Wrong statistics in images with periodic boundaries

[smhr]

I have a fits file with periodic BCs which shows projected density of a simulation snapshot. When I compute dendrogram, for structures that straddle the boundaries, the statistics seems to be wrong. please see the example.

import numpy as np
import matplotlib.pyplot as plt
from matplotlib.colors import LogNorm
from matplotlib.ticker import LogLocator

from astrodendro import Dendrogram
from astrodendro import periodic_neighbours
from astrodendro.analysis import PPStatistic
from astrodendro import pp_catalog

from astropy.io import fits
from astropy.table import Table
from astropy import units as u
############
fig1 = plt.figure(figsize=(12, 12))

periodic_axis = [0, 1]  # data wraps along the y and x axes

ax = plt.subplot(1, 1, 1)
image = fits.getdata('my_fits.fits')
d = Dendrogram.compute(image, min_value=0.05, min_delta=0.01, min_npix=1, 
                       neighbours=periodic_neighbours(periodic_axis))
p = d.plotter()
im = ax.imshow(image, origin='lower', interpolation='nearest', 
                cmap='Blues', norm=LogNorm(vmin=0.007, vmax=1.1))
fig1.colorbar(im)

for leaf in d.leaves:
    p.plot_contour(ax, structure=leaf, linewidths=0.3, colors='blue')
    p.plot_contour(ax, structure=560, linewidths=0.6, colors='red') ## this is one of straddling structures

    s = PPStatistic(leaf)
    ellipse = s.to_mpl_ellipse(edgecolor='green', linewidth=1, facecolor='none')
    ax.add_patch(ellipse)

metadata = {}
metadata['data_unit'] = u.Jy

cat = pp_catalog(d.leaves, metadata, verbose=False)
cat.write('test_catalog_periodic.txt', format='ascii', overwrite=True)

## load data from catalog 
iid, area, mass, x, y = np.loadtxt('test_catalog_periodic.txt', 
                                   skiprows=1, usecols=(0,2,3,8,9), unpack=True)

for _ in range(0, iid.size):
    if iid[_] == 560:
        ax.scatter(x[_], y[_], s=10, color='k')
        ax.annotate(iid[_], (x[_], y[_]))

plt.show()

This is the result.