GRACE's warts – new peer reviewed paper suggests errors and adjustments may be large

Below is a GRACE satellite map. The Earth looks like a warty ball, with red bumps and blue pits that represent measured fluctuations in the planet’s gravity. Note Greenland in the red. We’ve covered GRACE before, suggesting it may not be a good tool to measure ice loss in Greenland. See this WUWT story.

The red spots represent measurements where Earth’s gravity is stronger. The blue ones are where it is measured to be weaker. The universal force of gravity itself does not vary, but the pits and bumps are a local indication that Earth’s mass distribution isn’t smooth and uniform. As seen on the image above, tectonic mountain building in South America produces red zones; elsewhere, tectonic movements produce thin, blue, ones.

Even more interesting is the fact that the map changes over time, Earth as we know is not static.

CO2 science reviews this new paper, which suggests that for sea level rise and ocean mass, the signal to noise ratio is high low and adjustments further complicate the issue. It also suggests some studies aren’t appropriately correcting for these issues. For example, GRACE measurements related to Greenland and West Antarctica (which we also criticized in WUWT here and here):

“…non-ocean signals, such as in the Indian Ocean due to the 2004 Sumatran-Andean earthquake, and near Greenland and West Antarctica due to land signal leakage, can also corrupt the ocean trend estimates.”

Ocean Mass Trends (and Sea Level Estimates) from GRACE Reference

Quinn, K.J. and Ponte, R.M. 2010. Uncertainty in ocean mass trends from GRACE. Geophysical Journal International 181: 762-768.

Background

The authors write that “ocean mass, together with steric sea level, are the key components of total observed sea level change,” and that “monthly observations from the Gravity Recovery and Climate Experiment (GRACE) can provide estimates of the ocean mass component of the sea level budget, but full use of the data requires a detailed understanding of its errors and biases.”

What was done

In an effort designed to provide some of that “detailed understanding” of GRACE’s “errors and biases,” Quinn and Ponte conducted what they describe as “a detailed analysis of processing and post-processing factors affecting GRACE estimates of ocean mass trends,” by “comparing results from different data centers and exploring a range of post-processing filtering and modeling parameters, including the effects of geocenter motion, PGR [postglacial rebound], and atmospheric pressure.”

What was learned

The two researchers report that the mean ocean mass trends they calculated “vary quite dramatically depending on which GRACE product is used, which adjustments are applied, and how the data are processed.” More specifically, they state that “the PGR adjustment ranges from 1 to 2 mm/year, the geocenter adjustment may have biases on the order of 0.2 mm/year, and the atmospheric mass correction may have errors of up to 0.1 mm/year,” while “differences between GRACE data centers are quite large, up to 1 mm/year, and differences due to variations in the processing may be up to 0.5 mm/year.”

What it means

In light of the fact that Quinn and Ponte indicate that “over the last century, the rate of sea level rise has been only 1.7 ± 0.5 mm/year, based on tide gauge reconstructions (Church and White, 2006),” it seems a bit strange that one would ever question that result on the basis of a GRACE-derived assessment, with its many and potentially very large “errors and biases.” In addition, as Ramillien et al. (2006) have noted, “the GRACE data time series is still very short,” and results obtained from it “must be considered as preliminary since we cannot exclude that apparent trends [derived from it] only reflect inter-annual fluctuations.” And as Quinn and Ponte also add, “non-ocean signals, such as in the Indian Ocean due to the 2004 Sumatran-Andean earthquake, and near Greenland and West Antarctica due to land signal leakage, can also corrupt the ocean trend estimates.”

Clearly, the GRACE approach to evaluating ocean mass and sea level trends still has a long way to go — and must develop a long history of data acquisition — before it can ever be considered a reliable means of providing assessments of ocean mass and sea level change that are accurate enough to detect an anthropogenic signal that could be confidently distinguished from natural variability.

References

Church, J.A. and White, N.J. 2006. A 20th-century acceleration in global sea-level rise. Geophysical Research Letters 33: 10.1029/2005GL024826.

Ramillien, G., Lombard, A., Cazenave, A., Ivins, E.R., Llubes, M., Remy, F. and Biancale, R. 2006. Interannual variations of the mass balance of the Antarctica and Greenland ice sheets from GRACE. Global and Planetary Change 53: 198-208.