S-101-Portrayal-subWG / Working-Documents

17 stars 5 forks source link

Display of the seaward edge of a mangrove in the intertidal area (DCEG 5.12.1) #34

Closed alvarosanuy closed 2 years ago

alvarosanuy commented 3 years ago

The DCEG subWG has requested the Portrayal subWG to discuss changes to portrayal rules in order to symbolise the seaward edge of mangrove areas encoded in intertidal regions so they display using similar portrayal than Coastline with categoryOfCoatsline=7 (mangrove). It's important to note than in S-57, the UOC instructs producers to encode COALNE (CATCOA=7; QUAPOS=4) on the seaward edge of these objects as a workaround to solve display. This encoding guidance has now been removed from the DCEG with the expectation that display will be now managed by portrayal rules. In both, S-57 and S-101, the landward edge of the vegetation feature shares spatial with the coastline.

DavidGrant-NIWC commented 3 years ago

Edit: - I missed that this was specific to intertidal regions. What is the expected co-incident feature(s) on the seaward edge? If there is no shared spatial component then the portrayal will not be able to implement this.

It might be easier to add an InformationType similar to SpatialQuality which can be associated with the edge to indicate this condition: image

I don't understand the purpose of this requested change. The seaward edge will always be suppressed by the coastline assuming the curve geometry is shared (the coastline is always in the same or higher display plane, and always has a higher drawing priority). See S-100 9-11.1.8 Line Instruction paragraph 2:

image

Feature Attribute Geometry Viewing Group Drawing Priority Display Plane Line Style Line Width Line Color Area Fill
Coastline categoryOfCoastline:7 Curve 12410 21 via RadarOverlay context parameter dash 0.32 mm CSTLN -
Vegetation categoryOfVegetation:7 Curve 32030 9 via RadarOverlay context parameter dash 0.32 mm LANDF -
Vegetation categoryOfVegetation:7 Surface 32030 9 Under RADAR dash 0.32 mm LANDF VEGATN04
alvarosanuy commented 3 years ago

The intent is to make the seaward edge of the Vegetation/ categoryOfVegetation=7 to display as Coastline / categoryOfCoastline=7. This means dashed grey (CSTLN) line. The proper Coastline (boundary between Land and Intertidal) will be a continuous grey (CSTLN) line. The interior of the Vegetation surface will be made of a pattern (VEGATN04) of buff (LANDF) mangrove symbols.

What about adding the Boolean attribute 'inTheWater' to Vegetation? This would help with portrayal and also make this Vegetation surface be part of ECDIS Base display (which makes sense because it is a dangerous area to be avoided). Having inTheWater=True could be the trigger to change the portrayal of the seaward boundary of the Vegetation from buff dashes to grey dashes. The overlap with the Coastline wouldn't be a problem because Vegetation would be masked out (due to Coastline higher display priority).

This approach would require changes to the FC and DCEG and may also require a new Validation check (Critical?) to avoid encoders from using inTheWater=True when Vegetation does not overlap DepthArea.

DavidGrant-NIWC commented 3 years ago

From S-101 DCEG v1.1.0_Working 5.12 Vegetation:

[...] Where the source indicates that a mangrove area is in the intertidal area, a Vegetation feature, with attribute category of vegetation = 7 (mangroves) should be encoded on top of the intertidal area (Depth Area with attributes depth range minimum value = -H and depth range maximum value = 0 – see clause 11.7.3)). The seaward spatial type(s) of the mangrove area should have the attribute quality of horizontal measurement = 4 (approximate). The landward edge of the mangrove area should be encoded as Coastline (see clause 5.3), having no value populated for the attribute category of coastline, and no value for quality of horizontal measurement on the related spatial type(s).. [...]

I believe there are no technical issues with implementing the desired portrayal given the latest DCEG guidance; the portrayal can look for SpatialQuality on each curve associated with the Vegetation feature and output the appropriate drawing instruction(s).

SylviaSpohn-BSH commented 3 years ago

The proposal with the Boolean atrribute 'inthewater' for vegetation should be tested for the depiction. We have cases in non-intertidal waters where reed beds also go in the water and we need this edge of vegetation. Usually we depict it as a unsurveyed coastline.

DavidGrant-NIWC commented 3 years ago

Summary of agreement at mtg:

From the test data set

Top left: Vegetation.categoryOfVegetation=7 Bottom middle: Vegetation.categoryOfVegetation=21 image

alvarosanuy commented 2 years ago

Decisions and Actions at 01DEC2021 (post 2nd meeting in November 2021):

image

SylviaSpohn-BSH commented 2 years ago

As mangroves are usually combined with the land area and if the navigation through this vegetation is not allowed or dangerous. It could be charted as land with an unsurveyed (dashed) coastline. We have the same situation with reed beds in the Baltic Sea. Or the attribute "in the water" is taken to differentiate such vegetation. For navigation issues it seems not so important and to extend the land area seems a better solution. It is also a matter of scale. Obstructions are usually artificial and man-made. Also underwater volcanos should not be charted as a category of obstruction, see #71.

alvarosanuy commented 2 years ago

Decision made at Portrayal subWG meeting on 12/7/22

  1. Implement portrayal so Obstructions with categoryOfObstruction=23 (mangrove) are depicted as follows: image
alvarosanuy commented 2 years ago

This issue has been re opened until changes are implemented in PC 1.0.2 by NIWC.

mikan66 commented 2 years ago

Decision made at Portrayal subWG meeting on 12/7/22

  1. Implement portrayal so Obstructions with categoryOfObstruction=23 (mangrove) are depicted as follows: image

Implemented as: image

alvarosanuy commented 2 years ago

Implemented in PC 1.0.2

alvarosanuy commented 2 years ago

Can NIWC produce screenshots of an Obstruction with categoryOfObstruction=23 sitting on an intertidal area (sharing full geometry and partial overlapping)? We need to see portrayal in all ECDIS display modes. Safety Contour setting must be deeper than 0m so we can see the impact of CSP OBSTRN07 as well. According to DCEG, this category of obstruction MUST be encoded in intertidal areas only.

If test datasets are required, please submit request to Tom Richardson.

DavidGrant-NIWC commented 2 years ago

Using fake data (changed SeabedArea to Obstruction with categoryOfObstruction = 23)

Note:

Safety Contour Base Std Other Notes
5 ShoreECDIS 2022-11-02T193438Z ShoreECDIS 2022-11-02T193430Z ShoreECDIS 2022-11-02T193420Z Safety alert highlights disabled in Base
0 ShoreECDIS 2022-11-02T193244Z ShoreECDIS 2022-11-02T193254Z ShoreECDIS 2022-11-02T193301Z Safety alert highlights disabled in Base
alvarosanuy commented 2 years ago

Using fake data (changed SeabedArea to Obstruction with categoryOfObstruction = 23)

Note:

  • Safety contour suppresses boundary of mangroves when co-incident
  • CSP OBSTRN07 does not apply / is not called for mangroves

    • Alerts are triggered by the underlying depth area and / or safety contour

Looks good to me. I understand that you have Safety Alert/Highlights disabled for all features when in Base display (or is there something not working for the 'Mangrove' encoding only. Alerts/Highlight should be triggered by DEPARE/SC anyway)???

DavidGrant-NIWC commented 2 years ago

I understand that you have Safety Alert/Highlights disabled for all features when in Base display (or is there something not working for the 'Mangrove' encoding only.

The alert highlights are (currently) in viewing groups 101-115, which are part of viewing group layer 10b ("Alert Highlights (Standard)"). These are part of the standard display: image image

The indication at the bottom of the screen reflects the fact that the highlights are disabled: image

To match S-52 we should probably move all of the alert highlights to viewing group 53010 (mariners standard). The current VG's were created to support testing of the following 61174 requirement (which is not supported by the single S-52 VG). The current viewing group numbers do not follow the convention that standard display VGs are in the range [20000, 30000) for chart elements, and [50000-60000) for mariners' elements. image

Alerts/Highlight should be triggered by DEPARE/SC anyway)???

Correct. The mangroves do not trigger an alert. Alerts will be triggered by the underlying depth area.