The purpose of this report is to compare different methods using Global Navigation Satellite System Real Time Kinematic (GNSS RTK) to establish control points to be used for the establishing of a free total station (in the next step). The objectives are to evaluate quality measures for different methods for multiple occupations and the averaging method “180-seconds”. The quality measures used in the study is expanded uncertainty (U95; with 95% level of confidence) and maximum deviation from the true value (“risk”), i.e. the maximum horizontal distance from the mean.From the results in this study, it is clear that it is not only the number of occupations that matters, also the length of the observation periods is important in order to minimize the risk. Extending from one occupation to two (or more) in order to be ‘safe’ is to give a false sense of security.Janssen et al. (2012) stated that an observation window of 1-2 minutes reduces the effects of extreme outliers as much as possible in the shortest time frame. They also concluded that averaging for a longer period than 1-2 minutes does not appear to provide any significant further improvement. In our study, however, increasing the observation window from 1-2 to 3 minutes, are motivated by a decrease in risk (cf. Appendix 1). Further, 180 seconds seem like an eternity for RTK users in the field; consequently, they will use supporting legs for their antenna pole. Using a shorter averaging time (60-120 s), this is not always obvious for the user. Consequently, extending the observation window to 3 minutes is motivated by a decrease in risk and a decrease in centering error. Therefore, the recommendation is to use observation periods of at least 180 seconds (3 min) of data. This is according to the recommendations given in Edwards et al. (2010).There is a trade-off between the recommendation of using as many observations as possible, i.e. at least two occupations with at least 3 minutes length of every observation periods, and productivity. This task must be carefully balanced by the surveyor in a case-by-case evaluation.Regarding productivity, averaging over 180 seconds of data at only one occupation seems to be a proper balance for cadastral surveying. According to this study it is not significantly worse than the mean of the eleven different multiple occupations methods in this study.