Hyderabad P-Receiver Function Crustal Thickness and Vp/Vs Robustness Study
A completed geology and seismology study estimating first-order crustal thickness and Vp/Vs beneath Hyderabad from public teleseismic waveforms, with explicit data quality and robustness checks.

Software compatibility
The released analysis, retained results, figures, tests, and documentation use ObsPy 1.5.0 with the pinned Python 3.14 environment and the checksummed GEOSCOPE HYB dataset. Other versions, stations, data windows, or processing parameters require a new verification run.
Project definition
Problem statement
Receiver functions can estimate bulk crustal thickness and Vp/Vs from converted seismic phases, but the result depends on waveform quality, polarity, deconvolution, assumed crustal velocity, phase weights, and source direction.
The geology problem is to produce a traceable station-average estimate beneath Hyderabad while separating event-sampling precision from model and directional sensitivity.
Project objectives
- Acquire and preserve a balanced public teleseismic dataset for GEOSCOPE station HYB.
- Remove instrument response, rotate three-component records, and estimate radial and transverse receiver functions.
- Apply explicit quality rules and retain every metric and rejection reason.
- Estimate crustal thickness and Vp/Vs through H-kappa stacking.
- Test event resampling, assumed Vp, phase weights, backazimuth sectors, and synthetic recovery.
Project structure
Project components
Data acquisition
Retrieves the event catalogue, StationXML, and balanced three-component waveforms with checksums and provenance.
Signal processing
Corrects instrument response, filters, rotates to LQT, applies the radial polarity convention, and performs water-level deconvolution.
Quality control
Measures signal-to-noise ratio, transverse energy, deconvolution fit, and receiver-function amplitude for every event.
Crustal estimator
Calculates event-specific converted-phase times and the weighted H-kappa stack over the declared search grid.
Robustness analysis
Runs 1,000 bootstraps, five assumption cases, four directional sectors, and four synthetic noise experiments.
Methodology
Project workflow
- 01Verify retained data
The pipeline validates the StationXML, waveform inventory, event manifest, and declared checksums.
- 02Build receiver functions
Each accepted triplet is corrected, filtered, rotated, deconvolved, measured, and retained with its event metadata.
- 03Search H and Vp/Vs
Predicted Ps, PpPs, and later-multiple times are sampled for every accepted event and grid cell.
- 04Test robustness
Bootstrap, model, weight, directional, and synthetic experiments distinguish stable evidence from assumptions.
- 05Retain evidence
CSV, JSON, NPZ, PNG, Word, and PDF outputs preserve the complete chain from input records to conclusions.
Demonstration scenario
The retained sample accepts 48 of 55 receiver functions. With Vp fixed at 6.4 km/s, the reference stack estimates H at 32.4 km and Vp/Vs at 1.750. Event bootstraps span 32.2 to 32.6 km, while plausible fixed-Vp cases move H from 31.2 to 33.8 km, showing why model sensitivity must remain separate from sampling precision.
Engineering
Tools and method
- Tools
- The project uses Python 3.14, ObsPy 1.5.0, NumPy, SciPy, Pandas, Matplotlib for subject analysis, simulation, and results.
- Seismic data layer
- ObsPy reads MiniSEED and StationXML, calculates travel times, removes response, and performs coordinate rotation.
- Numerical layer
- NumPy and SciPy implement spectral deconvolution, interpolation, H-kappa contributions, resampling, and diagnostics.
- Experiment layer
- A configuration-controlled Python pipeline runs the reference study and every declared robustness case.
- Evidence layer
- Pandas, JSON, CSV, NPZ, and Matplotlib retain event decisions, numerical arrays, summaries, and labelled figures.
- Verification
- Automated tests, coverage, dependency audit, Docker execution, document audits, and visual review verify the delivery.
Testing
Evaluation
Evaluation measures
- Accepted waveform count and each per-event quality-control metric
- Reference H-kappa maximum, score, H, and Vp/Vs
- Conditional 2.5th and 97.5th event-bootstrap percentiles
- Sensitivity to fixed Vp and phase-family weights
- Directional spread across four backazimuth sectors
- Synthetic recovery at four noise levels
- Automated tests, coverage, dependency audit, document audit, and container execution
Project boundaries
- The result is a single-station, one-dimensional, one-layer H-kappa appraisal rather than a unique crustal velocity-depth model.
- The bootstrap interval measures accepted-event sample stability under fixed processing and model assumptions.
- Four broad sectors can reveal directional variation but cannot uniquely resolve dip, anisotropy, or lateral structure.
- Geological interpretation beyond the retained evidence requires independent velocity, gravity, active-source, mapping, or multi-station constraints.
Included
- 01Complete Python and ObsPy source code
- 02Fifty-five retained public three-component GEOSCOPE waveform records
- 03StationXML, event manifest, checksums, and per-event quality-control evidence
- 04Forty-eight accepted radial and transverse receiver functions
- 05H-kappa surface, 1,000 bootstrap estimates, five sensitivity cases, and four backazimuth-sector estimates
- 06Four synthetic recovery cases and sixteen generated project figures
- 07Twenty-three automated tests with 97.30 percent branch-aware coverage
- 08Complete project files, calculations, results, and analysis material in a private GitHub repository
- 0999-page project documentation in PDF and editable Word formats
- 1012-page setup and usage guide in PDF and editable Word formats
- 11Forty-five annotated references and two attributed current literature figures
Project record
No information is collected on this page.
- Permanent project ID
- GP-GE-1HWIV61
- Catalogued
- 21 Aug 2026
- Completed
- 28 Aug 2026
- Verified
- 28 Aug 2026
- Demonstration
- Included in repository
Handover
After purchase
- 01Payment is confirmed
The project is marked unavailable and cannot be purchased again.
- 02Repository access is granted
The buyer's submitted GitHub account receives access to the private repository.
- 03The purchase record is delivered
The certification sheet is prepared from the reviewed buyer details and sent privately by email.