Coordinate Systems¶

OVRO-LWA data uses multiple coordinate systems. Understanding them is essential for analysis.

Pixel Coordinates (l, m)¶

The primary coordinate system uses direction cosines:

l: East-West direction cosine (dimensionless)
m: North-South direction cosine (dimensionless)
Range: typically -1 to +1
Center: (l=0, m=0) is the phase center

import ovro_lwa_portal as ovro

ds = ovro.open_dataset("path/to/data.zarr")

# Access pixel data
intensity = ds['SKY'].sel(l=0.1, m=0.2, method='nearest')

World Coordinate System (WCS)¶

If WCS information is present, you can work with celestial coordinates:

Check for WCS¶

# Check if dataset has WCS
if ds.radport.has_wcs:
    print("WCS available")
else:
    print("No WCS information")

Pixel to Sky Coordinates¶

Convert pixel indices to RA/Dec at a given time (same epoch as time_idx on the dataset):

# Nearest pixel to l≈0.1, m≈0.2 (use indices from coords or searchsorted)
l_idx = int(abs(ds.coords["l"] - 0.1).argmin())
m_idx = int(abs(ds.coords["m"] - 0.2).argmin())
ra, dec = ds.radport.pixel_to_coords(l_idx, m_idx, time_idx=0)
print(f"RA: {ra:.3f}°, Dec: {dec:.3f}°")

Sky to Pixel Coordinates¶

Convert RA/Dec to pixel coordinates:

# Convert (RA, Dec) to (l, m)
l, m = ds.radport.coords_to_pixel(ra=180.0, dec=20.0, time_idx=0)
print(f"l: {l:.3f}, m: {m:.3f}")

Plotting with WCS¶

# Plot with WCS overlay
ds.radport.plot_wcs(
    time_idx=0,
    freq_idx=0,
    projection='rectangular'  # or 'aitoff', 'mollweide'
)

Indexing Methods¶

The accessor provides helper methods to find nearest indices:

Nearest Frequency¶

# Find index for frequency closest to 40 MHz
freq_idx = ds.radport.nearest_freq_idx(freq_mhz=40.0)

# Use it
ds.radport.plot(freq_idx=freq_idx)

Nearest Time¶

# Find index for time closest to MJD
time_idx = ds.radport.nearest_time_idx(mjd=59000.5)

# Use it
ds.radport.plot(time_idx=time_idx)

Nearest Pixel¶

# Find pixel indices closest to (l, m)
l_idx, m_idx = ds.radport.nearest_lm_idx(l=0.1, m=0.2)

# Extract light curve at that location
lc = ds.radport.light_curve(l_idx=l_idx, m_idx=m_idx)

Time Coordinates¶

Time is stored as Modified Julian Date (MJD):

# Access time coordinates
times = ds.coords['time']

# Convert to datetime if needed
from astropy.time import Time
times_dt = Time(times.values, format='mjd').datetime

Frequency Coordinates¶

Frequencies are in Hz:

# Access frequency coordinates
freqs = ds.coords['frequency']

# Convert to MHz for display
freqs_mhz = freqs.values / 1e6

Beam Coordinates¶

Some datasets include beam information:

# Check for beam
if ds.radport.has_beam():
    beam = ds['BEAM']
    print(f"Beam shape: {beam.shape}")

Example: Multi-Coordinate Analysis¶

import ovro_lwa_portal as ovro

ds = ovro.open_dataset("path/to/data.zarr")

# Find indices
freq_idx = ds.radport.nearest_freq_idx(freq_mhz=40.0)
time_idx = ds.radport.nearest_time_idx(mjd=59000.5)

# Convert sky position to pixels (requires an epoch)
l, m = ds.radport.coords_to_pixel(ra=180.0, dec=20.0, time_idx=time_idx)
l_idx, m_idx = ds.radport.nearest_lm_idx(l=l, m=m)

# Extract data at that point
value = ds['SKY'].isel(
    time=time_idx,
    frequency=freq_idx,
    l=l_idx,
    m=m_idx
).values

print(f"Intensity at RA=180°, Dec=20°, 40 MHz: {value:.2f}")