GeoEnergyMath Laplace's Tidal Equation modeling and fitting software.

Use the bundled PowerShell script to install/build on Windows (GNAT Community 2021 at C:\GNAT\2021\bin) and to stage the executable for both the CLI and GUI.

1. Install GNAT Community 2021 (the script can open the download page).
2. Run .\setup_with_gnat.ps1 from the repository root.
3. The script builds enso_opt and deploys:
   • run\lt.exe (CLI/runtime)
   • experiments\Feb2026\lt.exe (GUI launcher)

The experiments/Feb2026 directory already ships with a precompiled lt.exe so you can launch the GUI immediately, but rebuilding with the script keeps the binary current with source changes.

The main Ada driver is src/enso_opt.adb. It is built into lt.exe (or lt on Linux/macOS) via the GNAT project file lte.gpr.

1. Build: gprbuild -P lte.gpr enso_opt (or gprbuild -P lte.gpr), or run setup_with_gnat.ps1 on Windows.
2. Stage the executable: copy obj/enso_opt.exe to run/lt.exe and (for the GUI) experiments/Feb2026/lt.exe (see update_exe.bat, update_exe.ps1, or setup_with_gnat.ps1).
3. Runtime flow:
   • Reads environment configuration such as NUMBER_OF_PROCESSORS, DLOD_DAT, TIMEOUT, DLOD_SCALE, and EXPECT.
   • Loads the dLOD reference data (default ../dLOD3.dat) and parameter files through the shared primitives (the run/lt.exe.par* and run/lt.exe.resp* files are examples).
   • Starts the solver tasks, periodically printing status and checking for console input.
   • Interactive console controls: q/s stop, x exits without saving, and 1–9 trigger stored solution outputs.
   • The TIMEOUT cycle stops the run when the elapsed time is exceeded.

For a more detailed walkthrough, see the GEM-LTE wiki:

The interactive GUI lives in experiments/Feb2026/lte_gui.py and is launched from a Windows command prompt (Linux support is planned; currently the launcher uses cmd.exe). It helps you select climate indices and mean sea level (MSL) sites to model and simulate.

1. Run the GUI from the experiments\Feb2026 level (the script expects to run at that level and searches subdirectories). The GUI launches the experiments\Feb2026\lt.exe binary that ships with the dataset or is updated by setup_with_gnat.ps1: python .\lte_gui.py.
2. Choose the root directory that contains index folders (for example, experiments\Feb2026). The GUI lists each directory except locs, scripts, and rlr_data.
3. Select an index directory. If the directory name matches an entry in ID.yml, the GUI shows the MSL site name/country metadata (or climate index).
4. Configure the TIMEOUT, metric (CC/DTW), and test/training interval.
5. Click Run lt to launch lt.exe in a new console with environment variables (METRIC, TIMEOUT, TRAIN_START, TRAIN_STOP, CLIMATE_INDEX, IDATE) set for the run. Once finished, can close it.
6. Click Run plot to execute plot.py for the selected index and Refresh PNG to preview the newest plot in the GUI.

Selecting the Warnemunde MSL station, index #11

after Refresh PNG

Notes to mmyself and potential maintainers. Releases should package the compiled lt.exe, the run/ data/parameter files, and the experiment directories used with lte_gui.py, including the staged experiments/Feb2026/lt.exe copy. Probbaly should tag the commit and create a release from the tag in GitHub.