Summary: Marine geophysical exchange files for R/V Kilo Moana: 2002 to 2018 includes 328 geophysical archive files spanning km0201, the vessel's very first expedition, through km1812, the last survey included in this data synthesis. Data formats (you will likely require only one of these): MGD77T (M77T): ASCII - the current standard format for marine geophysical data exchange, tab delimited, low human readability MGD77: ASCII - legacy format for marine geophysical data exchange (no longer recommended due to truncated data precision and low human readability) GMT DAT: ASCII - the Generic Mapping Tools format in which these archive files were built, best human readability but largest file size MGD77+: highly flexible and disk space saving binary NetCDF-based format, enables adding additional columns and application of errata-based data correction methods (i.e., Chandler et al, 2012), not human readable The process by which formats were converted is explained below. Data Reduction and Explanation: R/V Kilo Moana routinely acquired bathymetry data using two concurrently operated sonar systems hence, for this analysis, a best effort was made to extract center beam depth values from the appropriate sonar system. No resampling or decimation of center beam depth data has been performed with the exception that all depth measurements were required to be temporally separated by at least 1 second. The initial sonar systems were the Kongsberg EM120 for deep and EM1002 for shallow water mapping. The vessel's deep sonar system was upgraded to Kongsberg EM122 in January of 2010 and the shallow system to EM710 in March 2012. The vessel deployed a Lacoste and Romberg spring-type gravity meter (S-33) from 2002 until March 2012 when it was replaced with a Bell Labs BGM-3 forced feedback-type gravity meter. Of considerable importance is that gravity tie-in logs were by and large inadequate for the rigorous removal of gravity drift and tares. Hence a best effort has been made to remove gravity meter drift via robust regression to satellite-derived gravity data. Regression slope and intercept are analogous to instrument drift and DC shift hence their removal markedly improves the agreement between shipboard and satellite gravity anomalies for most surveys. These drift corrections were applied to both observed gravity and free air anomaly fields. If the corrections are undesired by users, the correction coefficients have been supplied within the metadata headers for all gravity surveys, thereby allowing users to undo these drift corrections. The L&R gravity meter had a 180 second hardware filter so for this analysis the data were Gaussian filtered another 180 seconds and resampled at 10 seconds. BGM-3 data are not hardware filtered hence a 360 second Gaussian filter was applied for this analysis. BGM-3 gravity anomalies were resampled at 15 second intervals. For both meter types, data gaps exceeding the filter length were not through-interpolated. Eotvos corrections were computed via the standard formula (e.g., Dehlinger, 1978) and were subjected to identical filtering of the respective gravity meter. The vessel also deployed a Geometrics G-882 cesium vapor magnetometer on several expeditions. A Gaussian filter length of 135 seconds has been applied and resampling was performed at 15 second intervals with the same exception that no interpolation was performed through data gaps exceeding the filter length. Archive file production: At all depth, gravity and magnetic measurement times, vessel GPS navigation was resampled using linear interpolation as most geophysical measurement times did not exactly coincide with GPS position times. The geophysical fields were then merged with resampled vessel navigation and listed sequentially in the GMT DAT format to produce data records. Archive file header fields were populated with relevant information such as port names, PI names, instrument and data processing details, and others whereas survey geographic and temporal boundary fields were automatically computed from the data records. Archive file conversion: Once completed, each marine geophysical data exchange file was converted to the other formats using the Generic Mapping Tools program known as mgd77convert. For example, conversions to the other formats were carried out as follows: mgd77convert km0201.dat -Ft -Tm # gives mgd77t (m77t file extension) mgd77convert km0201.dat -Ft -Ta # gives mgd77 mgd77convert km0201.dat -Ft -Tc # gives mgd77+ (nc file extension) Disclaimers: These data have not been edited in detail using a visual data editor and data outliers are known to exist. Several hardware malfunctions are known to have occurred during the 2002 to 2018 time frame and these malfunctions are apparent in some of the data sets. No guarantee is made that the data are accurate and they are not meant to be used for vessel navigation. Close scrutiny and further removal of outliers and other artifacts is recommended before making scientific determinations from these data. The archive file production method employed for this analysis is explained in detail by Hamilton et al (2019).

{"references": ["Chandler, M. T., and P. Wessel (2012), Errata-based correction of marine geophysical trackline data, Geochem. Geophys. Geosyst., 13, Q10004, doi:10.1029/2012GC004294.", "Dehlinger, P. (1978), Marine Gravity, Elsevier Oceanography Series, 22, Elsevier Sci. Publ. Co.", "Hamilton, M., P. Wessel, B. Taylor, and J. Luis (2019), Producing marine geophysical archive files from raw underway data, Comp. and Geosci., 133, 104321, doi: 10.1016/j.cageo.2019.104321."]}

Funding was provided by the University of Hawai'i at Manoa, School of Ocean and Earth Science and Technology, by U.S. National Science Foundation grants 1458964 and 1558403, and by Michael Hamilton. Paul Wessel, Brian Taylor and Joaquim Luis contributed to the development of the methodology. Scott Ferguson, Joyce Miller and Sandy Shor assisted with data/metadata access.