Projecting coordinates in Python

Sometimes we have to manipulate huge spatial databases, and it uses to be a headache if we must to convert the spatial point to a different projection. Well, an excellent way to solve this sort of issues is taking advantage of the libraries of python. In this case, I am going to describe how to use the package utm for it.

First, we are going to simulate some spatial data in a decimal projection.

1 2 3 4 5 6 7 8 9 10 11 12 13 14

import matplotlib.pyplot as plt import numpy as np import pandas as pd import utm lat = np.linspace(20, 40, 100) long = np.linspace(60, 70, 100) index = np.arange(1,101) d = {'x': pd.Series(lat, index=index), 'y': pd.Series(long, index=index)} df = pd.DataFrame(d) print(df)

Done!, we have already created a Dataframe with one hundred of spatial points in decimal degrees. Now, we are going to project them to UTM. First, we make a temporal "temp" variable which will store our projected XY coordinates. Afterwards, we are going to use the function utm.from_latlon() in a loop. This instruction will put each project coordinate in your variable "temp" that you created recently.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

import matplotlib.pyplot as plt import numpy as np import pandas as pd import utm lat = np.linspace(20, 40, 100) long = np.linspace(60, 70, 100) index = np.arange(1,101) d = {'x': pd.Series(lat, index=index), 'y': pd.Series(long, index=index)} df = pd.DataFrame(d) #print(df) temp = [[None]*1 for ii in range(len(df.x))] ## Use the utm package to transform coordinates for i in np.arange(0, len(df.x)): temp[i] = utm.from_latlon(df.y[i+1], df.x[i+1]) # Transform in a dataframe data = pd.DataFrame(temp) data.columns = ['X', 'Y', 'Zone', 'Code'] print(data)

Finally, we convert the projected data to a DataFrame and assign this to a new variable "data", and that's all. This procedure helps me to transform a database with two hundred thousand locations, so it is useful to project massive databases.