Simulated observations of total electron content (TEC) along ray paths from Global Positioning System (GPS) satellites have been used to validate the estimation of TEC using GPS measurements. The Sheffield University plasmasphere ionosphere model (SUPIM) has been used to create electron densities that were integrated along ray paths from actual configurations of the GPS constellation. The resultant slant electron contents were then used as inputs to validate the self-calibration of pseudo-range errors (SCORE) process for the determination of TEC from GPS observations. It is shown that if the plasma resides only in the ionosphere below 1100 km, then the SCORE procedure determines the TEC to a high degree of accuracy. When the contribution of the electrons in the protonosphere above 1100 km is included, the analysis results in TEC estimates that are high by some 2 TEC units (TECU) for conditions appropriate to European midlatitudes at solar minimum. However, if a restriction is placed in the analysis on use of observations equatorward of the station, then allowance can be made for the effect of the protonosphere. It is shown that with appropriate selection of the boundary for the observations, TEC can be estimated by SCORE to better than 1 TECU for the conditions of the simulation. Sample results are included from actual experimental observations using GPS to demonstrate the effect of compensation for the protonospheric plasma. ¿ 1999 American Geophysical Union