Estimating vertical land motion and sea level change using satellite altimetry, tide gauges and GPS

Abstract:

Vertical land motion (VLM) is often poorly understood and sparsely observed yet is the connection between absolute sea level (ASL) observed by satellite altimetry and relative sea level (RSL) observed by tide gauges. The combination of altimetry, tide gauges and Global Positioning System (GPS) data offers the potential to refine estimates of land- and sea-level change, yet these are often limited in a regional context by short records and geographically correlated systematic errors. We develop a flexible methodology based on a Kalman filtering and smoothing framework to simultaneously estimate both location-specific VLM and mission-specific systematic errors. We incorporate altimeter minus tide gauge, altimeter crossover and GPS height observations within a spatio-temporal framework. We apply our method to the Baltic Sea and assess its performance. We show the technique can separate VLM at tide gauges from small but geographically correlated systematic errors in altimetry. We quantify local variation in VLM at TGs that otherwise cannot be derived from geophysical models or interpolated from GPS-based velocity fields. We apply these VLM estimates to TG RSL and find that the variability in derived ASL rates is reduced by ~20%. Together, understanding local variation in VLM at TGs, as well as small but significant regionally correlated errors in altimetry is required to advance this approach and improve our geophysical interpretation of these dynamic phenomena.