Extract LVM Target Spectra

This repository contains a small Python tool to automatically extract spectra from SDSS-V Local Volume Mapper (LVM) observations for a list of targets.

The script takes as input the cross-match table produced by Knox Long's find_obs.py routine, identifies the corresponding LVM exposures, and extracts spectra from the nearest fibers to each target.

Only the relevant spectra are saved, avoiding the need to download or store entire LVM datasets.

---

Overview

The script performs the following steps:

1. Reads the match file produced by the drpall cross-match (e.g. candidatos_lvm.matched.txt).
2. Groups targets by exposure number (expnum).
3. Downloads the corresponding lvmSFrame file from the SDSS Science Archive Server (SAS).
4. Identifies the nearest N fibers to each target position.
5. Extracts the spectra from those fibers.
6. Saves the extracted spectra in compressed .npz files.
7. Deletes the temporary lvmSFrame file to minimize disk usage.

This allows efficient extraction of spectra for large target lists without downloading full LVM datasets.

---

Requirements

Python ≥ 3.8

Required packages:

numpy
pandas
astropy
requests

Install them with:

pip install numpy pandas astropy requests

---

Input

The script expects the cross-match output produced by Knox Long's find_obs.py script.

Example file:

candidatos_lvm.matched.txt

This file contains the list of targets and the LVM exposures that overlap their positions.

---

Usage

Example command:

python extract_lvm_target_spectra.py   --matched candidatos_lvm.matched.txt   --outdir extracted_spectra   --redux-version 1.2.1   --radius-arcsec 15   --nclosest 2

---

Parameters

--matched

Cross-match file produced by find_obs.py.

Example:

candidatos_lvm.matched.txt

---

--outdir

Directory where the extracted spectra will be saved.

Example:

extracted_spectra/

---

--redux-version

LVM redux version stored on the SDSS SAS.

Example:

--redux-version 1.2.1

---

--radius-arcsec

Maximum radius (arcseconds) used to select nearby fibers.

Example:

--radius-arcsec 15

---

--nclosest

Maximum number of closest fibers to extract per target.

Example:

--nclosest 2

---

Output

The script produces:

Extracted spectra

Compressed .npz files containing:

• wavelength array
• flux
• flux error
• fiber IDs
• fiber coordinates
• angular separation from the target

Example file:

cand_001_exp15294.npz

---

Summary table

A CSV file listing all extracted fibers:

extracted_spectra_summary.csv

Columns include:

• target name
• target coordinates
• exposure number
• tile ID
• fiber ID
• angular separation
• output file name

---

Data Access

The script downloads lvmSFrame files directly from the SDSS Science Archive Server (SAS).

Access requires valid SDSS credentials.

Credentials must be provided in the script using:

USER = "your_sdss_username"
PASSWORD = "your_sdss_password"

---

Notes

• The script downloads only the necessary exposures and deletes them after processing to reduce disk usage.
• Multiple targets falling within the same exposure are processed simultaneously.
• The number of fibers extracted per target can be controlled using --nclosest.

---

Acknowledgments

This tool relies on the cross-matching routine developed by Knox Long (find_obs.py) to identify LVM exposures overlapping the target positions.

LVM data are part of the SDSS-V Local Volume Mapper survey.

---

License

MIT License (or specify your preferred license).