. Further by downloading this data the user acknowledges that they agree with the NERC data policy (), and the following conditions:

1. To cite both the data and literature reference in any publication as follows:

DATA REFERENCE

Becker, D., Jordan, T., Corr, H. & Robinson, C. "Strapdown aerogravity survey across the Brunt Ice Shelf, 2017" (2018) Polar Data Centre, Natural Environment Research Council, UK Research & Innovation, UK, doi:10.5285/79e63097-f5dc-41ff-8ca5-36bc4f95a6ff

LITERATURE REFERENCE

Jordan and Becker 2018 In prep.

2. The user recognizes the limitations of data. Use of the data is at the users' own risk, and there is no warranty as to the quality or accuracy of any data, or the fitness of the data for your intended use. The data are not necessarily fully quality assured and cannot be expected to be free from measurement uncertainty, systematic biases, or errors of interpretation or analysis, and may include inaccuracies in error margins quoted with the data.

Methodology:

Gravity data during the Brunt 2017 survey were collected using a strapdown approach. This technique was first developed in the 1990's but advances in sensor design and data processing have only recently made this a viable method for field data collection. The strapdown gravity data was processed using an 18-state Kalman filter in conjunction with a Rauch-Tung-Striebel (RTS) smoother. Besides the 15 states of a typical IMU/GNSS integrating Kalman filter used for navigation (3-D position, velocity, attitude, and six inertial sensor biases), additional states are used to model the gravity disturbance with respect to GRS80 normal gravity. GPS coordinates were processed with a standard precise point positioning software package (Novatel Waypoint GrafNav 8.60), giving an average estimated positional error of 7.1 cm in the vertical and 3.1 cm in the horizontal. The GPS positional estimates were introduced as observations to the Kalman filter for the strapdown gravity calculation. The Kalman filter (and smoother) provides gravity estimates in a one-step evaluation, i.e. no additional low-pass filtering step is required. This method is sometimes referred to as the indirect method of strapdown gravimetry, because gravity is determined indirectly by introducing GNSS positions to the Kalman filter, rather than computing GNSS accelerations in a pre-processing step and manually combining them with the specific forces measured by the accelerometers in order to determine gravity.

In order to minimise thermal effects on the QA2000 accelerometers, the IMU was warmed up for at least two hours before each flight. However, a further thermal correction was applied to compensate reproducible thermal effects arising from internal sensor temperature changes along the flights. Details on the strapdown gravity data processing and the thermal calibration methods can be found in (Becker, 2016).

Data collection:

Data was collected using the BAS aerogeophysicaly equipped twin otter VP-FBL. Gravity data were recorded using an iMAR RQH-1003 inertial measurement unit (IMU), consisting of three Honeywell QA2000 accelerometers (mounted in mutually perpendicular directions), and three Honeywell GG1230 ring laser gyroscopes. Coincident GPS data were recorded with a NovAtel receiver.

Data quality:

The standard deviation of the strapdown gravity crossover errors for the 2017 season was 2.5 mGal, consistent with a Route Mean Square Error (RMSE) of 1.8 mGal. The estimated GPS positional errors were 7.1 cm in the vertical and 3.1 cm in the horizontal.

Resolution: The line spacing for the main survey area is 5km, with regional lines spaced ~18 km apart. The optimum along track resolution of the strapdown gravity system is approximately 100 seconds, consistent with a full wavelength spatial resolution for the 2017 Brunt survey of ~6 km (at an aircraft speed of approximately 60 m/s).