CoAuthors

Michael N. Tsimplis (National Oceanography Centre, UK, United Kingdom); Don P. Chambers (College of Marine Science, University of South Florida, United States)

Global mean sea level (GMSL) is one of the most useful variables to monitor climate change as it reflects both mass addition from melting ice sheets and heat content increase in the oceans. Satellite altimetry measurements and their nearly global spatial coverage have provided a unique tool to study variations in GMSL. Unfortunately, accurate altimetric observations have been available only since 1992, which is not long enough to assess variability on decadal or longer timescales. Prior to 1992, tide gauges were the only means to measure sea level with sufficient accuracy. Tide gauges provide long records of sea level, but they are mainly located at the coast and thus measure coastal sea level, which can differ significantly from GMSL. With the aim of overcoming those limitations and producing a long (>100 years) time series of the GMSL, a lot of efforts have recently been devoted to reconstruct the global sea level fields by combining the tide gauge and satellite altimetry data. The most widely used technique for this purpose is a reduced space optimal interpolation (RSOI). There are a number of factors that affect the skill of this technique and thus introduce errors in the reconstruction, but the discrepancies between tide gauges and nearby altimetry due to the limitations of altimetry in the coastal zone is one of the most important ones. Our goal here is to assess the impact of such discrepancies on the skill of the RSOI to reconstruct the GMSL from tide gauges and satellite altimetry. To address this goal, we explore the analytical solution of the reconstruction technique and then perform a series of numerical experiments using both observations and modeled data. We will discuss the implications of the limited accuracy of coastal altimetry on the sea level reconstructions and assess the benefits brought by improved coastal altimetry data. We will show that better coastal altimetry observations will lead to a better reconstruction of both the rate of rise and the variability in GMSL.