The LIGO-Virgo-KAGRA network has now detected 90 signals from black holes and neutron stars colliding, and the next generation of detectors will see even more. But current methods for analysing these signals are too slow and too imprecise to extract everything the data can tell us. This programme tackles that bottleneck by building faster, more accurate tools. The team will develop their machine learning code DINGO to estimate the properties of colliding objects in minutes rather than days, and create a new framework for testing whether Einstein’s general relativity holds up under extreme gravitational conditions. They will also run numerical simulations to model what gravitational waves would look like if new physics is present, then use those simulations to calibrate their search methods. If successful, this work will let researchers test fundamental physics with far greater precision than is possible today. This is fundamental science: it will not change daily life directly, but it could reveal whether our current understanding of gravity is incomplete—a question that, if answered, would reshape physics itself.
View original technical description
During the first three Observing Runs, the LIGO-Virgo-KAGRA (LVK) network of gravitational wave detectors observed 90 signals from pairs of black holes and compact stars spiraling into each other and eventually coalescing. The Fourth Observing Run is ongoing and the Fifth is expected to follow in 2027. Every run follows an upgrade that allows the LVK to "see" further into the universe and further into the past, drastically increasing the number of observed events. More sensitive ground-based detectors and a space-borne detector are expected to start operating in the 2030s and continue this trend. Several research communities, including astrophysics, cosmology, and theoretical physics, are already considering gravitational waves as one of the main discovery probes in their search for new physics. Satisfying their needs and harnessing the full discovery potential of current and future observations requires faster data analysis methods, more accurate parameter estimation, and high-precision modelling of the new physics one hopes to discover. Our programme is designed to push the boundaries of all three of these frontiers. We will develop machine learning methods and our code DINGO for fast parameter estimation that is more accurate than existing methods. We will introduce a new approach for parametrising deviations from the predictions of general relativity in the inspiral phase, which will enable the LVK to efficiently and consistently combine multiple signals to test general relativity. We will also employ powerful numerical simulations to model gravitational wave signals that include such deviations and use them to benchmark our parametrization and calibrate waveform models. Finally, we will train DINGO to search for astrophysical environments and for new fundamental physics in gravitational wave data. Our goal is to maximise LVK capacity to discover new physics, broaden its exploitation potential and its science community, and set the stage for the exciting era of gravitational wave astronomy with the next generation of detectors.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know