Airborne gravity data adjustment using a cross-over adjustment with constraints

• Main Authors: C. Srimanee, P. Dumrongchai, N. Duangdee
• Format: Journal
• Published: 2020
• Subjects: Earth and Planetary Sciences
• Online Access: https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85084060729&origin=inward; http://cmuir.cmu.ac.th/jspui/handle/6653943832/70501
• Institution: Chiang Mai University

© Geoinformatics International. The Royal Thai Survey Department (RTSD), in cooperation with Chiang Mai University, directed airborne gravimetry across Thailand from 2016 to 2017. This made it possible to conduct surveys in inaccessible locations such as mountainous and coastal areas. The gravimeter that was integrated with GPS/GNSS receiver on board together with one gravity base station on the ground measured airborne free-air anomalies (FAA) at an average of 4,000m flight height above MSL with a speed of 200 knots and a 20 Hz measuring rate. The main flight lines were in a north-south direction, while east-west direction lines were supplementary and used to determine the observation quality of the measured FAA. Theoretically, the intersections between the main and supplementary flight lines should have been equal. However, gravity data is greatly affected by air turbulence from weather variations, causing large differences in values at those points. The study area included three operating blocks of geoid development projects conducted in the Chao Phraya River basin. These projects covered the northern and central regions of the country. This study focuses on improving airborne gravity through cross-over adjustments using a least-square adjustment with constraints. The observations were first inspected and edited, and then the FAA differences (FAAD) of 477 intersections were adjusted. The results from the adjustment of three blocks show that the root mean square (RMS) decreased from 3.9 to 2.6 milligals (mGals). Geoid models from both adjusted and non-adjusted airborne gravity data were determined. The geoidal heights of both models were compared with the geoidal height of 184 GNSS/Leveling co-points. The comparative results showed an improvement in accuracy of the adjusted airborne geoid model at a centimetre level.