Impact of network geometry, observation schemes and telescope structure deformations on local ties: simulations applied to Sardinia Radio Telescope

The 64-m Sardinia Radio Telescope (SRT) is currently under construction in Sardinia (Italy). To ensure future surveying and monitoring operations at an utmost level of accuracy, we aim at selecting the optimal design and the most cost-effective solution for the establishment of the local ground control network (LGCN). We simulate and test 45 data sets corresponding to 5 different network configurations. We investigate the influence of 2 LGCN geometries (14 or 8 ground markers) and 3 terrestrial observation schemes (based on redundant forward intersections or side shots) on the precision and accuracy of the conventional reference point (CRP) of SRT and the simulated tie vector with a global navigation satellite system (GNSS) station. In addition, thermal and gravitational deformations of the radio telescope structure are simulated as systematic errors introduced into the observations and their effects on the CRP estimates are quantified. The state-of-the-art of CRP surveying and computation, based on terrestrial indirect methods, is applied. We show how terrestrial indirect methods can estimate the position of the radio telescope CRP to the millimeter precision level. With our simulations, we prove that limiting the LGCN to a 8-point configuration ensures the same precision on the CRP obtained with a 14-point network. Furthermore, we demonstrate that in the absence of telescope deformations, side shots, despite the lower redundancy, preserve a precision similar to that of redundant forward intersections. We show that the deformations due to gravitational flexure and thermal expansion of the radio telescope cannot be neglected in the tie vector computation, since they may bias the CRP estimate by several millimeters degrading its accuracy but not impacting on its formal precision. We highlight the dependency of the correlation matrices of the solutions on the geometry of the network and the observation schemes. Similarly, varying the extent of telescope deformations, we show that the CRP estimate again depends on the combination of the network geometry and the observation schemes.