc.org.uk and ask him to contact the relevant LRCs for you to make them aware the determinations need checking, leaving you with just a few organisations such as RSPB and Scottish Wildlife Trust?

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

5 Reply by Tony Witts

  • Tony Witts
  • Member
  • Offline
  • From: Kent & Medway BRC
  • Registered: 08-04-2009
  • Posts: 21

Re: Quedius (Quedius) balticus - Huge number of data entry errors

We have 4 records which originated from the RSPB's 'Bringing Reedbeds to Life' project, which we downloaded from the Gateway.  The dataset also includes a single record of Q. molochinus.  The exec summary of the project cites Q. balticus as one of the high-lights of the survey.  All records are for Stodmarsh NNR. I'm not sure whether they are correct or not!

6 Reply by stuart.warrington

  • stuart.warrington
  • Member
  • Offline
  • Registered: 15-12-2013
  • Posts: 22

Re: Quedius (Quedius) balticus - Huge number of data entry errors

Thankyou everyone for the replies. Good to see some data being updated -
Tony - can you check carefully who recorded Quedius balticus from the report (& checked the identification, the species has to be dissected to identify it) ? If its a top Coleopterist, then the Quedius balticus should be correct for Stodmarsh and it is a really good discovery.
Murdo - Q. balticus has never been found in Scotland (yet) so its most unlikely unless a specialist fen survey has been undertaken?

7 Reply by BDeed

  • BDeed
  • Veteran
  • Offline
  • From: Bolton
  • Registered: 01-03-2011
  • Posts: 451

Re: Quedius (Quedius) balticus - Huge number of data entry errors

I have just been made aware of this thread by a concerned recorder (cheers Steve!). We hold two records which have mismatched. interestingly, the name on the record is Q. molochinus so this wasn't caused on input. In any case the records have now been re-matched against the Beetles checklist and appear to be OK. They will disappear in the next update.

Perhaps the remaining data providers should be contacted? If i had known about this i could have it sorted by now!

Natural History & Biodiversity Data Enthusiast

8 Reply by martinsanford

  • martinsanford
  • Veteran
  • Offline
  • From: Suffolk
  • Registered: 02-11-2007
  • Posts: 52

Re: Quedius (Quedius) balticus - Huge number of data entry errors

Ditto for Suffolk records

martinsanford's Website

9 Reply by mhorlock

  • mhorlock
  • Member
  • Offline
  • From: Norwich
  • Registered: 07-01-2009
  • Posts: 17

Re: Quedius (Quedius) balticus - Huge number of data entry errors

Hi

I've looked at the Norfolk records. All are from fen sites and most seem to originate from the NE East Anglian Fen Invertebrate Survey via our County Recorder. The rest come via Broads Authority and RSPB but are by a known recorders one of which worked on the aforementioned survey.

I welcome your thoughts on whether I should be amending these or not.

Regards
Martin

Martin Horlock
Norfolk Biodiversity information Service

mhorlock's Website

10 Reply by Tony Witts

  • Tony Witts
  • Member
  • Offline
  • From: Kent & Medway BRC
  • Registered: 08-04-2009
  • Posts: 21

Re: Quedius (Quedius) balticus - Huge number of data entry errors

Hi Stuart,
The 'Bringing Reedbeds to Life' pit fall trap coleoptera were detted by Mark Telfer. So all good there.

Cheers
Tony

11 Reply by charlesroper

  • charlesroper
  • charlesroper
  • Digital Development Manager
  • Offline
  • From: Field Studies Council
  • Registered: 26-03-2006
  • Posts: 906

Re: Quedius (Quedius) balticus - Huge number of data entry errors

We had three records exhibiting this problem which have now been corrected. It seems to be a dictionary issue (sort of). If you search for Quedius molochinus in Recorder in the BEETLES checklist, it returns Quedius molochinus sensu Last, 1952 non (Gravenhorst, 1806) [Spp]:

[img]http://i.imgur.com/9xkOsSR.png[/img]

But rather than being a synonym of Q. (Quedius) molochinus, it seems to be a synonym of Q. (Quedius) balticus:

[img]http://i.imgur.com/AUKbOkn.png[/img]

Here's the NBNG entry:

https://data.nbn.org.uk/Taxa/NHMSYS0001719964

According to Duff's Checklist of Beetles of the British Isles 2nd Edition (2012) molochinus sensu Last, 1952 non (Gravenhorst, 1806) is listed as a synonym of balticus Korge, 1960. So according to that the dictionary is correct, but the problem is that it's currently much too easy for non-beetle experts (i.e., the vast majority of users) to inadvertently choose the wrong thing.

I can confirm that redetermining the record as Quedius (Quedius) molochinus does fix the issue. I think it would also be useful to fix the dictionary somehow to make it clear that Quedius molochinus sensu is not the same thing as Quedius (Quedius) molochinus. Or, at least, return Quedius (Quedius) molochinus when searching for "Quedius m..."

Charles Roper
Digital Development Manager | Field Studies Council
http://www.field-studies-council.org | https://twitter.com/charlesroper | https://twitter.com/fsc_digital

charlesroper's Website

12 Reply by MikeWeideli

  • MikeWeideli
  • Administrator
  • Offline
  • From: Littlefield Consultancy
  • Registered: 05-07-2006
  • Posts: 2,645

Re: Quedius (Quedius) balticus - Huge number of data entry errors

As far as  Import Wizard matching is concerned the sub genus is now ignored in matching so automatic matching will find more than one option and fail force a manual decision.  Saved matches against the preferred  list will have been cleared in version 6.18 so should not leave a problem.  I will give some thought to the best way of removing other saved matches for this species.

Recorder now  displays the attributes when searching for taxa which it didn't before, which should help.

Mike Weideli

MikeWeideli's Website

13 Reply by Gordon Barker

  • Gordon Barker
  • Veteran
  • Offline
  • Registered: 15-05-2006
  • Posts: 344

Re: Quedius (Quedius) balticus - Huge number of data entry errors

Is there a way to identify all of the records that might have been affected by this issue? I have used a fairly crude method to put together a list of over 400 species comprising ~1300 records which might have a problem. Probably a lot of false positives in that list, only about 60 of the species have subgenera but the others may need checking

Gordon Barker
Biological Survey Data Manager
National Trust

14 Reply by DaveSlade (edited by DaveSlade 27-11-2014 16:30:02)

  • DaveSlade
  • Veteran
  • Offline
  • Registered: 15-05-2006
  • Posts: 332

Re: Quedius (Quedius) balticus - Huge number of data entry errors

I've corrected the records we have at SEWBReC for this species, and like Gordon would welcome a routine that would help identify these errors rather than relying on people spotting them on Gateway (as happened earlier this month with a rare weevil that had been entered matched to the synonym (Sitona puberulus instead of Coelositona cambricus).

Our gateway data will be refreshed in due course, when time permits.

Dave Slade
Senior IT & Records Officer,
15 Talbot Road, Talbot Green, Pontyclun, CF72 8AD
www.sewbrec.org.uk, www.sewbrecord.org.uk

15 Reply by Gordon Barker

  • Gordon Barker
  • Veteran
  • Offline
  • Registered: 15-05-2006
  • Posts: 344

Re: Quedius (Quedius) balticus - Huge number of data entry errors

Returning to this question, I thought there might be a solution by querying the IWMatched_Species table, but that didn't find one error that I know is definitely in there. I assume now that this would be because the error was down to a system-generated match, not a user-generated one. Does that sound right?

Has anyone got any other suggestions to track down the dodgy matches?

Gordon Barker
Biological Survey Data Manager
National Trust

16 Reply by Gordon Barker

  • Gordon Barker
  • Veteran
  • Offline
  • Registered: 15-05-2006
  • Posts: 344

Re: Quedius (Quedius) balticus - Huge number of data entry errors

I have attempted a query that I think works. Any comments welcome

    SELECT     TAXON_DETERMINATION.TAXON_OCCURRENCE_KEY, INDEX_TAXON_NAME.ACTUAL_NAME AS AN, INDEX_TAXON_NAME.COMMON_NAME,
                          INDEX_TAXON_NAME.PREFERRED_NAME, LEFT(INDEX_TAXON_NAME.PREFERRED_NAME, CHARINDEX(' ', INDEX_TAXON_NAME.ACTUAL_NAME)) AS 'LA',
                          LEFT(INDEX_TAXON_NAME.COMMON_NAME, CHARINDEX(' ', INDEX_TAXON_NAME.COMMON_NAME)) AS 'LC', RIGHT(INDEX_TAXON_NAME.PREFERRED_NAME,
                          LEN(INDEX_TAXON_NAME.PREFERRED_NAME) - CHARINDEX(' ', INDEX_TAXON_NAME.PREFERRED_NAME)) AS 'RA', RIGHT(INDEX_TAXON_NAME.COMMON_NAME,
                          LEN(INDEX_TAXON_NAME.COMMON_NAME) - CHARINDEX(' ', INDEX_TAXON_NAME.COMMON_NAME)) AS 'RC'
    FROM         TAXON_DETERMINATION INNER JOIN
                          INDEX_TAXON_NAME ON TAXON_DETERMINATION.TAXON_LIST_ITEM_KEY = INDEX_TAXON_NAME.TAXON_LIST_ITEM_KEY
    WHERE LEFT(INDEX_TAXON_NAME.PREFERRED_NAME,charindex(' ',INDEX_TAXON_NAME.PREFERRED_NAME))=LEFT(INDEX_TAXON_NAME.COMMON_NAME,charindex(' ',INDEX_TAXON_NAME.COMMON_NAME))
    AND RIGHT (INDEX_TAXON_NAME.PREFERRED_NAME,(LEN(INDEX_TAXON_NAME.COMMON_NAME)-charindex(' ',INDEX_TAXON_NAME.COMMON_NAME)))<> RIGHT(INDEX_TAXON_NAME.COMMON_NAME,(LEN(INDEX_TAXON_NAME.COMMON_NAME)-charindex(' ',INDEX_TAXON_NAME.COMMON_NAME)))
   

and LEFT(INDEX_TAXON_NAME.PREFERRED_NAME,charindex(' ',INDEX_TAXON_NAME.PREFERRED_NAME))<>''

order by AN

Gordon Barker
Biological Survey Data Manager
National Trust