Reconciling muted NH4 seasonality with observed concentrations in 2-D riparian transect model with low-K soils (silty clay loam)

[Bisesh Joshi]

Dear Pflotran experts,

I am modeling reactive transport in a fine-grained riparian sediment (silty clay loam) using PFLOTRAN.

Using Kx = 1–5 × 10⁻¹⁴ m² and Kz = 1–5 × 10⁻¹⁵ m² produces steady-state average ammonium concentrations that are reasonably consistent with observations. However, the simulated Darcy velocities are quite low (~0.02–0.1 m/yr), corresponding to porewater velocities of roughly ~0.04–0.2 m/yr assuming porosity ≈ 0.5. This results in long residence times and muted seasonal variability in transient simulations.

When I increase permeability by one to two orders of magnitude to improve connectivity and advective transport, ammonium is flushed out and no longer matches observations.

I am trying to understand the most defensible way to reconcile realistic solute concentrations with more realistic transport dynamics. Has anyone encountered a similar tradeoff?

Would you recommend:

• Using an effective/mobile porosity instead of total porosity for transport calculations?

• Representing preferential flow paths using a dual-domain or multiple-continuum approach?

• Keeping the lower permeability and interpreting the system as transport-limited?

• Another approach altogether?

Any suggestions or experiences with similar riparian reactive transport systems would be greatly appreciated.

Thank you!

Best regards,

Bisesh

--

Bisesh Joshi

PhD student

Water Science and Policy

Graduate College

University of Delaware

901-297-6761