search

BiologicalRecordsCentre / UKBMS-online

Issue tracking for UKBMS online recording site
2 stars 0 forks source link

Improvements required for Estimate Table #15

Open IanMiddlebrook opened 7 years ago

IanMiddlebrook commented 7 years ago

For the Estimates Table to provide a more accurate prediction of the annual index, we should exclude walks done in sub-optimal conditions, replacing them with standard estimates for those weeks. This was done in the old Transect Walker, and similarly sub-optimal walks are removed by CEH before trend analysis and site-index calculations. This would also require an 'Upland' attribute for each Site, as weather criteria are different for Upland Sites. There should also be an option (probably a default option) to only include weeks 1-26 when calculating the 'total plus estimates' - which would then provide a better surrogate for annual index.

DavidRoy commented 7 years ago

@IanMiddlebrook can you separate this out into the three issues as they need some discussion:

  1. Exclusion for sub-optimal walks
  2. Categorisation of upland sites
  3. Default to 1-26 weeks for summary tables
Gary-van-Breda commented 7 years ago

Whilst considering this area, should the estimates be created for all species (as they are now), including Butterflies, Moths, Odonata and Others?

IanMiddlebrook commented 7 years ago

I'd say it's only worth showing estimates for other species that are recorded consistently. This might apply to day-flying moths and odonata, but I wouldn't worry about others if they will slow things down.

DavidRoy commented 7 years ago

It is probably less effort to do this for everything. I also suspect it won't make much performance issue doing it on everything, given the remainder is a relatively small proportion of data?

Gary-van-Breda commented 7 years ago

I suspect there would be no noticable impact on performance. I supose I am qustioning how valid it is to apply these estimate formulae to non butterfly species?

DavidRoy commented 7 years ago

I think it's ok, at least there's no objective way of deciding what's in or out

DavidRoy commented 5 years ago

@IanMiddlebrook I think this issue is with you to re-arrange into sub-tasks. Can you separate this out into separate issues and then close this more general issue, e.g.

  1. Exclusion for sub-optimal walks
  2. Categorisation of upland sites
  3. Default to 1-26 weeks for summary tables
DavidRoy commented 5 years ago

@IanMiddlebrook to generate new issues for this

Gary-van-Breda commented 2 years ago

@IanMiddlebrook : the 'total plus estimates' is only available on the Estimates, not Summary. I've added an extra column to the Estimates table for the In-season totals (wks 1>26), and deployed to Test for you to check. If you want a similar column on the Summary table, I can do that as well.

IanMiddlebrook commented 2 years ago

Thanks @Gary-van-Breda No need for a similar column on the Summary Table.

Gary-van-Breda commented 2 years ago

Pt3 deployed to Live: 'In season total plus estimates' on the Estimates table.