Reading CSV files · CESM-diagnostics

I have the following code. It appears to create a numpy array of the values I want from the dataframe. However, when I do the plot, it remembers the time axis from the dataframe. So, it plots the values versus real years (1979 -) I want to reset years to 0 to 40 say. I have tried reset_index() and this appeared to do what I wanted, but I still got a plot versus the real years. I also tried to_numpy(). Same thing. Does anyone have experience with dataframes here?

### Read in the NSIDC data from files

path_nsidc = '/glade/campaign/cesm/development/pcwg/ice/data/NSIDC_SeaIce_extent/'

sep_nsidc = pd.read_csv(path_nsidc+'N_09_extent_v3.0.csv')

sep_area = sep_nsidc.iloc[:,5].values

sep_area

plt.plot(sep_area)

Katelyn FitzGerald (Jan 31 2024 at 17:57):

Is possible you're working in a notebook where there's some holdover from a prior plot? The .values should just give the values in a numpy array. That's odd behavior.

David Bailey (Jan 31 2024 at 18:06):

I even tried doing:

x =np. linspace(1,np.size(sep_area),np.size(sep_area))

plt.plot(x,sep_area)

and same result.

Katelyn FitzGerald (Jan 31 2024 at 18:44):

Does something like this help at all?

plt.clf() # clear figure
x = np.linspace(1,np.size(sep_area),np.size(sep_area))
plt.plot(x,sep_area)

David Bailey (Jan 31 2024 at 19:55):

This did not help unfortunately.

Michael Levy (Jan 31 2024 at 20:02):

What else is going on in your notebook? And what environment / package versions are you running? I copied and pasted your code into a new notebook using NPL 2024a and it doesn't put the years on the axes:

dave-issue.png