POISSON project I. Emission lines as accretion tracers in young stellar objects: results from observations of Chamaeleon I and II sources

Context. We present the results of the analysis of low-resolution optical-near IR spectroscopy (0.6-2.4 mu m) of a sample (47 sources) of Class I and Class II young stellar objects in the Chamaeleon I and II star-forming clouds. These data are part of the POISSON project (Protostellar Optical-Infrared Spectral Survey On NTT). Aims. The aim of the observations is to determine the accretion luminosity (L-acc) and mass accretion rate ((M) overdot(acc)) of the sources through the analysis of the detected emission features. Taking advantage of the wide wavelength range covered by our spectra, we also aim at verifying the reliability and consistency of the existing empirical relationships connecting emission line luminosity and L-acc. Methods. We employ five different tracers ([OI] lambda 6300, H alpha, CaII lambda 8542, Pa beta, and Br gamma) to derive the accretion luminosity, and critically discuss the various determinations in the light of the source properties. Results. The tracers provide L-acc values characterised by different scatters when plotted as a function of L-*. The Br gamma relation appears to be the most reliable, because it gives the minimum dispersion of L-acc over the entire range of L-*, whereas the other tracers, in particular H alpha, provide much more scattered L-acc results, which are not expected for the homogeneous sample of targets we are observing. The direct comparison between L-acc(Br gamma) and the accretion luminosity obtained from the other four tracers also shows systematic differences in the results provided by the empirical relationships. These may probably be ascribed to different excitation mechanisms that contribute to the line emission, which may vary between our sample and those where the relationships have been calibrated, which were mostly based on observations in Taurus. Adopting the accretion luminosities estimates derived from the Br gamma line, we infer L-acc in the range 0.1 L-*-1 L-* for all sources, and (M) overdot(acc) of the order 10(-7) - 10(-9) M-circle dot yr(-1), in the range of values commonly obtained for Class II objects. The mass accretion rates derived in Cha I are roughly proportional to M-*(2), in agreement with the results found in other low-mass star-forming regions. We find that the discrepancies observed in the case of L-acc(Br gamma) and L-acc(Pa beta) can be related to different intrinsic Pa beta/Br gamma ratios. The derived ratios point to the existence of two different emission modalities, one that agrees with predictions of both wind and accretion models, the other suggesting optically thick emission from relatively small regions (10(21)-10(22) cm(2)) with gas at low temperatures (<4000 K), the origin of which needs additional investigation.