spelacew 0.1.0

Interactive spectral line fitting and Equivalent Width measurement tool

SpelacEW

Interactive spectral line fitting and Equivalent Width (EW) measurement tool for stellar spectroscopy.

SpelacEW is a Python-based interactive software designed for the analysis of stellar absorption lines through manual continuum placement and Gaussian profile fitting. The code allows the user to inspect spectral regions individually, perform single-line or multi-line (blended) fits, and export reproducible measurements in a structured format.

The project was developed for high-control spectroscopic analysis workflows where visual inspection and user-guided continuum selection are essential.

---

Features

• Interactive visualization of stellar spectra
• Support for:
  • FITS spectra
  • ASCII/TXT/CSV spectra
• Manual continuum placement
• Gaussian fitting of absorption lines
• Multi-Gaussian fitting for blended lines
• Automatic calculation of:
  • Equivalent Width (EW)
  • Full Width at Half Maximum (FWHM)
  • Reduced chi-square (χ²)
• Reference continuum visualization from external datasets
• Interactive spectral exploration mode
• Automatic generation of:
  • CSV results tables
  • Multi-page PDF reports
• Session accumulation:
  • Existing results are preserved between runs
  • PDFs are automatically merged across sessions

---

Scientific Motivation

Precise measurement of spectral line parameters is a fundamental task in stellar spectroscopy and abundance analysis. While fully automated EW pipelines are efficient for large datasets, many scientific applications still require manual inspection due to:

• blended spectral features
• continuum uncertainties
• local noise structures
• line asymmetries
• crowded spectral regions

SpelacEW was designed as a semi-manual high-control alternative, prioritizing:

• reproducibility
• visual inspection
• interactive continuum definition
• robust treatment of blended lines

This makes the code particularly useful for:

• stellar abundance studies
• chemical tagging
• solar reference comparisons
• educational spectroscopy workflows
• spectroscopic quality control

---

Installation

Clone repository

git clone https://github.com/USERNAME/spelacew.git
cd spelacew

Install package

Recommended installation:

pip install -e .

This enables the command-line interface:

spelacew

---

Dependencies

Main dependencies:

• numpy
• scipy
• matplotlib
• pandas
• PyAstronomy
• pypdf

Python >= 3.9 is required.

---

Input Data

Spectrum

Supported formats:

• .fits
• .fit
• .txt
• .dat
• .ascii
• .csv

ASCII files must contain:

wavelength flux

in two columns.

---

Line List

The line list must be provided as a CSV file.

Minimum required column:

wavelength

Optional columns:

• element
• species
• ep
• gf

Example:

wavelength,element,species,ep,gf
6562.79,H,1,10.2,-0.3

---

Usage

Interactive mode

Run:

spelacew

The program will ask interactively for:

• spectrum file
• line list
• optional reference file
• optional width

---

Command-line mode

spelacew spectrum.fits lines.csv

Optional arguments:

spelacew spectrum.fits lines.csv width

or

spelacew spectrum.fits lines.csv width reference.csv

Example:

spelacew HD10700.fits lines_fe.csv 1.5 solar_reference.csv

---

Interactive Interface

The interface displays:

• observed spectrum
• theoretical line center
• crosshair cursor
• local spectral window
• automatic zoom panel after fitting
• reference continuum visualization (optional)

The user interacts entirely through keyboard commands.

---

Controls

Navigation

Key	Action
n	Next spectral line
p	Previous spectral line

---

Normal Fitting Mode

Key	Action
k	Select continuum points

Procedure:

1. Press k on left continuum region
2. Press k on right continuum region
3. Fit executes automatically

The continuum is estimated through a linear interpolation between the selected points.

---

Blending Mode

Key	Action
b	Enable/disable blending mode
d	Define fitting region
g	Add Gaussian component center
Enter	Execute multi-Gaussian fit

This mode is intended for partially or strongly blended spectral features.

Each Gaussian component is fitted simultaneously.

---

Exploration Mode

Key	Action
x	Enable exploration mode
c	Manual zoom input
w	Change visualization width

Exploration mode allows free navigation through the spectrum independently of the predefined line list.

---

Other Commands

Key	Action
r	Full reset
a	Cancel input
q	Save and quit

---

Outputs

For each session, the program generates a directory:

ajustes_<spectrum_name>_<date>/

Example:

ajustes_HD10700_2026-05-06/

---

CSV Results

File:

resultados.csv

Contains:

Column	Description
wavelength	Central wavelength
wave_left	Left fitting boundary
wave_right	Right fitting boundary
left_continuum	Left continuum level
right_continuum	Right continuum level
element	Atomic species
species	Ionization state
ep	Excitation potential
gf	Oscillator strength
ew_ref	Reference EW
ew	Measured EW
FWHM	Full Width at Half Maximum
Chi2R	Reduced χ²
hpf	Reserved parameter

Equivalent Width values are reported in milli-Angstroms (mÅ).

---

PDF Report

File:

resultados.pdf

Contains one page per fitted line including:

• observed spectrum
• continuum model
• Gaussian fit
• zoom visualization
• fit diagnostics

PDF sessions are automatically merged between runs.

---

Fitting Methodology

Continuum Normalization

The continuum is defined manually using two user-selected points.

A linear continuum model is constructed through:

F_c(\lambda) = a\lambda + b

The observed spectrum is normalized as:

F_{norm} = \frac{F}{F_c}

---

Gaussian Model

Single-line fits use:

F(\lambda) =
1 - A \exp
\left(
-\frac{(\lambda-\mu)^2}{2\sigma^2}
\right)

where:

• (A) = line depth
• (\mu) = line center
• (\sigma) = Gaussian width

---

Multi-Gaussian Blending

For blended lines:

F(\lambda)=
1-
\sum_i
A_i
\exp
\left(
-\frac{(\lambda-\mu_i)^2}{2\sigma_i^2}
\right)

Each component is fitted simultaneously.

---

Equivalent Width

Equivalent Width is computed numerically as:

EW =
\int
(1-F_{norm}) d\lambda

and reported in milli-Angstroms (mÅ).

---

Recommended Usage

For more stable fits:

• select continuum regions wider than the line core
• avoid noisy edge regions
• inspect blended features manually
• verify continuum placement visually

Broader continuum selections generally improve normalization stability and reduce sensitivity to local noise fluctuations.

---

Current Status

Current implemented features:

• Interactive continuum selection
• Single Gaussian fitting
• Multi-Gaussian blending
• EW/FWHM/χ² calculations
• PDF and CSV export
• Reference EW comparison
• Spectral exploration mode
• Session accumulation

Planned future developments:

• automatic continuum estimation
• Voigt profile support
• radial velocity correction tools
• abundance pipeline integration
• GUI improvements
• batch-processing utilities

---

Citation

If you use SpelacEW in scientific work, please cite the repository and acknowledge the software appropriately.

---

Author

Daniel Pérez
Universidad de Concepción

---