algorithmic modeling for Rhino
So I received an inquiry about using Meerkat with GHowl to coordinate multiple shape file projections into the same point space in Rhino/Grasshopper. Here are my initial findings.
I used a Meerkat gis crop file (.mkgis) to create a rectangular boundary of a shapefile crop in point space. Meerkat also outputs the lat long coordinates of the bottom left and upper right vertices of the boundary. Using these as inputs, I tested GHowls lat long and xyz conversion tools, by converting points in Rhino to lat long and then back to Rhino point space from lat long. I then mapped out the discrepancy between the points.
I tested two grids of points one is the size of the meerkat crop rectangle (100 x 100 points of roughly 300 x 300 meters). The other is a(100 x 100 grid of points 10 x 10 kilometers). I realized that a non-uniform scale can drastically reduce the discrepancy between the original points and the GHowl double-translated points.
In image below, blue = original point, green = original Ghowl, magenta = adjusted GHowl. These differences are most likely due to the projection differences and how the surface of the Earth is translated (though both run in WSG84).
IMPORTANT NOTE: Units must be in meters to work with GHowl. In my gh file you'll see a multiplier (via scale) to change from feet to meters.
Ghowl and Meerkat can be combined to link multiple GIS projections into the same Rhino point space but user beware of the discrepancies and be aware of the margin of error. Please weigh in if you have more info.
Release 1.5 solves this dilemma. Two new components allow projection back and forth between point space and latitude-longitude.
Thanks so much for this- the extra features in r1.5 are super helpful.
What kind of information is the "projection file" looking for in the new components? Just trying leverage these a little bit...