1 Reply by SteveHannah

  • SteveHannah
  • Veteran
  • Offline
  • From: Midlothian
  • Registered: 02-09-2010
  • Posts: 84

Re: Hybrids with incorrect status on export

When I've exported data out of recorder for conversion to GIS, I've found that several hybrid species have "inherited" statuses from one of the parent species.  You can't see this within Recorder as it correctly doesn't give it a species if browsing the taxon dictionary, but it does crop up on export.

For example Tilia x. Europaeus gets a Nationally Scarce status because one of the parent species, Tilia platyphyllos is Nationally Scarce.  Another one is hybrid bluebells (Hyacinthoides non-scripta x. hispanica) which gains the statuses of Hyacinthoides non-scripta.

-----------------------------------------------------------------------
Steve Hannah - GIS and Data Infrastructure Officer
The Wildlife Information Centre

SteveHannah's Website

2 Reply by charliebarnes

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

Re: Hybrids with incorrect status on export

See http://forums.nbn.org.uk/viewtopic.php?id=1838

Oh for a real bug tracking system.

Charlie Barnes
Information Officer
Greater Lincolnshire Nature Partnership

charliebarnes's Website

3 Reply by MikeWeideli

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

Re: Hybrids with incorrect status on export

The Taxon Designation dictionary does not hold levels below species, so the query used to generate Index_Taxon_Designation picks up the levels below. It seems that in some cases thsi is right and in some cases not.  I have suggested before that it would be possible to exclude specific taxon ranks (eg hybrids) while allowing others to be included (eg. sub species). Woudl this help ?

Mike Weideli

MikeWeideli's Website

4 Reply by MikeWeideli

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

Re: Hybrids with incorrect status on export

The following query is provided for someone to try to see if this improves the reporting of designations. It removes the hybids. I don't think it will remove anything which is required.  The normal  sitiuation can be restored by rebuilding the Index_Taxon_Designation table. Feedback on the effect would be appreciated.  Query needs to be run in SQL Server under NBNData.


DELETE FROM Index_Taxon_Designation
FROM Index_Taxon_Designation ITD
INNER JOIN TAXON_LIST_ITEM TLI
ON TLI.TAXON_LIST_ITEM_KEY = ITD.TAXON_LIST_ITEM_KEY
WHERE TLI.TAXON_RANK_KEY  IN('NBNSYS0000000029','NBNSYS0000000030','NHMSYS0000808785','NHMSYS0000377392')

Mike Weideli

MikeWeideli's Website

5 Reply by AndyFoy (edited by AndyFoy 22-07-2011 08:46:45)

  • AndyFoy
  • Consultant
  • Offline
  • Registered: 09-08-2006
  • Posts: 219

Re: Hybrids with incorrect status on export

Hi Mike,

Thanks for this, it works for me although I needed to change the SQL slightly in MS Access to run it.  I ended up with:

DELETE ITD.*
FROM Index_Taxon_Designation ITD
INNER JOIN TAXON_LIST_ITEM TLI
ON ITD.Taxon_List_Item_Key = TLI.TAXON_LIST_ITEM_KEY
WHERE TLI.TAXON_RANK_KEY In ('NBNSYS0000000029','NBNSYS0000000030','NHMSYS0000808785','NHMSYS0000377392')

I haven't run this in on our live system yet as I'd like to investigate any knock-on effects first.  Having discussed the problems with hybrids inheriting designations from parent taxons with John Van Breda I think removing them is the right thing.  Obviously it would much be better if they were never added to the index during the index rebuild by Recorder.

As an interim solution this SQL will work, but the only problem is that if there are hybrids that are specifically designated in their own right they will also be removed.  It would be more useful if the SQL could track back through the taxons and designations to determine which index entries are there in their own right and which are there because they inherited their designations.  John did write some SQL for me to do this but I'm not sure if it would work for everyone.

Cheers,
Andy

Andy Foy
Systems Manager
Greenspace Information for Greater London (GiGL) CIC
www.gigl.org.uk

6 Reply by RobLarge

  • RobLarge
  • RobLarge
  • Veteran
  • Offline
  • From: Wiltshire
  • Registered: 15-07-2008
  • Posts: 812

Re: Hybrids with incorrect status on export

The fault surely lies in the checklists then, since while the hybrid bluebell, for example is clearly genetically a child of the native bluebell, in the strict taxonomic/heirarchical sense used above it is not actually a 'child' in the way that a subspecies is.

If it can be said that the hybrid is a child of one parent then it must be a child of the other also and this would lead to a taxon occurring in two different places in the heirarchy. The hybrid should in most cases occur at the same heirarchical level (i.e. species-level) as both its parents (except in the case of a transgeneric hybrid, when I guess it should occur at species-level within its own trans-genus within the parent family).

I was also going to point out that there may be some hybrids which are designated in their own right, but you got there before me Andy.

Rob Large
Wildlife Sites Officer
Wiltshire & Swindon Biological Records Centre

RobLarge's Website

7 Reply by RobLarge

  • RobLarge
  • RobLarge
  • Veteran
  • Offline
  • From: Wiltshire
  • Registered: 15-07-2008
  • Posts: 812

Re: Hybrids with incorrect status on export

It seems to me that so many of these problems track back to the Recorded 3.3 checklist. Must be time to think of a way of doing away with the old thing.

Rob Large
Wildlife Sites Officer
Wiltshire & Swindon Biological Records Centre

RobLarge's Website

8 Reply by AndyFoy

  • AndyFoy
  • Consultant
  • Offline
  • Registered: 09-08-2006
  • Posts: 219

Re: Hybrids with incorrect status on export

Hi Rob,

I have to agree ... it is the species lists that are ultimately 'at fault' for listing hybrids as children of one or both of their parent species.  But I guess until that is resolved we'll have to agree an interim solution.

Mike's suggestion of excluding all hybrids is a nice and simple solution.  John has also suggested another solution but it would require more development I think - if I understood him correctly it would be to add a flag to each designation to say if it also applies to any child taxons.

A combination of the two solutions would give the greatest control so that one flag could indicate if designations apply to child taxons (e.g. sub-species when designated at the species level or species and sub-species if designated at the genus level) or just strictly to the designated taxon itself.  Another flag (or another value on the same flag) could indicate if the designation also applied to hybrid species - this would be unlikely for species level designations, but might be useful for genus level designations or higher where the designatio is more likely to apply to hybrids.

Andy

Andy Foy
Systems Manager
Greenspace Information for Greater London (GiGL) CIC
www.gigl.org.uk

9 Reply by RobLarge

  • RobLarge
  • RobLarge
  • Veteran
  • Offline
  • From: Wiltshire
  • Registered: 15-07-2008
  • Posts: 812

Re: Hybrids with incorrect status on export

I would have thought that if a species is designated, logically the designation would apply to all subspecies.

Another point worthy of consideration however, is the situation of a species where one or more (but not all) subspecies is designated. How then would you handle a record at the species level (which may or may not be of a designated taxon)? It seems to me that we would need to be able to say that the record may be of a designated taxon, but clearly indicate the uncertainty. I guess that the same would apply to records at genus or higher level where some of the contained species are designated. Some kind of routine to search the subordinate taxa for designations perhaps.

In either case the system would need to be able to consider these issues against a designation set, rather than (or as well as) all designations in order to ensure fleixibility with regard to designations which (e.g.) do not apply locally.

Rob Large
Wildlife Sites Officer
Wiltshire & Swindon Biological Records Centre

RobLarge's Website

10 Reply by charliebarnes (edited by charliebarnes 26-07-2011 08:57:39)

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

Re: Hybrids with incorrect status on export

The hybrid should in most cases occur at the same heirarchical level

Couldn't we just edit the taxon dictionary (TAXON_LIST_ITEM?) to change the parent of the hybrid and rebuild the taxon indices? As an interim solution, at least, until the taxon dictionary is fixed.

Charlie Barnes
Information Officer
Greater Lincolnshire Nature Partnership

charliebarnes's Website

11 Reply by MikeWeideli

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

Re: Hybrids with incorrect status on export

The problem with adding fields to the table is that someone has to maintain them and I don't know at the present time where the resource would come from to do this.  JNCC have expanded the Designation list down to species which has been helpful, but ideally it should be expanded to include sub-species or below where applicable,. If this was done then there would be no need to expand the list in Recorde 6.  I can't see this being any more difficult to do than trying to add indicators to taxa, however, it is question of resource.

I think it will be easy enough to modify the query which generates Index_Taxon_Designation to exclude the hybrids and this can be made more sophisticated so that it doesn't remove hybrids if they actually appear on the JNCC (or local) designation list.  The query is in the stored procedure,  usp_Index_Taxon_Designation_Rebuild and this could be altered locally as a temporary  measure.  The problem with the expansion is that all lists are taken into account, so badly constructed lists affect the results. If we can do what we hope with the dictionary early in 2012, then we should  be able to expand the list without relying on these ureliable lists.

Mike Weideli

MikeWeideli's Website