Partitioning Regional Sea Level in the Bay of Bengal from a Global Grace and Jason-1/-2 Joint Inversion

Abstract:

In Bangladesh, large areas are located just above sea level. Present-day sea level rise in combination with land subsidence, poses a major threat to the coastal regions, home of about 30 million people. Consequently, monitoring of sea level and knowledge of all recurrent effects are crucial for coastal protection. As part of the Belmont-project “Bangladesh Delta: Assessment of the Causes of Sea-level Rise Hazards and Integrated Development of Predictive Modeling Towards Mitigation and Adaptation” (BAND-AID) a global inverse method is employed to estimate the different contributors to sea level, such as melting of glaciers and ice-sheets, hydrology, glacial isostatic adjustment, as well as shallow and deep steric effects from Jason-1/2 altimetry and GRACE data.

In the global inverse method, spatial patterns (fingerprints) are computed a-priori for each of the contributing process, applying the sea level equation for mass fingerprints, and empirically (PCA) for steric fingerprints from ARGO data. Temporal GRACE gravity data and along-track Jason-1/ -2 altimetry is then combined to estimate the temporal evolution of these patterns, which allows the partitioning of altimetric sea level into individual sources. This method largely mitigates truncation and leakage problems associated with GRACE resolution. Globally, our estimates are close to others, although they point at a somewhat larger deep steric effect.

In this work we provide preliminary results for the Bay of Bengal / Bangladesh region by confronting global inversion with local measurements. Estimated sea level trends are compared to trends from tide gauges and differences are interpreted in terms of unmodeled regional effects, such as land subsidence. Initial results provide an indication on the magnitude of the contributions from the different sources at the coast of Bangladesh / in the Bay of Bengal; e.g. the contribution from the Greenland ice-sheets between 2003 and 2011 (0.69 mm/a) is significantly larger compared to that of Antarctica (0.15 mm/a), but the biggest effect results from steric sea level changes (-1.5 to 6 mm/a).