The code-carrier phase divergence property of the ionosphere at GPS fr
equencies is used to derive the group delay or, alternatively, the car
rier phase advance, using single frequency observations. The model use
d to separate the carrier phase ambiguity from the initial ionospheric
delay is reviewed. The results of tests conducted with P code L1 and
narrow correlator spacing C/A code L1 receivers are presented and anal
ysed. The latter receiver type delivers pseudorange observations at th
e same level of accuracy as that of P code L1 observations. These resu
lts show that 90% of the relative effect of the ionosphere is removed
using the divergence method. The initial delay can be estimated with a
n accuracy of a few metres. The method is also tested to estimate the
effect of the ionosphere on baseline determination. Double difference
carrier phase observations are used and only the relative effect of th
e ionosphere is required in this case. The results show that 90% of th
e ionospheric effect is removed. The conditions which have to be met t
o apply the divergence successfully are summarized in the conclusions.