To advance our understanding, and the application of surface nuclear magnetic resonance (NMR), we need to conduct Earth’s magnetic field (EF) laboratory NMR measurements of the acquired relaxation times. In EF laboratory NMR, prepolarization is used to achieve detectable signal levels. This involves exposing the sample to a static magnetic field to increase the equilibrium magnetization associated with the nuclear spins of protons in the pore fluid. We found that prepolarization can also have a significant impact on the commonly measured relaxation time . We studied this impact on using a set of sand samples taken from boreholes in the U.S. High Plains aquifer. Using rock-magnetic measurements, we found that prepolarization at 25 mT increased the magnetization of the solid phase of the samples up to a factor of 10. For these samples, we observed that decreased from 30 to 3 ms with increasing magnetization. Surface NMR data collected at the borehole site indicate significantly longer (50–80 ms) at the depths from which the samples were taken. We attributed our observations to prepolarization inducing remanent magnetization in the solid phase of the sample, which results in elevated internal magnetic fields that reduce . In contrast, we found good agreement between EF laboratory and surface NMR measurements of the relaxation time . Because is unaffected by internal fields, this supports our conclusion that prepolarization is significantly impacting the laboratory measurement of . To mitigate these undesired effects, we have developed the inclusion of a demagnetization pulse after prepolarization in EF laboratory experiments. By applying the demagnetization pulse along the same direction of the static field only the remanent magnetization of the solid phase of the sample would be removed, whereas the prepolarized state of the nuclear spin magnetization of the liquid phase would be retained.