VERDI2.0-Beta: Adding External GIS Shapefile Layers

[Patrick.C.Campbell]

Dear CMAS Community:

Shapefiles that come packaged with Verdi v2.0-Beta, such as the huc03.shp, are easily added as GIS layers to tile plots; however, externally downloaded shapefiles do not work, even when all the shapefile layers, .shx, .prj, and .dbf are present.

For example those downloaded at: https://www.naturalearthdata.com/downloads/10m-physical-vectors/ When adding the downloaded shapefiles (e.g., the “Physical Label Area” shapefiles) to VERDI (tile plot of NOx emissions, clicking on GIS Layers, clicking on Add Map Layers, and then clicking on Other), the polygons from the shapefile create an erroneous plot as you can see below:

Seems likely it could be an issue with directly adding a different shapefile projection.

Has anyone else run into this issue or been successful displaying externally downloaded shapefiles as GIS layers VERDI v2.0-Beta?

Thank you,

Patrick

[lizadams]

Hi Patrick,

I can’t seem to view the image that you attached to this issue. I did download the ne_10m_rivers_lake_centerlines from the link that you provided and was able to successfully load that shapefile and see the rivers on a tileplot.

First I selected the dataset and variable, and added it as a formula, and then created the tile plot. To add the shapefile to the tileplot, I selected the menu option:
GIS Layers > Add Map Layers > Other

Then from the pop-up window that was titled “Open”, I navigated to the directory where I had downloaded the shapefile and selected the following file, and clicked “Open”
ne_10m_rivers_lake_centerlines.shp

The Shapefile was added to the Tile Plot:

So, it is possible, that the dataset that I chose was of the same projection as the shapefile, and that the dataset that you are loading is a different projection.

I will try to load a dataset with a different projection, and see if I can reproduce the error that you are seeing.

Can you send me the projection information for your dataset? You can obtain it using the following command:

ncdump -h 

[Patrick.C.Campbell]

Thank you Liz for having a look at this.

Yes, I thought maybe it was a projection issue, however, I haven’t had any luck adding those shapefiles to a number of different projections, maybe which all are different.

For the example I provided, I was just looking at the world NOx file that is provided as example data with VERDI2.0-beta, VERDI_2.0_beta/data/model/world_NOX_latlon.ncf Seems like it is just a rectilinear lat/lon projection.

I can also successfully add the ne_10m_rivers_lake_centerlines as you did for your example to the rectilinear lat/lon NOx world map, but I did not download this and it was already packaged with the VERDI2.0-beta distribution.

However, as before, when trying to add the shapefile I download from that website, Physical Label Area” polygon shapefiles, then I get the erroneous GIS layer data in the tile plot.

I also tested both the above shapefiles with the CMAQ east_Ozone model output packaged with VERDI2.0-beta, VERDI_2.0_beta/data/model/cmaq.cdc.east_Ozone_20060703-1.nc, which is of course a Lambert-Conformal, curvilinear projection, and I get the same results as above, i.e., OK when adding the “Rivers_Lakes_Centerlines” shapefile, but erroneous when adding the downloaded "Physical Label Area” polygon shapefiles:

You can see how I leave open the manage GIS layers GUI window in the snapshots to show the GIS layers I have been adding the in above examples.

Thanks again for your help with this,

Patrick

[Patrick.C.Campbell]

@lizadams

et that you are loading is a different projection.

Hi Liz,

For some reason the CMAS Forum limits me to only one image in my previous reply. Here is the similar world NOx example on the rectilinear lat/lon grid, which gives me the same result for the shapefiles tested:

Thanks again,

Patrick

[lizadams]

Hi Patrick,

I was able to successfully plot the ne_10m_rivers_lake_centerlines.zip using the world_NOX_latlon.ncf.

I obtained the same set of lines as you did for the ne_10m_geography_regions_polys

I also obtained a heavy line at the equator when adding the ne_10m_lakes.shp to the tileplot of NOX from the world_NOX_latlon.ncf file.

I’ve forwarded this issue to the development team.
It may be that these types of spurious lines are most visible on tileplots that display the full extent of the earth, and not as much of an issue on tileplots of sub-domain regions.

[Patrick.C.Campbell]

Thank you Liz for forwarding this to your development team as there does seem to be an issue with adding and displaying GIS shapefile layers. I also obtain a heavy line at the equator for the ne_10m_lakes.shp as you do.

Patrick

[esm]

Hi

I want to analysis my results, but I don’t know how can I show them on my city map. Could you guide me, please?
Thank you