Add X-Y Coordinates to Generated PECASZonesI.csv file

[bettinardi]

jeabraham commented on Mar 23, 2016

@bstabler or @gregmacfarlane:
PECASZonesI.csv.zip

In comparing my calibration directory against what Graham sent me via SWIM1, I noticed that the PECASZonesI.csv that's generated in a model run on SWIM1 doesn't have the x-y coordinate of the centroid of the LUZs, which we need to generate our compressed flow visuals.

Can you make VISSUM add these in? The values I am using, which look like lat long in the WGS84 projection, are in the attached file. If you want to use a different projection or a different set of centroids, let me know.

gregmacfarlane commented on Mar 23, 2016

This was an issue back on JIRA, and we closed it with wontfix when we moved to GitHub. As I remember, this decision was part triage (we need a model that runs, and this isn't a blocker) and part "can't we just write out a file that we can join to the tables that need it?"

Maybe I'm wrong about both of these, but would love to hear perspectives.

bettinardi commented on Mar 24, 2016

We do write-out x-y coordinates for nodes (link shaping). I think for consistency and ease of use of the data we should plan on writing out zone x-y coordinates as well. @bstabler needs to go in and revise the SWIM-main script anyway, this should be a very simple add-on. So while it is not results critical I would suggest that we take care of it now.

jeabraham commented on Mar 24, 2016

Since we're rarely changing the Beta zones, can we just add two attributes (Lat, long) to the table of beta zone attributes in Vissum? That seems like it should be simple.

We did talk about making the flow visual clustering a separate process, so it could be run after AA, but it will still need x-y coordinates.

jeabraham commented on Apr 5, 2016

@bettinardi @gregmacfarlane can you add the existing XY coordinates from the original attachment as zonal attributes now, or disable the writing of PECASZonesI by Vissum (copy it in from the base year again like before)? Otherwise we'll have to overwrite the PECASZonesI.csv files in every year manually. @graham97 Lets just write a simple python script to iterate through all the years and rerun AA in each year generating flow visuals, then no-one has to touch the tsteps file or the main model run system and risk breaking it. Also a generic script will be useful for other PECAS installations.

gregmacfarlane commented on Aug 22, 2016

I'm putting all of these visuals into the TLUMIP6 label; it doesn't seem critical for the peer review or last haul towards delivery.