Impact of baseline length on uncertainty in static relative GNSS positioning

Static relative Global Navigation Satellite Systems (GNSS) positioning is commonly used to achieve mm-level positioning uncertainty. However, few studies have examined the impact of baseline length to the positioning uncertainty. This study assesses 1,500 baselines across the five consecutive observation days and two plate motion inclusion options (15,000 baseline processing), to obtain the relationship between the baseline length and the positioning uncertainty. These baselines connect 25 International GNSS Stations (IGS) stations with the 60 Indonesia Continuously Operating Reference Stations (InaCORS) stations, continuous GNSS network in Indonesia. The 24-h GNSS data is processed using Bernese 5.2 to obtain daily coordinates solution in International Terrestrial Reference Frame 2014 (ITRF2014) with the uncertainty in the east/north/up component. Results show that longer baselines produce higher uncertainties, following an exponential trend. The east component has the lowest uncertainty due to the predominant east baseline azimuths. Baseline azimuths and plate motion considerations slightly influence uncertainty, but baseline length plays a much more significant role. Excluding plate motion noticeably increase uncertainty for baselines over 5,000 km. Specific uncertainty levels can be inferred from the exponential function, such as the baselines under 3,800 km are required for horizontal uncertainties under 3 mm. This exponential function also serves as a basis for selecting the constrained station in the GNSS data processing with the static relative GNSS positioning method.