siim — Sliding Ice Incision Model

Coupled glacial–fluvial landscape evolution. siim models bedrock incision by sliding ice and rivers together, resolving how glaciers and channels jointly shape mountain topography over geologic time. It ships three coordinated front ends over one physical core:

• siim.siim1d — fast 1D longitudinal-profile model
• siim.siim2d — full 2D landscape-evolution model (in-house driver, D8/D-inf routing, FFT flexure, ADI hillslope diffusion)
• siim.analytical — closed-form steady-state profiles and regime diagrams from the underlying theory

Python with NumPy/SciPy/Numba — no Fortran and no framework lock-in for the standalone models.

Install

pip install siim-lem

siim runs the full 1D and 2D models standalone on NumPy 2. An optional adapter lets you drive the 2D model through fastscape/xsimlab instead of the built-in driver; that stack is conda-only:

conda env create -f environment.yml   # optional fastscape adapter

Requires Python ≥ 3.12.

Quickstart

from siim import siim2d

params = {
    'U': 1e-3,                          # rock uplift rate (m/yr)
    'zELA': 1500, 'P': 2, 'beta': 1e-2, # climate: ELA + mass balance
    'ce': 2e-5, 'Ko': 1e-6,             # glacial + fluvial erodibility
    'Lx': 100e3, 'Ly': 50e3,            # domain size (m)
    'nx': 201, 'ny': 101,               # grid
    'T': 5e6, 'nt': 501, 'nt_out': 51,  # 5 Myr run
    'seed': 0,                           # reproducible initial relief noise
    'mode': 'C',                        # bed-memory + sub-grid glacier carving
}

model = siim2d.siim(params)
model.run()
model.plot.landscape(i=-1)             # final topography

The 1D model (from siim.siim1d import siim) and the analytical solutions (from siim.analytical import GeneralProfile) follow the same shape. See the documentation for the full parameter reference, guides, and the theory background.

Citing

If you use siim, please cite it via CITATION.cff. A methods paper is in preparation.

License

MIT © Eric Deal