Matched field processing (MFP) was introduced sometimes ago for source localization based on the replica field for a hypothesized source location that best matches the acoustic data received on a vertical line array (VLA). A data-based matched-mode source localization method is introduced in this paper for a moving source, using mode wavenumbers and depth functions estimated directly from the data, without requiring any environmental acoustic information and assuming any propagation model to calculate the replica field. The method is in theory free of the environmental mismatch problem since the mode replicas are estimated from the same data used to localize the source. Besides the estimation error due to the approximations made in deriving the data-based algorithms, the method has some inherent drawbacks: (1) it uses a smaller number of modes than theoretically possible, since some modes are not resolved in the measurements, and (2) the depth search is limited to the depth covered by the receivers. Using simulated data, it is found that the performance degradation due to the above approximation/limitation is marginal compared with the original matched-mode source localization method. Certain aspects of the proposed method have previously been tested against data. The key issues are discussed in this paper.