Seabed DAS Saturation Detection

This repository provides Python workflows for analyzing seabed Distributed Acoustic Sensing (DAS) data.
It focuses on:

• First-break picking using robust, noise-adaptive methods
• Waveform coherence analysis for detecting dynamic range saturation
• Signal quality evaluation for DAS datasets

---

Repository Structure

1. First-break picking

• :contentReference[oaicite:0]{index=0}
  Contains multiple picking strategies:

• Bandpass filtering for DAS data

• Envelope-based picking using Hilbert transform

• Threshold-based picking methods

• Robust outlier removal (local + slope constraints)

• Iterative pick repair using neighbor interpolation

• Final interpolation (PCHIP) to obtain smooth arrival curves

---

2. Waveform coherence analysis

• :contentReference[oaicite:1]{index=1}

Implements neighbor-based waveform coherence using arrival-aligned windows:

• Asymmetric window around first-break arrival
• Multiple coherence metrics:
  • Pearson correlation
  • Absolute correlation
  • Cosine similarity
  • Spearman correlation
  • Cross-correlation with lag tolerance
  • Normalized MSE similarity
• Aggregation across neighboring channels

---

Requirements

numpy
scipy
dascore