Plotting Geographical Data using Basemap

2 min read

Basemap is a Matplotlib toolkit, a collection of application-specific functions that extends Matplotlib functionalities, and its complete documentation is available at http://matplotlib.sourceforge.net/basemap/doc/html/index.html.

Toolkits are not present in the default Matplotlib installation (in fact, they also have a different namespace, mpl_toolkits), so we have to install Basemap separately. We can download it from http://sourceforge.net/projects/matplotlib/, under the matplotlib-toolkits menu of the download section, and then install it following the instructions in the documentation link mentioned previously.

Basemap is useful for scientists such as oceanographers and meteorologists, but other users may also find it interesting. For example, we could parse the Apache log and draw a point on a map using GeoIP localization for each connection.

We use the 0.99.3 version of Basemap for our examples.

First example

Let’s start playing with the library. It contains a lot of things that are very specific, so we’re going to just give an introduction to the basic functions of Basemap.

# pyplot module import
import matplotlib.pyplot as plt
# basemap import
from mpl_toolkits.basemap import Basemap
# Numpy import
import numpy as np

These are the usual imports along with the basemap module.

# Lambert Conformal map of USA lower 48 states
m = Basemap(llcrnrlon=-119, llcrnrlat=22, urcrnrlon=-64,
  urcrnrlat=49, projection='lcc', lat_1=33, lat_2=45,
  lon_0=-95, resolution='h', area_thresh=10000)

Here, we initialize a Basemap object, and we can see it has several parameters depending upon the projection chosen.

Let’s see what a projection is: In order to represent the curved surface of the Earth on a two-dimensional map, a map projection is needed.

This conversion cannot be done without distortion. Therefore, there are many map projections available in Basemap, each with its own advantages and disadvantages. Specifically, a projection can be:

equal-area (the area of features is preserved)
conformal (the shape of features is preserved)

No projection can be both (equal-area and conformal) at the same time.

In this example, we have used a Lambert Conformal map. This projection requires additional parameters to work with. In this case, they are lat_1, lat_2, and lon_0.

Along with the projection, we have to provide the information about the portion of the Earth surface that the map projection will describe. This is done with the help of the following arguments:

Argument	Description
llcrnrlon	Longitude of lower-left corner of the desired map domain
llcrnrlat	Latitude of lower-left corner of the desired map domain
urcrnrlon	Longitude of upper-right corner of the desired map domain
urcrnrlat	Latitude of upper-right corner of the desired map domain

The last two arguments are:

Argument	Description
resolution	Specifies what the resolution is of the features added to the map (such as coast lines, borders, and so on), here we have chosen high resolution (h), but crude, low, and intermediate are also available.
area_thresh	Specifies what the minimum size is for a feature to be plotted. In this case, only features bigger than 10,000 square kilometer