PSRCHIVE

PSRCHIVE
Name	PSRCHIVE
Developer	Centre for Astrophysics and Supercomputing, Swinburne University of Technology
Released	1999
Programming language	C++
Operating system	Linux, macOS, FreeBSD
Genre	astronomy software
License	GNU General Public License

PSRCHIVE PSRCHIVE is a software library and application suite designed for the analysis of pulsar astronomical data, particularly folded pulse profiles and single-pulse data from radio telescopes. Originating from collaborative efforts at the Australia Telescope National Facility and the Centre for Astrophysics and Supercomputing at Swinburne University of Technology, it interconnects with instrumentation and analysis pipelines used at facilities such as the Parkes Observatory, Arecibo Observatory, Green Bank Telescope, MeerKAT, and the Very Large Array. PSRCHIVE underpins workflows that include data reduction, calibration, timing, and polarization analysis used by research groups at institutions like Jodrell Bank Observatory, Max Planck Institute for Radio Astronomy, MIT, Caltech, and NASA centers.

Overview

PSRCHIVE provides a modular framework for handling observational products produced by backend instruments such as the CPSR2, GUPPI, ROACH, and CASPER-based systems deployed at observatories including Parkes Observatory, Effelsberg Observatory, Lovell Telescope, and GMRT. It integrates with time and frequency standards referenced to International Atomic Time, UTC, and uses ephemerides from services such as JPL and TEMPO2 for barycentering and timing model comparisons. The project has been cited in studies published in journals like the Monthly Notices of the Royal Astronomical Society, the Astrophysical Journal, and Astronomy & Astrophysics, and is used in collaborations involving teams from CSIRO, European Southern Observatory, Harvard-Smithsonian Center for Astrophysics, and Kavli Institute for Astronomy and Astrophysics.

Data formats and architecture

The core of PSRCHIVE is an object-oriented data model implemented in C++ that represents pulse archives, sub-integrations, frequency channels, and polarization states compatible with formats produced by pipelines from SIGPROC, PRESTO, DSPSR, and archival standards adopted by repositories such as the Virtual Observatory and the NASA/IPAC Infrared Science Archive. Files commonly use extensions produced by backends at Arecibo Observatory, Green Bank Telescope, Westerbork Synthesis Radio Telescope, and LOFAR. The architecture exposes APIs enabling bindings to Python, Fortran, and Perl; it interoperates with software like PSRPOP, PsrPlib, cython wrappers, and toolchains used by groups at University of Manchester, University of Amsterdam, and Potsdam Institut für Klimafolgenforschung for multi-instrument pipelines. The design supports metadata standards referenced by FITS conventions and integrates timing metadata compatible with TEMPO, TEMPO2, and PINT.

Functionality and features

PSRCHIVE implements functionality for pulse profile manipulation, polarization calibration, flux calibration, radio frequency interference excision, and single-pulse searches. Utilities support processes such as coherent dedispersion tied to Dispersion Measure solutions derived from surveys like the Parkes Multibeam Pulsar Survey and the PALFA survey at Arecibo. It offers routines for template matching used in timing analyses in conjunction with TEMPO2, PINT, and statistical packages used in publications by teams at University of British Columbia, Haverford College, and University of California, Berkeley. PSRCHIVE includes visualization tools analogous to those in matplotlib-based workflows and integrates with plotting libraries used by researchers at Princeton University and Stanford University for quality assurance and presentation.

Development and software ecosystem

Development has been coordinated by contributors affiliated with institutions such as Swinburne University of Technology, CSIRO, Jodrell Bank Observatory, and the University of Manchester, with source control historically managed with systems like Subversion and Git. The project interacts with build systems and continuous integration tools used at organizations including GitHub, GitLab, and Travis CI in collaborative projects with groups from Max Planck Institute for Radio Astronomy, NRAO, and CSIRO Astronomy and Space Science. PSRCHIVE forms part of an ecosystem alongside packages such as DSPSR, PSRCHIVE-utils, PRESTO, SIGPROC, PsrTools, and engages with data archives maintained by ATNF, Vizier, and ADS for reproducible research. Contributor networks include researchers from Monash University, University of Sydney, University of Western Australia, and international collaborators at South African Radio Astronomy Observatory and International Centre for Radio Astronomy Research.

Applications and usage in pulsar astronomy

Researchers use PSRCHIVE for timing arrays like the Parkes Pulsar Timing Array, the European Pulsar Timing Array, and collaborative efforts forming the International Pulsar Timing Array for gravitational-wave searches reported by teams at NANOGrav, EPTA, and IPTA. It supports polarimetric studies relevant to investigations by groups at Leiden University, University of Amsterdam, and Max Planck Institute for Astrophysics into magnetospheric emission and propagation effects. Surveys such as the High Time Resolution Universe survey and targeted campaigns at facilities including FAST and MeerKAT employ PSRCHIVE-based pipelines for candidate verification, single-pulse transient searches, and discovery papers authored by teams at Peking University, Southampton University, and University of Manchester.

Performance and benchmarking

Performance evaluations compare PSRCHIVE routines with real-time and offline systems like DSPSR, PRESTO, and custom backends deployed at Parkes Observatory and Green Bank Observatory; benchmarks measure I/O throughput, calibration convergence, and template-matching latency on hardware ranging from workstation clusters at Swinburne University of Technology to HPC facilities at National Computational Infrastructure and cloud resources provided by Amazon Web Services and Google Cloud Platform. Profiling tools used in optimization efforts include gprof, valgrind, and perf; comparisons reported by groups at CSIRO and MIT evaluate scaling across multicore servers and integrations with GPU-accelerated pipelines developed by teams at NVIDIA research collaborators.

Category:Astronomy software