Getting Started

Installation

You can install GeoMesher using pip:

$ pip install geomesher

or using conda (mamba):

$ conda install -c conda-forge geomesher

Quick start

The following example shows how to generate a mesh from a GeoDataFrame using both the gdf_mesher function and the Mesher class.

We start by getting a GeoDataFrame of South America from the Natural Earth website. Then, we reproject it to a projected coordinate system (UTM zone 20S). Finally, we add a new column called cellsize that will be used to set the maximum size of the mesh elements.

We use the gdf_mesher function to generate the mesh with default parameters and use Mesher to generate the mesh with MESH_ADAPT algorithm. We also use the area_interpolate function to remap the POP_EST column from the source GeoDataFrame to the generated mesh.

import geopandas as gpd
import geomesher as gm

world = gpd.read_file(
    "https://naciscdn.org/naturalearth/110m/cultural/ne_110m_admin_0_countries.zip"
)

south_america = world[world["CONTINENT"] == "South America"]
south_america = south_america.explode(ignore_index=True).to_crs(32620)
south_america["cellsize"] = 500_000.0

mesh_auto = gm.gdf_mesher(south_america, intensive_variables=["POP_EST"])

mesher = gm.Mesher(south_america)
mesher.mesh_algorithm = "MESH_ADAPT"
mesh_adapt = mesher.generate()
mesh_adapt = gm.area_interpolate(south_america, mesh_adapt, intensive_variables=["POP_EST"])