Influence of meteorological data and horizontal gradient of tropospheric model on precise point positioning

Using GPT2 derived meteorological data and actual meteorological observations can achieve the same positioning precision in the most areas worldwide except for the Antarctic region. However, the improvement of the actual meteorological observations on the positioning result is significant comparing to using GPT2 derived meteorological data in Antarctic. In the case of 5° elevation cut-off angle, the height precision can be improved by 25%. Furthermore, when the elevation cut-off angle is lower, the effect of the actual meteorological observations on the positioning precision is more significant in Antarctic due to the retention of low elevation angle observations. This study also shows that the influence of tropospheric horizontal gradient correction can improve the PPP precision. Under the lower elevation cut-off angle and higher humidity conditions, especially in summer time and low-latitudes area, the usefulness of the horizontal gradient correction is remarkable. The average improvement of N, E and U directions can reach up to 51%, 15% and 30%, respectively.