Skip to contents

Kinematic Analysis

Source: vignettes/kinematic-analysis.Rmd
kinematic-analysis.Rmd

Capivara separates three questions that are often mixed together:

  1. Which spaxels belong together in emission-line kinematics?
  2. Is an axisymmetric rotating disc an adequate comparison model?
  3. Is there evidence for a more specific perturbation, such as a bar?

The first two apply to both barred and unbarred galaxies. The bar model is optional.

1. Kinematic-aware segmentation

segment_kinematics() makes line maps and clusters flux, velocity, dispersion, and profile-shape information. Full-spectrum segmentation is handled by segment() and segment_large().

library(capivara)

segments <- segment_kinematics(
  cube_path = "/data/galaxy.fits",
  redshift = 0.03,
  emission_line = "halpha",
  segmentation_mode = "kinematic",
  knn_k = 50,
  n_segments = 25,
  show_plots = TRUE
)

Choose segmentation_mode = "path_signature" to add a path-signature segmentation. The native kinematic maps are still returned.

support_mode = "starlet" is the conservative default and retains the white-light galaxy footprint. For emission-line work in a field with appreciable sky, set support_mode = "line_flux"; Capivara then uses a robust border-noise threshold on the measured line-flux map and keeps its main connected component. line_flux_sigma = 3 is a useful strict starting point; lower it if faint extended line emission is scientifically important.

2. Axisymmetric disc comparison

The default axisymmetric model returns the observed velocity map, disc model, residual map, and circular-speed profile.

result <- run_kinematic_analysis(
  cube_path = "/data/galaxy.fits",
  redshift = 0.03,
  emission_line = "halpha",
  model = "axisymmetric",
  segmentation_mode = "kinematic",
  knn_k = 50,
  n_segments = 25,
  show_plots = TRUE
)

plot(result, which = "model")

For a MaNGA LOGCUBE, leave redshift = NA_real_ to read it from the local metadata or header. For another IFU cube, provide the redshift.

3. Model modules

Use kinematic_models() to see the installed dynamical modules. Each module uses the same native maps and kinematic segmentation, so new models such as a spiral perturbation can be added without changing the upstream workflow.

kinematic_models()
#>             model             label requires_bar_angle components_plot
#> 1    axisymmetric Axisymmetric disc              FALSE           FALSE
#> 2 bisymmetric_bar   Bisymmetric bar               TRUE            TRUE

The bar-specific workflow is documented separately in vignette("bisymmetric-bar-model", package = "capivara").

Return to Get Started, or view the MaNGA examples. Region labels are categorical; velocity and dispersion are separate quantities.