Theoretical studies of ionospheric structure and dynamics require knowledge of the underlying thermal structure of the ionosphere since it affects the chemical reaction rates, recombination rates, and pressure gradients. Measurements of ionospheric temperatures have been made for decades with a variety of ground- and space-based techniques. This thesis is motivated in particular by the recent improvements in the temperature measurements made by the Jicamarca Radio Observatory (JRO) using the incoherent scatter radar (ISR) technique. Modern ionospheric models all have widely different treatments of ionospheric energetics, and none can produce satisfactory quantitative agreement with the JRO measurements even in quiet conditions. This thesis explores the energy balance calculations in the widely used, open source SAMI2 model in detail, and shows that it is the oversimplification of the treatment of nonlocal heating by photoelectrons in particular which is preventing this model from predicting JRO measurements. This thesis presents an extended version of the SAMI2 model, called SAMI2-PE, which includes a newly developed photoelectron transport model. The model uses finite volume methods which guarantee conservation of particles and energy, incorporates the magnetic field geometry and magnetic mirroring effects, and can be extended to any spatial, energy, and pitch-angle resolution. The new model shows promising agreement with the JRO measurements.