Impact of Cosmic Filaments on Galaxy Morphological Evolution and Predictions of Early Cosmic Web Structure for Roman

We leverage the IllustrisTNG cosmological simulations to test how the large-scale cosmic web shapes galaxy morphology and to forecast the early cosmic web structure that the Nancy Grace Roman Space Telescope will reveal. In the hydrodynamic TNG50 and $N$-body TNG50-Dark runs, we reconstruct the cosmic web at redshifts $z=0$, 0.5, 1, 2, 3, and 4 with the Monte Carlo Physarum Machine density estimator and the DisPerSE structure identification framework. We confirm that dark matter halos start out predominantly prolate (elongated) and their shapes are aligned with their nearest filaments; prolate galaxies retain strong shape-alignment with their outer halos to later times. The fraction of prolate galaxies and halos increases toward lower stellar mass, higher redshift, and lower-density filaments. Oblate and spheroidal galaxies show weaker trends with filament density, but spheroidal halos preferentially reside in higher-density filaments. We also find that higher-density filaments favor extended rotationally-supported disks, whereas lower-density filaments more often host smaller dispersion-supported systems. Then, generating mock galaxy samples from TNG100 and TNG50, we predict the early cosmic web accessible to Roman. We find that the spectroscopic emission-line depth planned for the High-Latitude Wide-Area Survey (HLWAS) yields a highly incomplete galaxy sample that does not accurately trace the $z=1$ cosmic web. A survey $\geq2.5\times$ deeper over a few square degrees would enable a proper reconstruction and reveal qualitatively correct filament-galaxy morphology relationships. Nevertheless, the planned HLWAS Deep field should still identify most galaxy overdensities; targeted deeper spectroscopy of these regions would efficiently and adequately map the early filamentary structure.