Note
Click here to download the full example code
ContoursΒΆ
The pygmt.Figure.contour
method can plot contour lines from a table of points by direct triangulation.
The data for the triangulation can be provided using one of three methods:
x
,y
,z
1dnumpy.ndarray
data columns.data
2dnumpy.ndarray
data matrix with 3 columns corresponding tox
,y
,z
.data
path string to a file containing thex
,y
,z
in a tabular format.
The parameters levels
and annotation
set the intervals of the contours and the
annotation on the contours respectively.
In this example we supply the data as 1d numpy.ndarray
with the x
, y
,
and z
parameters and draw the contours using a 0.5p pen with contours every 10 z
values and
annotations every 20 z
values.
Out:
<IPython.core.display.Image object>
import numpy as np
import pygmt
# build the contours underlying data with the function z = x^2 + y^2
X, Y = np.meshgrid(np.linspace(-10, 10, 50), np.linspace(-10, 10, 50))
Z = X ** 2 + Y ** 2
x, y, z = X.flatten(), Y.flatten(), Z.flatten()
fig = pygmt.Figure()
fig.contour(
region=[-10, 10, -10, 10],
projection="X10c/10c",
frame="ag",
pen="0.5p",
# pass the data as 3 1d data columns
x=x,
y=y,
z=z,
# set the contours z values intervals to 10
levels=10,
# set the contours annotation intervals to 20
annotation=20,
)
fig.show()
Total running time of the script: ( 0 minutes 1.106 seconds)