Alvaro Santamaría Gómez (2014)

The awardee, Alvaro Santamaría-Gómez, took his first steps in geodesy during his undergraduate degree in land surveying at the University of Salamanca in 2002 and then during his graduate degree in geodesy at the Technical University of Madrid in 2005.

In 2006 he joined the geodetic department at the National Geographic Institute of Spain and in 2007 he started his PhD studies at the Geodesy Research Laboratory of the National Geographic Institute of France. His PhD research focused on the correction of vertical land motion in tide gauge records using GPS velocities. In 2012, while working on the awarded paper, he obtained a Marie Curie International Outgoing Fellowship at the University of La Rochelle and the University of Tasmania. The objective of his present research is to advance in the understanding of vertical land motion errors and its impact on sea level change estimates from tide gauges and satellite altimetry.

The award-winning paper presents an improvement of the method to estimate linear trends of vertical land motion (VLM) at tide gauges using mean sea level observations. Former methods were based on differences between pairs of tide gauge records or differences between a tide gauge record and its corresponding nearby satellite altimetry record. The improved method in this paper is based on double differences between pairs of tide gauge records and their corresponding nearby pairs of satellite altimetry records. The estimated relative VLM trend between redundant (multiple inter-connected) pairs of tide gauges is then adjusted while taking into account their spatial correlation. The VLM trend at each tide gauge is finally obtained by adjusting the origin or datum of the relative VLM estimates using vertical velocities from GPS stations co-located near some of the inter-connected tide gauges.

One of the main advantages of this method is that the geocentric VLM of many inter-connected tide gauges can be estimated from a lower number of co-located GPS stations, i.e. more than one tide gauge per co-located GPS station. When redundant (multiple) GPS velocities are available their relative vertical velocities can be compared against the double-differenced results, resulting in an independent method to assess the quality of GPS vertical velocities to determine the VLM at the tide gauges. Furthermore, by using differences between pairs of altimetry records, the impact of relative geocentric sealevel trends between pairs of tide gauges is reduced while also reducing the spatially-correlated trend errors arising from altimeter bias drift, satellite orbits or sea-surface pressure. More than a thousand tide gauges were considered in the paper. However, due to the shortness of the satellite altimetry records, only pairs of tide gauge and satellite altimetry having an extremely high spatial correlation were used, which reduced the number of tide gauges used to 86. With the extension of the satellite altimetry data span in the future, the correlation threshold can be loosened resulting in more pairs of inter-connected tide gauges being included in the double differences. The estimated VLM at the tide gauges has a mean formal uncertainty of 0.7 mm/yr including the uncertainty of

adjusting the datum of the relative VLM using sparse GPS velocities. With a larger number of tide gauge pairs and colocated GPS velocities, this uncertainty could be ubstantially reduced.