Week 4 Exercises - Astro RPS

In this set of exercises, you’ll get more practice writing routines in Python and working with numpy arrays, lists, and dictionaries.

You may find it useful to use some if-statements, some list indexing, and the np.min/np.max functions.

Exercise 2: Finding Peaks

Create a script peak_finder.py.

The goal of this script is to locate the peaks in an array. In this case, it will be a 1D array, but this issue generalizes: stars in an image are peaks, and we often need to find them!

Below, I provide a list, flux_values, and then I plot it’s y values. Note that matplotlib assumes x-values are an arange if not provided. Your goal is to write code which can find the peaks in this array, where a peak is defined as any location where the flux value is higher than both neighboring values (i.e., in [1,2,1], index 1 (the two) would be a peak).

The answer you are looking for (see the plot) is [4,13,16]. In an array this small, you could find this by eye or trial and error, but in a real array of thousands of elements, you may find it useful to have a peak finding code around!

Hint

There is a looping based solution to this, and a non-looping based solution! If you ARE looping, you may need to be careful about how you handle the endpoints.

import numpy as np 
import matplotlib.pyplot as plt 


flux_values = [0,0,1,2,3,2,1,1,1,0,2,2,3,5,4,3,6,2,1,0]

fig,ax = plt.subplots()
ax.plot(flux_values)
ax.axvline(4,alpha=0.2)
ax.axvline(13,alpha=0.2)
ax.axvline(16,alpha=0.2)

Exercise 3: Load an Astronomical Image

In this exercise, we’ll view our first astronomical image! You can find it on Slack.

To open an astonomical image, we will use the astropy.io.fits module. Import it:

from astropy.io import fits

To obtain the header and image data from this file, we can do the following

with fits.open('/path/to/file') as hdu:
    header = hdu[0].header 
    image = hdu[0].data

We’ll talk more about what this means on Monday!

Once you’ve done this, you should have an image variable that is a 2D array. Use the imshow() command from matplotlib to view it.

Does it look like much?

You will probably need to adjust key parameters vmin and vmax to get a nicer looking image. I suggest using np.percentile() to compute the 1st and 99th percentile of the data, and then use those values when seeting the vmin and vmax arguments in imshow.