GNEISS Stackplot

This workspace contains scripts and prepared data products for an interactive stackplot of GNEISS rocket measurements. The Dash app aligns the panels on a shared x-axis and lets the viewer show or hide panels and switch between time since TG and magnetic latitude.

Workspace Layout

GNEISS_stackplot/
├── data/
│   ├── app_data/
│   │   ├── b_e_field_components_data.h5
│   │   ├── brightness_vs_time_20260210_101900_102848_step0p05.h5
│   │   ├── brightness_vs_time_20260210_101900_102848_step0p05_avg25.h5
│   │   ├── chimps_397_downgoing_data.h5
│   │   ├── erau_signal_data.h5
│   │   ├── erpa_hi_data.h5
│   │   ├── erpa_temp_data.h5
│   │   ├── exb_components_data.h5
│   │   ├── footpoint_brightness_data.h5
│   │   ├── pip3_0_voff_data.h5
│   │   ├── tg_to_maglat.csv
│   │   └── trajectory_keogram_green_20260210_101900_102848.h5
│   └── source_data/  # local raw/intermediate files; not distributed on GitHub
└── scripts/
    ├── build_app_data/
    │   ├── b_e_field.py
    │   ├── erau.py
    │   ├── erpa_hi.py
    │   ├── erpa_temp.py
    │   └── footpoint_brightness.py
    ├── stackplot_app.py
    └── requirements.txt

The build scripts create full-resolution HDF5 data products. The preferred lab layout stores source and app data outside this repository. Set:

export LAB317_DATA_ROOT="/Users/anniepflaum/lab317/data"

This resolves source data from raw/rocket/gneiss-stackplot and app data from processed/gneiss/stackplot. GNEISS_SOURCE_DATA_DIR and GNEISS_APP_DATA_DIR are direct overrides. Without these variables, scripts retain the repo-local data/ fallback. See config/example.env and the checksum catalogs in data/source_manifest.csv and data/app_manifest.csv.

Download the Prepared App Data

The prepared data files are distributed through this Google Drive folder, not through GitHub.

1. Open the Google Drive link in a browser. Sign in or request access if Google Drive prompts you to do so.
2. Select all files in the folder and choose Download. Google Drive may package the files into one or more ZIP archives.
3. Extract the downloaded archive or archives.
4. Copy the files themselves into GNEISS_stackplot/data/app_data/. Do not put an additional Google Drive folder or ZIP file inside app_data.
5. Confirm that the following paths exist directly inside data/app_data/:

b_e_field_components_data.h5
brightness_vs_time_20260210_101900_102848_step0p05.h5
brightness_vs_time_20260210_101900_102848_step0p05_avg25.h5
chimps_397_downgoing_data.h5
erau_signal_data.h5
erpa_hi_data.h5
erpa_temp_data.h5
footpoint_brightness_data.h5
pip3_0_voff_data.h5
tg_to_maglat.csv
trajectory_keogram_green_20260210_101900_102848.h5

exb_components_data.h5 may also be included in the download, but the ExB panels are currently hidden and the app does not require that file to start.

From the repository root, the installation can be checked with:

ls -lh data/app_data

If a required filename is missing or nested inside another directory, the app will stop during startup with a file-not-found error.

Run the Stackplot App

From the workspace root:

cd scripts
python3 -m pip install -r requirements.txt
python3 stackplot_app.py

Open http://127.0.0.1:8051 in a browser.

The app includes:

• ERAU PIP
• ERPA temperature
• ERPA hi
• CHIMPS downgoing electron energy-time spectrogram
• CHIMPS downgoing electron total counts
• PIP3 low-gain Voff for rockets 397 and 398
• (ExB)/B^2 east, north, and up (hidden as of 7/20/26)
• B north and E east
• B east and E north
• Footpoint brightness at 110 km
• Averaged-versus-unaveraged trajectory brightness at 110 km
• Trajectory keograms for rockets 398 and 397

Use the checkboxes to show or hide panels. Use the x-axis control to switch all panels between time since TG and magnetic latitude.

Required Data

scripts/stackplot_app.py expects the prepared HDF5 files and tg_to_maglat.csv listed above under data/app_data/.

data/source_data/ contains local raw and intermediate inputs used to regenerate the prepared HDF5 files. The stackplot app does not read that folder directly. Its contents are excluded by .gitignore and must not be uploaded to GitHub.

Data Processing Notes

The prepared HDF5 files retain the source resolution. The app applies a few display-oriented transformations:

• B/E field, ExB, ERPA hi, ERPA temperature, and ERAU PIP series are written to HDF5 at their full processed source cadence without export-time downsampling.
• ERAU PIP is additionally sorted by TG time, duplicate times are reduced to the first sample, the configured TG offset is applied, and a centered 100-sample rolling median is applied before export.
• CHIMPS 397 downgoing-electron data are sorted by time, duplicate times are reduced to the first sample, and log10_counts is stored as log10(max(counts, 1)) for the energy-time spectrogram. The total-count line is stored without smoothing.
• PIP3 low-gain Voff uses the positive time samples from the sigmoid-fit exports to match the source notebook, and stores one separate series each for rockets 397 and 398 without smoothing or downsampling.
• The Dash app does not interpolate panels onto a shared time grid. Lower-rate panels use their stored HDF5 timestamps directly. Full-view B/E and ERAU traces use an extrema-preserving display aggregate; zooming reads only the visible HDF5 interval and returns every native sample once the interval fits within the display point limit.
• Magnetic latitude is not an independent source measurement in the HDF5 series; it is linearly interpolated from data/app_data/tg_to_maglat.csv for the selected rocket and TG time samples.
• Keogram brightness is displayed as log10 brightness with clipping at the stored brightness limits. Footpoint brightness retains the finite timestamps from the prepared brightness CSV.