Characterization of local boiling trends, in addition to the typically reported area-averaged trends, is essential for the robust design and implementation of phase change technologies to sensitive heat transfer applications such as electronics cooling. This paper compares local and area-averaged boiling curves during a phase change process using non-intrusive quantitative infrared thermal visualization to highlight the importance of considering the former in design of phase change cooling systems. As an illustration, both pool and jet impingement boiling modes of heat transfer from a heated silicon surface are considered. A thin-film serpentine heater that allows for a circumferentially uniform but radially non-uniform heat flux distribution on the surface is chosen as the heat source in order to assess the effect of spatial variations in imposed boundary condition on the local and area-averaged boiling curves. Local convective heat fluxes are estimated from thermal maps using a control volume approach that accounts for axial conduction in the silicon substrate. Using the spatial information on surface temperatures and convective heat fluxes, local boiling curves are generated for different radial locations on the surface and compared with their corresponding area-averaged representations. As validation, it is shown that the net input electrical heat flux varies within 4-12% of the area-averaged results. Boiling curves averaged over regions of like imposed boundary condition are seen to provide a more realistic estimate of the local heat transfer conditions when compared with an area-averaged representation of the entire surface. Area-averaged thin-film sensor data are used to augment the heat transfer data by highlighting the differences in nucleation events on the surface in different regions. (C) 2012 Elsevier Ltd. All rights reserved.