In this study, we performed a comparative analysis, on different sites in French Polynesia, between 11 years of Tide Gauge (TG) and GNSS data and the 27 years of gridded sea level data from GLORYS (GL). The TG data provide Relative Sea Level variations (ΔRSL), while the GNSS provide Vertical Land Motions (VLM) and GL Absolute Sea Level Variations (ΔASL), at six sites (Papeete, Vairao, Rikitea, Tubuai, Makemo, and Rangiroa). In both data sets (TG and GL records), we found all the frequencies associated with climate indicators like the Interdecadal Pacific Oscillation (IPO), the Pacific Decadal Oscillation (PDO), and the El Niño Southern Oscillation (ENSO). A regional difference is found between a Northern group (TGs from the Society and Tuamotu archipelagos) and a Southern group (TGs from the Gambier and Austral islands). A strong signal of the IPO is detected at the Tubuai TG (Austral Islands), this signal could blur the secular trend. The main conclusions of this paper are: a) the inter-annual IPO variability is the only phenomenon affecting the ΔASLGL determination on the 27-year long GL series; b) due to the inter-annual variability, the 11-year time span of TG records is too short to determine reliable long-term ΔRSLTG trends; c) The subtraction of GL data from TG data, which eliminates the common inter-annual variability, provides a proxy for the determination of VLMTG, as -VLM = ∆RSL − ∆ASL and the VLMTG are reliable when compared to the VLM derived from GNSS data records; d) reliable ∆RSL=∆ASLGL-VLMTG trends are obtained from the combination of TG and GL data. ∆RSL values around 2.2 mm/yr are obtained at Vairao and Rangiroa. For the Southern group, much larger values are observed due mainly to the contribution of IPO. After correction we get 2.45 mm/yr at Rikitea and 3.33 mm/yr at Tubuai.
Ducarme et al. (Sun,) studied this question.