BaRaN

Ba(yesian) Ra(infall extremes) N(etwork design). This repository presents Bayesian, Maximum Likelihood, and L-moments inference methods' effect on IDF curve computation — and how those choices impact urban drainage design. This GitHub repository accompanies the manuscript Uncertainty due to limited extreme-rainfall records is consequential for infrastructure design adaptation across the US (submitted to Nature Communications).

---

🌧️ Overview

This repository implements end-to-end workflows to estimate Intensity–Duration–Frequency (IDF) curves using:

• Bayesian inference
• Maximum Likelihood Estimation (MLE)
• Linear moments (L-moments)

It compares the three approaches, quantifies method-dependent biases, and propagates those differences to drainage design decisions (pipe sizing).

Figure: Comparison of pipe sizes for 2–1000-year events using Bayesian, MLE, and L-moments methods (upper center panel); parameter traces (lower right panel); and return-levels (lower left panel) for Atlanta.

---

📁 Repository Structure

BMM-IDF4DRAINAGE/
├─ Data curation/
│  ├─ AORC/
│  │  └─ AORC_data.ipynb
│  └─ CMIP6/
│     └─ NASA/
│        └─ Projected_precipitation_daily.ipynb
├─ Environment/
├─ GIF/
├─ Model/
│  ├─ Bayesian.ipynb
│  ├─ MLE.ipynb
│  ├─ lmoments.ipynb
│  └─ pipe sizing.ipynb
├─ Numerical experiments/
│  ├─ Noisy Nonstationry
│  │   ├─ Bias_Bayesian.ipynb
│  │   ├─ Bias_MLE.ipynb
│  │   └─ Bias_lmoments.ipynb
│  ├─ Noisy Stationary 
│  │   ├─ Bias_Bayesian.ipynb
│  │   ├─ Bias_MLE.ipynb
│  │   └─ Bias_lmoments.ipynb
│  └─ Stationary
│      ├─ Bias_Bayesian.ipynb
│      ├─ Bias_MLE.ipynb
│      └─ Bias_lmoments.ipynb
├─ Visualization/
│  └─ Visualization.ipynb
├─ Licence/
└─ README.md

---

🔬 Methodology (High level)

• Data curation: Ingest AORC / CMIP6 / NASA precipitation.

• Fit distributions: computes IDF GEV parameters with Bayesian, MLE, and MOM.

• Bias analysis: Quantify differences across methods (Numerical experiments).

• Design impact: Translate IDF differences into hydraulic pipe sizing.

• Visualization: Plot comparative IDF curves, bias summaries, and pipe sizes.

---

Getting Started

To run the analysis:

1. Clone the repository:

   git clone https://github.com/omidemam/BaRaN.git
   cd BaRaN

📬 Contact

For questions, feedback, or collaboration opportunities, please email me at: omid.emamjomehzadeh@nyu.edu

📚 Citation

If you use this repository in your research or projects, please cite it as follows:

BibTeX format:

@misc{Emamjomehzadeh2026BaRaN,
  author       = {Emamjomehzadeh, Omid and Qureshi, Dawar and Cook, Lauren M. and Mascaro, Giuseppe and Aghakouchak, Amir and Mahoney, Kelly and Zarei, Seyedamirhossein and Wani, Omar},
  title        = {BaRaN: Bayesian Rainfall extremes Network design},
  year         = {2026},
  note         = {GitHub repository accompanying the manuscript ``Uncertainty due to limited extreme-rainfall records is consequential for infrastructure design adaptation across the US'' (submitted to Communications Earth & Environment)},
  howpublished = {\url{https://github.com/omidemam/BaRaN.git}},
}