Dunai Scaling in ACE

New to ACE? The Introduction is a good place to start.

A reviewer of our Quaternary Geochronology manuscript noted that when using Dunai (2001) scaling in ACE, HLSL production rates are about 10% lower than for other scalings.  This is verified by running two ACE experiments with different scaling and comparing calibrated output:

Click on image to expand

Click on image to expand

For the Demonstration experiment (using Desilets and Zreda 2003 scaling) the HLSL spallation scaling is ~5 a / g / yr and for the experiment using Dunai (2001) scaling the HLSL spallation scaling is ~4.5 a / g / yr.  Details of scalings are listed here, however this result is different to other results (eg Table 6 of Balco et al, 2008). Upon further investigation we found the difference was in the calculation of sea level pressure.  The following comes from DunaiScalingFunctions.py in the Components Directory:

slpressure = s["sea level pressure"] * (10. / g_0)

Meaning, ‘for sea level pressure take the NCEP reanalysis and convert to g cm-2. Although this seems a reasonable estimate of sea level pressure, in 2001 Dunai assumed a globally uniform value of 1030 g cm-2. Our NCEP derived sea level pressure is shown here. By changing the above line to 1030 g cm-2, we get a HLSL spallation production rate of ~5 a / g / yr, in line with other scalings.

So which is correct? Using a sea level pressure of 1030 g cm-2 is probably more in keeping with the spirit of the Dunai 2001 paper, but using reanalysis data give a better calibration, reflected by a lower chi-square (see calibration results in figure). What is the appropriate choice?

Leave a Reply

You must be logged in to post a comment.