Searching for gravitational-wave signals emitted by eccentric compact binaries using a non-eccentric template bank: implications for ground-based detectors

Most of the inspiralling compact binaries are expected to be circularized by the time their gravitational-wave signals enter the frequency band of ground-based detectors such as LIGO or VIRGO. However, it is not excluded that some of these binaries might still possess a significant eccentricity at a few tens of hertz. Despite this possibility, current search pipelines—based on matched filtering techniques—consider only non-eccentric templates. The effect of such an approximation on the loss of signal-to-noise ratio (SNR) has been investigated by Martel and Poisson (1999 Phys. Rev. D 60 124008) in the context of initial LIGO detector. They ascertained that non-eccentric templates will be successful at detecting eccentric signals. We revisit their work by incorporating current and future ground-based detectors and precisely quantify the exact loss of SNR. In order to be more faithful to an actual search, we maximized the SNR over a template bank, whose minimal match is set to 95%. For initial LIGO detector, we claim that the initial eccentricity does not need to be taken into account in our searches for any system with total mass M ∊ [2–45] M ⊙ if e 0 ≲ 0.05 because the loss of SNR (about 5%) is consistent with the discreteness of the template bank. Similarly, this statement is also true for systems with M ∊ [6–35] M ⊙ and e 0 ≲ 0.10. However, by neglecting the eccentricity in our searches, significant loss of detection (larger than 10%) may arise as soon as e 0 ⩾ 0.05 for neutron-star binaries. We also provide exhaustive results for VIRGO, Advanced LIGO and Einstein Telescope detectors. It is worth noting that for Einstein Telescope, neutron star binaries with e 0 ⩾ 0.02 lead to a 10% loss of detection.