There is an important summary article on sea level rise that has appeared in the AGU publication EOS. This article is
N. A. Schwadron et al., 2011: Two Millennia of Sea Level Data: The Key to Predicting Change. Eos, Vol. 92, No. 35, 30 August 2011
The abstract reads
A close look at the past 2000 years of sea level history may help reveal useful lessons for improving future sea level predictions.
Excerpts from the article are [highlights added] Note: GIA is used in the text for glacial isostatic adjustment
Sea level reconstructions spanning the late Holocene (the past 2000 years) provide a preindustrial context for understanding the patterns and causes of contemporary and future change. The Intergovernmental Panel on Climate Change (IPCC) assumed that global sea level change during the past two millennia (prior to the middle of the nineteenth century) was close to zero [Bindoff et al., 2007], but understanding of late Holocene sea level variability is limited. Glaciers and ice sheets changed significantly in size during this period, and therefore sea level likely oscillated on the order of several decimeters. In addition, ocean dynamics, solid Earth movements, steric (density) changes, and gravitational effects contributed to complex regional patterns of sea level change.
On any given coastline, relative sea level changes occur as a result of local, regional, and global processes. Examples of local processes are tidal range changes and sediment compaction. Regional processes are dominated by vertical land motion (isostatic and tectonic). Global processes include ocean basin volume change and exchange of mass between oceans and continental ice sheets and glaciers. A slow increase in ocean basin volume due to glacial isostatic adjustment (GIA), a process triggered by the rebound of the Earth following the melting of ice sheets that covered much of North America and Europe during the last glaciation, causes sea level to fall around most of the globe at a rate of about 0.3 millimeter per year [Gehrels, 2010]. The contribution of global ice melt during the past 2000 years is less well constrained but is assumed to be close to zero by GIA models [Milne et al., 2009] and 0.0–0.2 millimeter per year by the IPCC’s fourth assessment report [Bindoff et al.,2007]. Accurately estimating this rate has far-reaching implications because it provides a benchmark against which rates of sea level rise measured by tide gauges and satellites can be compared.
Contemporary observations of changes in the sizes of glaciers and ice sheets made by the Gravity Recovery and Climate Experiment (GRACE) demonstrate the relevance of a late Holocene context and the importance of robust sea level reconstructions [Bentley, 2010]. GRACE measures the distribution of mass on Earth’s surface. However, it is unable to distinguish directly between gravitational changes produced by changing ice volume and those caused by GIA. Corrections to compensate for GIA are therefore critical, but they can be as large as the ice ocean mass flux signal itself [Cazenave et al., 2009], making it difficult to isolate the meltwater contribution to sea level rise.
Several lessons can be drawn from global sea level patterns during the past two millennia. First, sea level change attributed to melting ice from 0 to 1000 C.E. was close to zero, but proxy sea level data from various regions appear to record a slowly falling sea level during much of the past millennium. Second, ongoing sea level rise represents a significant departure from late Holocene trends and began during the late nineteenth or early twentieth century.The preceding global sea level fall implies that the change to modern rates of sea level rise was greater in magnitude than assumed by the IPCC, which considered a rate of 0.0–0.2 millimeter per year for the past 2000 years and a rate of about 1.7 millimeters per year for the twentieth century.
Finally, regional sea level variability must be captured by climate models if there is to be any expectation that these models can accurately predict future sea level change. Herein lies a big challenge for future work. Single values of the magnitude of future global sea level rise are of little practical use for agencies concerned with defending coasts against future flooding. To obtain a better understanding of future sea level change, projections must take into account all known processes that contribute to patterns of relative sea level change at local to regional scales.
My conclusion from this article is that sea level assessments involve a range of uncertainties that have not yet been properly quantified and corrected for. Also, since the authors report a larger rise in sea level starting in the late 18oos or early 1900s, if this were due primarily from ocean heating due to the positive radiative forcing the input of CO2 and other greenhouse gases, we should have seen an acceleration of the sea level rise later in the 20th and early 21st centuries, which has not occurred. The authors, also report that the use of a global average sea level trend value is of little use in reducing the threat to sea level changes that coastal communities face.
source of image from the University of Colorado Sea Level Research Group