Helping Business Weather the Storm

Streamplotting Wind Coordinates

by Justin Stewart, on Sep 7, 2016 1:27:30 PM

Let's take a look at how we can create a matplotlib streamplot using Insight API's Wind Direction and Wind Speed location endpoints.

 It is recommended that you follow along with this example in a Jupyter notebook.


Calculating Wind Coordinates

The Insight API provides endpoints for both Wind Direction and Speed, but not for Wind Coordinates (at least not yet). Here's a simple Python function to accomplish this:

from math import sin, cos

RPD = 0.01745329

def calculate_uv(wind_speed, wind_direction):
    u = -wind_speed * sin(wind_direction * RPD)
    v = -wind_speed * cos(wind_direction * RPD)
    return (u, v)

Retrieving the Data

We'll need a bounding box to determine the lat/lon values we'll be using to make calls to the API. Let's use the bounding box surrounding Oklahoma. We'll represent our bounding box as follows:

#  (min_lon, min_lat, max_lon, maxlat) 
bbox = (-103.0, 33.0, -94.0, 37.0)

Let's start with a precision of one degree for the bounding box grid. We'll need to make a call to both Wind Speed and Direction endpoints for each coordinate:

import numpy as np
from skywiseinsight import HourlyWindSpeed as ws, HourlyWindDirection as wd

def create_uv_grid(bbox, valid_time, precision=1.0):

    Y, X = np.mgrid[bbox[1]:bbox[3]:precision, bbox[0]:bbox[2]:precision]
    V = X * 0
    U = X * 0

    for x in range(X.shape[0]):
        for y in range(Y.shape[1]):

            # Coords
            lat = Y[x][y]
            lon = X[x][y]

            # Call Insight API and Calculate UV
            wind_speed = ws.location(lat, lon, start=valid_time, end=valid_time).series[0]['value']
            wind_direction = wd.location(lat, lon, start=valid_time, end=valid_time).series[0]['value']
            u, v = calculate_uv(wind_speed, wind_direction)

            # Load Data
            U[x][y] = u
            V[x][y] = v

    return {
        'X': X,
        'Y': Y,
        'U': U,
        'V': V

Creating the Stream Plot

Now that we can obtain all of our necessary values, let's load them into a matplotlib streamplot. It will take some time to complete all of location requests to the Insight API:

We'll show you how to quickly make calls to Insight API asynchronously in a future tutorial.

%matplotlib inline

from datetime import datetime
import matplotlib.pyplot as plt

def plot():
    dt =
    grid = create_uv_grid(bbox, dt)
    fig0, ax0 = plt.subplots()

    X = grid['X']
    Y = grid['Y']
    U = grid['U']
    V = grid['V']
    strm = ax0.streamplot(X, Y, U, V, color=U, linewidth=2,


Your end result should look something like the following:

wind streamplot