Wrong vice county where straddles boundary

Submitted by charlie on

A number of my records that straddle vice-county boundaries have been assigned a single (wrong) one - whereas others that straddle a vice-county boundary are assigned to both. Eg 30394627

There also seems to be no way to edit this or have I missed it?


Submitted by James Emerson on Wed, 02/08/2023 - 19:31


You can't edit the vice county manually as far as I'm aware, it is applied automatically as a result of the record coordinates. 

The other part of your issue is one that other people seem to have as well, which is that if you use the app the record is recorded using lat and long, which is then converted into a grid reference in the main part of iRecord. Looking at the example record number you gave, the location has an undertainty of 500m, which is why there is the big circle around it, and presumably more of the circle is in Suffolk than Norfolk, so it has listed it as Suffolk. I think if you added a record at, e.g. a 6 figure reference and that square crossed the vice county boundaries, that is the sort of thing that would assign the record to both, but that's based on having seen a few similar records not on any inside knowledge on how the system works.

Submitted by iRecord support 3 on Thu, 03/08/2023 - 14:17


That is correct - the location of this record is recorded as a point (with uncertainty) rather than a grid square, and in these cases it is the central point location that is overlaid with the VC boundaries to assign a VC. You can't edit the VC manually, but you could edit the location slightly to endure the centre point is the right side of the VC boundary here: https://irecord.org.uk/enter-app-record?occurrence_id=30394627

Submitted by A G Duff on Mon, 07/08/2023 - 12:10


I strongly suspect that these sorts of problems are occurring because when the app calculates its lat/long from a grid ref it is defaulting to the southwest corner of the implied square. It should, in my opinion, add a northeast offset first so that it points to the centre of the square, not its southwest corner. As this is essentially an average position for where the record actually took place there is a greater chance that the calculated vice-county will be accurate. It should also eliminate the problem that I have been encountering (see Data Flow threads) where the recalculated grid reference is incorrect.

In any case, why does the app calculate lat/long in the first place? If I give it a grid ref that is what I want the record to have, not a grid ref recalculated from a lat/long which has been offset southwest by 70 metres.