GNSS Monitoring Network Optimization Case Study: Opak Fault Deformation, Yogyakarta

Abstract

Opak fault is a fault located in Opak River area, Bantul. The existence of the fault is one of the biggest causes of earthquake in Yogyakarta in 2006. The seismic potential caused by the active fault requires continuous geodynamic monitoring. The GNSS network (TGD, SGY, and OPK) have been developed since 2013 consists of 17 stations and in 2016 there was an additional number of four monitoring stations. Several high-precision monitoring stations distributed at the fault location are needed to monitor the fault movement. Optimal observation network is one of the factors to obtain high precision station coordinates. The GNSS network optimization has been carried out in the previous research partially on each network; namely the segment of TGD, SGY, and OPK. Therefore, this research conducts a thoroughly optimization for 17 monitoring stations either use old or new stations to obtain an optimal network based on the criteria of accuracy and reliability.

The network is designed widely from simple to complex combination and to combination between network segments. The computation uses least squares adjustment with parameter method. The value of the cofactor matrix parameter of the adjustment is applied to analyze the network based on the function of the accuracy criteria, namely A-Optimality, D-Optimality, E-Optimality, S-Optimality, and I-Optimality. Meanwhile, the value of the residual cofactor matrix is used for network configuration analysis based on the reliability objective function, namely the individual redundancy, external and internal reliabilities criteria. The result showed that the design of TGD, SGY and OPK network segments are optimized based on the criteria of accuracy and reliability if they use a network design with a complex baseline. The criteria for accuracy and reliability in the design with a combination of segments such as TGD and SGY, TGD and OPK, as well as TGD, SGY, and OPK are not much different from the optimization results performed by each segment. Therefore, if the measurements are carried out with a limited receiver, it is better to use each of segment designs.