Lithium-ion batteries currently dominate the light-duty electric vehicle (EV) market due to their high energy density, low self-discharge, and high efficiency. Because of these advantages, Li-ion batteries are now being explored for next-generation EV applications like heavy trucks and aerial vehicles. The new applications present design challenges somewhat different from light-duty, road vehicles. For instance, aerial EVs during take-off and landing or heavy trucks on lengthy inclines/declines require high power (or regenerative) pulses for >1 minute. These long-duration, high power pulses may lead to significant temperature rise that could damage the battery and/or produce unsafe conditions resulting in thermal runaway. Accurate estimations of the temperature rise during these pulses are important for initial studies focused on assessing the validity of Li-ion batteries in these and other applications.This work starts by demonstrating how the accuracy of approximating the temperature rise using Joule heating – i.e., I2R, where I is the applied current and R is the electrochemical resistance – decays considerably for high power pulses. A 3-D thermal-electrochemical model of an NMC532-Graphite pouch cell is then used to explain the causes of the deviation. It is shown that a combination of transport in the electrolyte and self-heating cause variations in the resistance during the pulse, which invalidate the Joule heating approximation. Insights from these results are used to develop a correlation for the temperature rise during high power pulses. The correlation provides an improved estimation of the temperature using minimal experimental data for the area specific resistance of the cell.