We propose a unitarity method for determining τ neutrino appearance probability P (νμ→ντ) in long-baseline (LBL) accelerator experiments and atmospheric neutrino observations. When simultaneous in situ measurements of P (νμ→νμ) and P (νμ→νe) proceed, as is typical in the LBL experiments, one can use unitarity to "measure" P (νμ→ντ) . A theorists' toy analysis for the model-independent determination of P (νμ→νμ) and P (νμ→νe) is presented by using the NOvA data. It is shown in our analysis that ≲5 % (8%) measurement of τ neutrino appearance probability in neutrino (antineutrino) mode is possible in the peak region 1.5 ≲Eν≲2.5 GeV . The ν SM -independent nature of determination of the probabilities is emphasized.