A surface energy balance model is used to calculate the surface heat budget terms from data gathered in austral summer 1997--1998 at seven locations near Svea (74¿35'S, 11¿13'W), located in the Heimefrontfjella, Dronning Maud Land, Antarctica. The modeled results are validated using the turbulent fluxes calculated with the profile method, those directly measured with sonic anemometer, the measured surface and subsurface temperatures, and the stake ablation readings to constrain the simulated budgets. The simulated quantities generally agree reasonably well with those observed. The seven measuring sites can be subdivided into three climatic regions: I, blue ice areas; II, the low-altitude coastal snow-covered plateau; and III, the high-altitude interior plateau. Spatial and temporal differences in the average surface heat budget and in the variability on daily and longer-term timescales are presented and discussed. The three regions exhibit quite different surface heat budget regimes induced by their climatic setting and the prevailing boundary layer flow, a well-established katabatic flow regime, and the associated typical surface heat fluxes being found over regions II and III. The differences between blue ice and snow surface are quite distinct, blue ice absorbing twice as much solar energy due to its lower albedo. This and the lower relative humidity values cause the latent heat flux to be much higher over blue ice than over snow, enabling strong sublimation and a negative mass balance. Differences in the turbulent heat fluxes between the three climatic regions can be well understood in terms of the temperature-dependent Bowen ratio. The surface heat budgets of the various sites are compared with estimates of other regions in the Antarctic. ¿ 2000 American Geophysical Union