Using the specific satellite line of sight geometry and station location with respect to the source, thomas et al. [Scientific Reports, https://doi.org/10.1038/s41598-018-30476-9] developed a method to infer the detection altitude of co-seismic ionospheric perturbations observed in Global Positioning System (GPS)-Total Electron Content (TEC) measurements during the Mw 7.4 March 9, 2011 Sanriku-Oki earthquake, a foreshock of the Mw 9.0, March 11, 2011 Tohoku-Oki earthquake. Therefore, in addition to the spatio-temporal evolution, the altitude information of the seismically induced ionospheric signatures can also be derived now using GPS-TEC technique. However, this method considered a point source, in terms of a small rupture area (~90 km) during the Tohoku foreshock, for the generation of seismo-acoustic waves in 3D space and time. In this article, we explore further efficacy of GPS-TEC technique during co-seismic ionospheric sounding for an extended seismic source varying simultaneously in space and time akin to the rupture of Mw 9.0 Tohoku-Oki mainshock and the limitations to be aware of in such context. With the successful execution of the method by Thomas et al. during the Tohoku-Oki mainshock, we not only estimate the detection altitude of GPS-TEC derived co-seismic ionospheric signatures but also delineate, for the first time, distinct ground seismic sources responsible for the generation of these perturbations, which evolved during the initial 60 seconds of the rupture. Simulated tsunami water excitation over the fault region, to envisage the evolution of crustal deformation in space and time along the rupture, formed the base for our model analysis. Further, the simulated water displacement assists our proposed novel approach to delineate the ground seismic sources entirely based on the ensuing ionospheric perturbations which were otherwise not well reproduced by the ground rupture process within this stipulated time. Despite providing the novel information on the segmentation of the Tohoku-Oki seismic source based on the co-seismic ionospheric response to the initial 60 seconds of the event, our model could not reproduce precise rupture kinematics over this period. This shortcoming is also credited to the specific GPS satellite-station viewing geometries.