WMS layers from EC2015 3.6 million face grid

[rsignell-usgs]

@USGS-CMG/wetlands-synthesis, I created some NcML to bring UGRID-ize the EC2015 data set from @BrianOBlanton.

To make it as speedy as possible, I copied the EC2015 netcdf file to gamone.whoi.edu:/usgs/vault0/models/tides/ec2015 and created the NcML file there beside it.

You can see the NcML version at https://gamone.whoi.edu/thredds/catalog/usgs/vault0/models/tides/ec2015/catalog.html?dataset=usgs/vault0/models/tides/ec2015/f53.ncml

Just the depth variable is UGRID-compliant at this point, but that should be sufficient for testing how well the WMS services work on this 2M+ node, 3M+ face grid.

[rsignell-usgs]

I snipped out the depth variable to a stand-along netcdf file:

ncks -v depth,ele,lon,lat,adcirc_mesh \
  "https://gamone.whoi.edu/thredds/dodsC/usgs/vault0/models/tides/ec2015/f53.ncml" depth.nc

and dropped that file location into the ncWMS2 server at https://gamone.whoi.edu/ncWMS/

So the good news is that ncWMS2 can make maps from this data:

The bad news is that they each take about 5 minutes to render. :cry:

[rsignell-usgs]

Oooh, filled contours in sci-wms are much faster. This renders in a few seconds: https://gamone.whoi.edu/terriamap/#clean&proxy/_60s/https://raw.githubusercontent.com/USGS-CMG/terriajs-dive/master/examples/ec2015.json

But the colorscalerange request is getting ignored in the getMap requests. Hopefully we'll figure this out here: https://github.com/sci-wms/sci-wms/issues/115

[rsignell-usgs]

This is working nicely now: https://gamone.whoi.edu/terriamap/#clean&proxy/_60s/https://raw.githubusercontent.com/USGS-CMG/terriajs-dive/master/examples/ec2015.json produces: and we can zoom in and rescale, as long as we AVOID ZERO AS A COLOR LIMIT!!!! This is a bug in sci-wms that hopefully will be fixed after merging this PR