Twentieth century sea-level rise inferred from tide gauge, geologically derived and thermosteric sea-level changes

Relative sea-level (RSL) changes at thirteen sites derived from tide gauge and/or salt-marsh sediment sequences, with information for RSL changes during at least the past ~200 years, are used to infer the rates and causes of the global sea-level rise in the twentieth century (pre-satellite era) by incorporating spatially non-uniform thermosteric sea-level change and recent estimates for the melting of mountain glaciers and both polar ice sheets. The main advantage of the method adopted here is that we can avoid a model-dependence in the GIA (glacial isostatic adjustment) correction and any steady tectonic effect depending on each location can be corrected for. We first estimate the acceleration of sea-level rise in the twentieth century at each RSL observation site, and then evaluate the residuals between the sea-level acceleration and thermosteric sea-level rise. The spatially non-uniform residuals are examined to infer the rates of RSL rise due to the melting of mountain glaciers and polar ice sheets. The comparison between the residuals and the predicted rates for the melting wholly supports the estimates for the melting by the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC 2007 Report), and does not require a significant melting from the Greenland ice sheet. However, the possible equivalent sea-level rise for mountain glaciers and/or Antarctic ice sheet may be ~0.3mmyr^-1 larger than the preferred estimate by IPCC 2007 Report (~0.7mmyr^-1), and the total one is ~1.0mmyr^-1. The additional melting for ~0.3mmyr^-1 is due to the comparison mainly for Brest with long tide gauge records of ~200 years. Also, the present study indicates that the global mean thermosteric sea-level rise before the rapid sea-level acceleration occurred at 1990 is ~0.3mmyr^-1.