gaia-dr3-plot-with-dust

name	gaia-dr3-plot-with-dust
description	Retrieve nearby Gaia DR3 stars (via the AIP TAP service) and produce a two-panel research-paper figure — RA/Dec on the left, a Galactic XY projection overlaid with the SFD all-sky dust-extinction map on the right (dustmaps package).
version	2.0.0
author	Hermi (sorgenfresser) / AIP
license	MIT
prerequisites	{"python":["pyvo","pandas","pyarrow","matplotlib","seaborn","dustmaps","scipy"]}
metadata	{"hermes":{"tags":["astronomy","gaia-dr3","dust","extinction","plotting","pyvo"],"category":"astronomy","related_skills":["gaia-dr3-tap-query","cmd-plotting","seaborn-paper-plots"]}}

Gaia DR3 — paper-style figure with dust overlay

When to Use

Use when you want a polished, manuscript-style figure of a Gaia sample set against the Milky Way's dust lanes. For plain Gaia queries (no dust), use gaia-dr3-tap-query.

Dependencies are pre-installed in this skill's dedicated venv. Run every command with {{SKILL_PYTHON}}; never pip install. The SFD dust map (~167 MB) downloads once on first use via dustmaps.sfd.fetch() (needs internet); it is cached after.

Procedure

1. Query Gaia DR3 (nearest stars)

import warnings, os
warnings.filterwarnings("ignore")
import pyvo, pandas as pd
service = pyvo.dal.TAPService("https://gaia.aip.de/tap/")
q = """SELECT TOP 100 source_id, ra, dec, l, b, parallax, phot_g_mean_mag, bp_rp
       FROM gaiadr3.gaia_source
       WHERE parallax > 10
       ORDER BY parallax DESC"""
df = service.run_sync(q).to_table().to_pandas()
out = "gaia-dr3-dust"; os.makedirs(out, exist_ok=True)
df.to_parquet(f"{out}/gaia_dr3_closest.parquet", index=False)

2. Build the dust-extinction background (SFD)

import dustmaps.sfd
dustmaps.sfd.fetch()                       # one-time download; cached afterwards
from dustmaps.sfd import SFDQuery
from astropy.coordinates import SkyCoord
import astropy.units as u
import numpy as np
l_grid = np.linspace(0, 360, 720); b_grid = np.linspace(-90, 90, 360)
L, B = np.meshgrid(l_grid, b_grid)
ebv = SFDQuery()(SkyCoord(l=L*u.deg, b=B*u.deg, frame="galactic"))   # E(B-V), mag

3. Two-panel figure

import matplotlib; matplotlib.use("Agg")
import matplotlib.pyplot as plt
fig, axs = plt.subplots(1, 2, figsize=(12, 6), facecolor="white")
# Left: RA/Dec
sc = axs[0].scatter(df["ra"], df["dec"], c=df["parallax"], cmap="viridis",
                    s=60, edgecolor="black", linewidth=0.4)
axs[0].set_xlabel("RA [deg]"); axs[0].set_ylabel("Dec [deg]")
axs[0].set_title("Gaia DR3 — closest stars (RA/Dec)")
fig.colorbar(sc, ax=axs[0], label="parallax [mas]")
# Right: Galactic XY + dust
axs[1].imshow(ebv, extent=[-1.1, 1.1, -1.1, 1.1], origin="lower",
              cmap="gray_r", alpha=0.4)
lr = np.radians(df["l"].values); br = np.radians(df["b"].values)
x = np.cos(br)*np.cos(lr); y = np.cos(br)*np.sin(lr)
sc2 = axs[1].scatter(x, y, c=df["parallax"], cmap="viridis", s=70,
                     edgecolor="black", linewidth=0.5)
axs[1].scatter(0, 0, c="gold", s=300, marker="o", edgecolor=)
axs[].annotate(, xy=(, ), xytext=(, ), color=, fontweight=,
                arrowprops=(arrowstyle=, color=))
axs[].set_xlabel(); axs[].set_ylabel()
axs[].set_title(); axs[].set_aspect()
axs[].set_xlim(-, ); axs[].set_ylim(-, )
fig.colorbar(sc2, ax=axs[], label=)
plt.tight_layout(); plt.savefig(, dpi=, bbox_inches=)
()

Pitfalls

• The SFD map downloads once (dustmaps.sfd.fetch()); needs internet the first time.
• Use TOP N (not LIMIT); a parallax floor (WHERE parallax > 10) bounds the sort.
• 0.5° grid steps give a smooth background at modest memory; coarsen for speed.
• Save outputs under the workspace topic folder — never /home/hermes/ or /tmp.

Verification

• {out}/gaia_dr3_closest.parquet and {out}/gaia_dr3_with_dust.png exist.
• The figure shows stars colour-coded by parallax over a faint dust background.