Review observations: direction_of_travel

peterdesmet commented 3 years ago

observations: direction_of_travel indicates the direction of travel with a letter code:

https://github.com/ices-tools-dev/esas/blob/0d294b4ea678e34f52409a16602f9133ac179dcf/_data/table-schemas/observations.yaml#L241-L255

[ ] @ices-tools-dev/data-and-information Does ICES already have a controlled vocabulary for this?
[x] @nicolasvanermen Wouldn't it be better to express all of this as degrees from north (e.g. heading)? Or are the number reserved for heading relative to the platform?

nicolasvanermen commented 3 years ago

Yes, during aerial surveys direction of travel is often recorded relative to the movement of the plane.

Nicolas Vanermen Wetenschappelijk medewerker Instituut voor Natuur- en Bosonderzoek Havenlaan 88, bus 73, 1000 Brussel 0486/361.582

On Wed, Jan 6, 2021 at 1:21 PM Peter Desmet notifications@github.com wrote:

observations: direction_of_travel indicates the direction of travel with a letter code:

https://github.com/ices-tools-dev/esas/blob/0d294b4ea678e34f52409a16602f9133ac179dcf/_data/table-schemas/observations.yaml#L241-L255

@ices-tools-dev/data-and-information https://github.com/orgs/ices-tools-dev/teams/data-and-information Does ICES already have a controlled vocabulary for this?

@nicolasvanermen https://github.com/nicolasvanermen Wouldn't it be better to express all of this as degrees from north (e.g. heading)? Or are the number reserved for heading relative to the platform?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ices-tools-dev/esas/issues/36, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABYKGXKCRMMO7SOO3NC62D3SYRISZANCNFSM4VXOWH2A .

peterdesmet commented 3 years ago

@nicolasvanermen I think this should be clearer defined in the definition, so that numbers always represent headings relative to the movement of the plane. My attempt:

Cardinal direction in which the bird/cetacean is travelling: ... Direction relative to the movement of the platform (plane/ship) in which the bird/cetacean is travelling as degrees in increments of 10°: ...

Agree?

peterdesmet commented 3 years ago

Oh, my mistake, the type of direction (relative or absolute) is indicated in direction_of_travel_type (#23). I that case I think we can always work with numeric values? With U Flying, no apparent direction being expressed as an empty value? Wouldn't that make analysis easier?

nicolasvanermen commented 3 years ago

OK, with a minor suggestion:

(Inter)cardinal direction in which the bird/cetacean is travelling: ... Direction relative to the movement of the platform (plane/ship) in which the bird/cetacean is travelling as degrees in increments of 10°: ...

Nicolas Vanermen Wetenschappelijk medewerker Instituut voor Natuur- en Bosonderzoek Havenlaan 88, bus 73, 1000 Brussel 0486/361.582

On Thu, Jan 7, 2021 at 11:55 AM Peter Desmet notifications@github.com wrote:

@nicolasvanermen https://github.com/nicolasvanermen I think this should be clearer defined in the definition, so that numbers always represent headings relative to the movement of the plane. My attempt:

Cardinal direction in which the bird/cetacean is travelling: ... Direction relative to the movement of the platform (plane/ship) in which the bird/cetacean is travelling as degrees in increments of 10°: ...

Agree?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ices-tools-dev/esas/issues/36#issuecomment-756043130, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABYKGXMBKGNXU6FTJYFQWYLSYWHILANCNFSM4VXOWH2A .

nicolasvanermen commented 3 years ago

Let's keep this in, for some data providers (that do not use the U, W, F codes in the distance column), this might be the only column indicating flight behaviour. It also means more than unknown direction, personally I would interpret this as 'hanging around', 'attached to the area',... for what it's worth.

Nicolas Vanermen Wetenschappelijk medewerker Instituut voor Natuur- en Bosonderzoek Havenlaan 88, bus 73, 1000 Brussel 0486/361.582

On Thu, Jan 7, 2021 at 12:01 PM Peter Desmet notifications@github.com wrote:

Oh, my mistake, the type of direction (relative or absolute) is indicated in direction_of_travel_type (#23 https://github.com/ices-tools-dev/esas/issues/23). I that case I think we can always work with numeric values? With U Flying, no apparent direction being expressed as an empty value? Wouldn't that make analysis easier?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ices-tools-dev/esas/issues/36#issuecomment-756046193, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABYKGXPDR7MWUH76WOXOBODSYWH6ZANCNFSM4VXOWH2A .

peterdesmet commented 3 years ago

To make this less confusing on a database level can we decide to:

[x] Only record whether this is absolute or relative directions in the direction_of_travel_type field?
[ ] Only allow 0 to 360 values?
[ ] Converting N to 0, NE to 45
[ ] Replace U by empty, but convince data providers to populate distance instead if they don't?

nicolasvanermen commented 3 years ago

I am not a fan - code U becomes the odd only letter code, only because you're dropping the (inter)cardinal codes N, NE, ..., U, which are widely used in the ESAS v6 (though they were coded as integers 0-10, with 0 for no data, 1 for no apparent direction, 2 for N and so on, that was confusing!).

On Fri, Jan 8, 2021 at 12:07 PM Peter Desmet notifications@github.com wrote:

To make this less confusing on a database level can we decide to:

Only record whether this is absolute or relative directions in the direction_of_travel_type field?

Only allow 0 to 360 values?

Converting N to 0, NE to 45

Replace U by empty, but convince data providers to populate distance instead if they don't?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ices-tools-dev/esas/issues/36#issuecomment-756697085, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABYKGXIV437ADUELMSBBJ4TSY3RN3ANCNFSM4VXOWH2A .

peterdesmet commented 3 years ago

Ok, agree to leave the codes as is. Final question:

Do we assume the N, NW, E values to always be absolute?
Do we assume the 0, 10, 20 values to always be relative?

If so, there is no need for a direction_of_travel_type field in trips.

nicolasvanermen commented 3 years ago

That's more or less the way it has been recorded up until now. I would leave all options open however. N, NE, ... NW are clearly absolute, direction in degrees can in theory be both, so it would be best to keep the direction_of_travel_type.

Nicolas Vanermen Wetenschappelijk medewerker Instituut voor Natuur- en Bosonderzoek Havenlaan 88, bus 73, 1000 Brussel 0486/361.582

On Tue, Jan 12, 2021 at 9:25 AM Peter Desmet notifications@github.com wrote:

Ok, agree to leave the codes as is. Final question:

Do we assume the N, NW, E values to always be absolute?

Do we assume the 0, 10, 20 values to always be relative?

If so, there is no need for a direction_of_travel_type field in trips.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ices-tools-dev/esas/issues/36#issuecomment-758492151, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABYKGXIN3AKZNQAC67ACV2LSZQBQXANCNFSM4VXOWH2A .

peterdesmet commented 3 years ago

direction in degrees can in theory be both

Could you check if that is the case in the data you currently have? I'd very much prefer to contain all the necessary info in just this field.

nicolasvanermen commented 3 years ago

OK Peter, but note that we already dropped 4 (yet unused) codes without consulting the parties that originally included these codes, which is fine because they can be added to the controlled vocabulary if wanted. Which will not be possible once the field is dropped fully. This should all have been discussed earlier...

On Wed, Jan 13, 2021 at 2:43 PM Peter Desmet notifications@github.com wrote:

direction in degrees can in theory be both

Could you check if that is the case in the data you currently have? I'd very much prefer to contain all the necessary info in just this field.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ices-tools-dev/esas/issues/36#issuecomment-759456016, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABYKGXKMR6RWHGCQEK5KAHTSZWPOPANCNFSM4VXOWH2A .

peterdesmet commented 3 years ago

After conversation with @nicolasvanermen: opted for N, NE, ... as absolute and 0, 10, 20 as relative directions. We therefore no longer need direction_of_travel_type. See 5e1decc.

peterdesmet commented 2 years ago

Implemented: observations have a TravelDirection field.

ices-tools-dev / esas

Review observations: direction_of_travel #36