Increase coordinate uncertainty for gbif dataset

[SanderDevisscher]

Due to a issue with coordinates flagged by Mieke Hoogewijs (via @damianooldoni) I've noticed the coordinate uncertainty is not correct. More precise it still is the value prior to the conversion to 1x1km squares (#91).

This value should be changed.

@damianooldoni what value do you mostly use in case of blurring to 1x1km ?

changes should be implemented here:

sk-analyse/script/dwc_mapping.Rmd

Line 328 in 2a2300c

occurrence %<>% mutate(dwc_coordinateUncertaintyInMeters = "30")

& here:

sk-analyse/script/dwc_mapping_bycatch.Rmd

Line 296 in 2a2300c

occurrence %<>% mutate(dwc_coordinateUncertaintyInMeters = "30")

[damianooldoni]

@SanderDevisscher: typically we should use the radius of the circle circumscribed a square of 1km x 1km. So, sqrt(2 * 0.5^2). @peterdesmet: is it correct?

[SanderDevisscher]

in meters, so sqrt(2 * 500^2) =~ 707m

[SanderDevisscher]

Most of the issues are caused by the ponds being in different gridcells, the crossing of several squares happens to pass through that area, see image below:

[damianooldoni]

Indeed, 707 meters is correct.

I would ask to the partners specifically involved with these occs if you may blurring less than 1x1km. Or not blurring at all 😄 Probably the concerns of few years ago have faded away?

[SanderDevisscher]

It's an issue from late last year 😖

However I've explained the cause of the issue to Mieke and she is going to try and convince the other members of the consortium to do away with the blurring 💪 if she succeeds I need to disable the blurring 🤞

[SanderDevisscher]

Blurring should remain 😢