#About

Python script to generate a graticule that will project cleanly (with smooth arcs) at world scale. The graticule should have a dense distribution of internal nodes along it's line, especially along the ±180, ±90 WGS84 bounding box.

Output format is geojson, because it's easy to write as a text file in python. Use ogr2ogr to convert to a SHP file.

By Alex Mandel Copyright 2012. Modifications by Nathaniel Vaughn KELSO.

#Usage

python pygraticule.py -g 1 -o outfile.geojson

There are two script arguments for controlling the density of nodes and the output file name:

• g is the grid interval is measured in decimal degrees: generally 69 miles at the equator, but varies at higher latitudes.
• s optional, is the step interval is measured in decimal degrees: defaults to every 0.5 a degree, adds a vertex after that distance. Future work might vary this by latitude with a cosine()?
• o is the outfile. If none is provided, file is auto named in this format: "graticule_1dd.geojson" where the number is the step interval.

Limitations

• The output is lines, not polygons feature classes.

• The output is a GeoJSON format file, not a SHP format Esri Shapefile.

Examples

When we project out of WGS84 to another coordinate system that is not cylindrical, we need to have enough intermediate nodes on the paths so the GIS application shows a "curve". Most GIS do not auto-densify stright lines during the projection so we need to add these extra nodes in the raw geodata.

Here we see Robinson using enough nodes:

Box results when nodes are sparse:

The two superimposed:

The proj4 string for Robinson is: +proj=robin +lon_0=0 +x_0=0 +y_0=0 +ellps=WGS84 +towgs84=0,0,0,0,0,0,0 +units=m +no_defs