To ensure the robustness of both civilian and military infrastructure, it is important to protect electric power grids, smart grids, and other electrotechnologies from known and possibly as‐of‐yet unknown space weather hazards. Space weather can generate intense geoelectric fields at the surface of the Earth, as well as large voltage gradients across long distances of the Earth. These voltage gradients can lead to geomagnetically induced currents (GICs), which are known to produce hazards to electric power grids. The finite‐difference time‐domain (FDTD) method is a powerful and versatile method that has already been applied to the study of geoelectric fields. The advantages of FDTD over other methods are that it can account for more geometrical complexities and realistic time waveforms and that it directly solves for geoelectric fields. Snell's Law predicts that any electromagnetic waves incident on the ground should essentially propagate straight downwards into the low resistivity ground. For this reason, vertical FDTD grid resolutions of 1/3 of a skin depth were usually chosen, while the horizontal grid resolution was relaxed. We find, however, that there is another important consideration for choosing an FDTD grid resolution applied to real‐world scenarios: localized field variations due to currents generated by ground features. It turns out the grid resolution requirements are much stricter when taking this physics into account.