FitteR for Accretion ProPErties of T Tauri stars (FRAPPE): A new approach to use class III spectra to derive stellar and accretion properties

Context. Studies of the stellar and accretion properties of classical T Tauri stars (CTTS) require photospheric spectral templates to be compared with. The use of low-activity, slowly rotating field dwarfs or model spectra can be advantageous for the determination of stellar parameters, but it can lead to an overestimate of the mass accretion rate, since both classes of templates do not include the emission of the active chromosphere present in young stars. Observed spectra of non-accreting young stars are best suited to this purpose. Using such templates comes with the downside of a limited number of available templates and observational uncertainties on the properties of the templates. Aims. For this work, we aimed to expand the currently available grid of wide-wavelength coverage observed spectra of non-accreting stars with additional new spectra and an interpolation method that allowed us to obtain a continuous grid of low resolution spectra ranging from spectral type G8 to M9.5, while also mitigating observational uncertainties. This interpolated grid was then implemented in the self-consistent method to derive stellar and accretion properties of CTTS. With the new templates, we aimed to estimate a lower limit on the accretion luminosities that can be obtained through a study of the UV excess emission using observed templates. Methods. We analyzed the molecular photospheric features present in the VLT/X-shooter spectra of the targets to perform a spectral classification, including estimates of their extinction. We applied a non-parametric fitting method to the full grid of observed templates to obtain an interpolated grid of templates. Both the individual templates and interpolated grid are provided to the community. We implemented this grid to improve the method to self-consistently derive stellar and accretion properties of accreting stars. We used the uncertainties on our interpolated grid to estimate a lower limit on the accretion luminosity that we can measure with this method. Results. Our new method, which uses a continuous grid of templates, provides results that are consistent with using individual templates but it significantly improves the reliability of the results in the case of degeneracies associated with the peculiarities of individual observed templates. We find that the measurable accretion luminosities range from ∼2.7 dex lower than the stellar luminosity in M5.5 stars to ∼1.3 dex lower for G8 stars. For young stars with masses of ∼1 M⊙ and ages of 3–6 Myr this limit translates into an observational limit of the mass accretion rate on the order of 10−10 M⊙/yr. This limit is higher than the lower limit on the measurable mass accretion rate when using the various emission lines present in the spectra of young stars to estimate the accretion rate. An analysis of these emission lines allows us to probe lower accretion rates, pending a revised calibration of the relationships between line and accretion luminosities at low accretion rates. Conclusions. The implementation of an interpolated grid of observed templates allows us to better disentangle degenerate solutions, leading to a more reliable estimate of accretion rates in young accreting stars.