Skip to content

Visualizing output with igm_viz

igm_viz is a Plotly/Dash app that renders a run's output as an interactive 3-D animation — glacier surface and bedrock, colored by any saved variable, steppable frame by frame or played back automatically. This tutorial walks through the app using the Aletsch forward-modelling tutorial output as an example.

Setup

pip install igm[viz]   # one-off install of the dash dependency

Run it from the same folder you ran igm_run in — it looks for output.nc under outputs/*/*/ by default:

cd igm-examples/aletsch
igm_viz

This opens a local web app at http://127.0.0.1:8050.

If your outputs/ folder isn't in the current directory, point igm_viz at it directly:

igm_viz --output_folder /path/to/other_experiment/outputs

The exact netCDF file to view is always picked afterward in the dashboard itself.

igm_viz overview

Picking a run and a property

Every past run found under outputs/*/*/ is listed in the model run picker at the top, so you can switch between runs without restarting the app. The property dropdown selects which variable colors the glacier surface — thickness, velocity, SMB, sliding coefficient, or any other 2-D+time field saved in the output.

Adjusting the view

  • Vertical exaggeration and glacier opacity sliders adjust the 3-D rendering.
  • Ocean plane and calving front checkboxes add a sea-level plane and vertical ice walls where the glacier meets open ocean.
  • Colorbar range can be clamped manually instead of scaling to the data's min/max.

Animating and statistics

Step through frames with the slider at the bottom of the 3-D view, or hit play to animate. Below the 3-D view, a stats panel tracks glacier volume and area over time (or over iterations, for data-assimilation output).

Data-assimilation output

igm_viz also reads data-assimilation outputs (optimize.nc, geology-optimized.nc), deriving bedrock elevation from usurf - thk when topg isn't saved. Pick one from the model run picker like any other file — no separate flag needed.

Next steps