Pygmt.plot possible bug

[tonino13]

Dear Forum.

I have the pygmt v0.10 installed within anaconda environment. I have a csv file (CANP_v2.dat) which contains data from a seismic fault.

I tried to plot and the results looks like my points from the csv close a polygon (output.png).

This is the code:

import os
import pygmt
import pandas as pd
import numpy as np

region_central = [-66.704, -63.44, -18.72, -16.51]  

grid = pygmt.datasets.load_earth_relief(resolution="03s", 
                                        region=region_central)
                                        
dgrid = pygmt.grdgradient(grid=grid, azimuth=270, normalize="e")

fig = pygmt.Figure()

pygmt.makecpt(cmap="geo", series=[float(grid.min()), float(grid.max())], 
                                                           continuous=True)
fig.grdimage(
    grid=grid, 
    projection="M6i",
    region=region_central, 
    frame=["a1f1", "WSne+t'Orocline Faults'",
           "xaf1f0.5+lLongitude",   # x-axis: major every 1°, minor 0.5°
           "yaf1f0.5+lLatitude"],   # y-axis: major every 1°, minor 0.5°
    shading=dgrid,
    cmap=True
)

# Add political boundaries, coastlines, and water color
fig.coast(
    region=region_central,
    projection="M6i",
    borders="1/1.5p,black",  # Country borders
    shorelines="1/0.8p,black",  # Coastline
    water="lightblue",
    map_scale="jTR+w50k+o0.4c/0.6c+f",
)


# # --- The CANP --- #
df_canp = pd.read_csv("/home/asus/Documents/PyTools/Points/PrincipalFaults/CANP_v2.csv", delimiter=",")  

fig.plot(
     x=df_canp["Lon"],
     y=df_canp["Lat"],
     pen="2p,black",
     )

faultnames = {"CANP" : (-66.603, -16.558),
              "CBE" :(-65.794, -16.738),
              "CFP" : (-66.364, -16.642),
              "MAF" : (-65.010, -17.247)
              }

# # Plot names of faults in squares 

fig.plot(
     x=[lon for lon, lat in faultnames.values()], 
     y=[lat for lon, lat in faultnames.values()], 
     style="s0.2c", 
     color="black", 
     pen="1p,white" 
 )

fig.text(
     text=list(faultnames.keys()),
     x=[lon for lon, lat in faultnames.values()], 
     y=[lat for lon, lat in faultnames.values()], 
     font="8p,Helvetica-Bold,black",
     justify="LM",  
     fill="white@60", 
     offset="0.2c/0c" 
 ) 
 
fig.show()

This is the output using pygmt.plot (see the solid black line):

However, If I use Matplotlib I have the fault as it is (good_plot.png)

Any idea is appreciated
CANP_v2.dat (4.4 KB)

[yvonnefroehlich]

Hi @tonino13,

fig.plot(
     x=df_canp["Lon"],
     y=df_canp["Lat"],
     pen="2p,black",
     )

In case you like to plot your data points as symbols you have to use the style parameter, e.g., style="c0.2c" for circles with a size of 0.2 centimeters. If you only use pen the data points are connected by lines. In combination with style, pen sets the outline of the symbol. The fill parameter allows to fill the symbol with a color, e.g., fill= "blue".

BTW we just released v0.17.0 .

[tmerry1]

The problem is in your csv — lines 2–59 should be sequentially after lines 60–288 (note that line 288 is practically the same as line 2). I think the line joining the points defined by line 59 and line 60 is probably the problem.

[tonino13]

@tmerry1 true, thank you for your time reviewing the csv file