No results yet
Go to the Input tab to load data and run analysis
Tracer Metrics:
-
Pore Water Vel. (cm/day)
TTD Model Comparison
| Model | R² | RMSE | AIC |
|---|
| Lognormal | - | - | - | - |
| Gamma | - | - | - | - |
About TTD Toolkit
Purpose
This toolkit fits transit time distribution (TTD) models to tracer breakthrough curves from pulse-labeling experiments. It implements Nelder-Mead optimization to fit lognormal and gamma distributions, providing model comparison via AIC/BIC criteria.
Our transit time method is similar to Evaristo et al. (2019, Water Resources Research) paper. Briefly, under approximately steady water fluxes during tracer passage, the first temporal moment of a concentration breakthrough curve (BTC) is proportional to the first moment of the corresponding mass-flux BTC, so our concentration-based “mean transit times” can serve as useful approximate summaries of water residence or transit times.
Input Requirements
- CSV with Date and δ²H columns
- Reference date (tracer injection)
- Sample depth (for velocity calculation)
- Background/pulse δ²H if not normalized
Output Metrics
- t_peak, t_mean, f_pulse
- Pore water velocity
- Model parameters (shape, scale)
- R², RMSE, AIC, BIC
Citation
Evaristo, J., Wright, C.W., Bauser, H., Knighton, J.K., Johnson, D., Kim, M. (2026). Tracer Labeling and Transit Time Modeling in Soil-Plant Systems: Perspectives and a Call for Broader Dialogue in Ecohydrology. Ecohydrology. DOI: 10.1002/eco.70182
Technical Notes
This web implementation uses Nelder-Mead optimization with multi-start initialization to match scipy.optimize.curve_fit behavior. For bootstrap uncertainty analysis and publication-quality figures, use the companion Python toolkit.