Enable center to be defined in lat/lon

[jourdain]

That way when we reset the view (or something else) we ensure the given location to be centered in the view.

Maybe that is just panning to put that location in center without touching at the current zooming.

[jourdain]

Do we want to mark an area as a region on interest (min/max of lat/long) box? And use that box bounds to do a reset camera?

I'm not sure exactly how a single lat/lon center should be handle as a "center"?

[huiwanpnnl]

You're absolutely right. When talking about a "center longitude" we were thinking about global maps, and we assumed that we would always show the full range of latitudes (-90 to 90) centered at the equator. In a sense, this was just about rotating the Earth in the east-west direction.

And yes, for a lat/lon box, it would make sense to simply reset camera, so that the center of the view is located at the midpoint of each direction (lat and lon).

[huiwanpnnl]

I just realized that when we start to work on extending QuickView to support E3SM's surface models (see issue #108), the feature discussed here will become important. For example, the land ice model developers will often want to center their map/view at the South Pole or somewhere over Greenland, while a land modeler will likely want to center their map/view over a contienent.

[huiwanpnnl]

As of early August, 2026, center lat/lon can be specified for the spherical projection (but not for the other projections).

At the SciDAC PI meeting, @jourdain and I agreed that we need to update the slider for longitude cropping so that the range shown becomes (center_lon - 180, center_lon + 180).