Negative wind speeds in windfield

[ddusseau]

I am using the STORM dataset to create windfields with the tropical cyclone package. When I explore the windfield output, I see that the magnitude portion of the velocity vector is sometimes negative. I made sure to check I was grabbing the magnitude and not the direction component of the vector. Why would there be negative wind speeds?
I have attached an example of the magnitude for one track.

tropcyc = TropCyclone.from_tracks(tracks, centroids=cent,
                                pool=pool, max_dist_inland_km=500,
                                max_dist_eye_km = 300, model="H10", store_windfields=True)
for i in tropcyc.windfields:
        if i.nnz != 0:
          test = i.toarray()
          test = test.reshape((test.shape[0], cent.size, 2))
          plt.imshow(test[:,:,0], cmap='viridis')
          plt.colorbar()
          plt.xlabel('Centroids')
          plt.ylabel('Positions')
          plt.show()

[chahank]

Dear @ddusseau ,

what you are plotting is the second component of the windfield vectors, not their magnitude.

As stated in the docstring, the TropCyclone.windfields stores the vector components in a sparse matrix.

For each track, the full velocity vectors at each centroid and track
position are stored in a sparse matrix of shape (npositions,
ncentroids * 2) that can be reshaped to a full ndarray of shape (npositions,
ncentroids, 2).

In order to get the magnitude, you would need to take the norm: np.linalg.norm(windfield.toarray().reshape(windfield.shape[0], centroids.size,2), axis=-1)

[ddusseau]

Dear @chahank ,

Thank you for the clarification!
And to get the wind direction angle would this be correct?

wind = windfield.toarray().reshape(windfield.shape[0], centroids.size,2)
angle = np.degrees(np.arctan(wind[:,:,0]/wind[:,:,1]))

[chahank]

The angle with respect to where? Please refer to a trigonometry textbook on how to calculate the angle between two vectors. For further non-coding related questions, I kindly refer you to more suitable resources such as StackOverflow. I hope you were able to resolve your problem.