consistent inter-species offsets and reconstructions of pH versus depth ( Pearson and Palmer 1999 ) which would not have been possible if we were not roughly in the right ballpark , so to speak . We were also able to reconstruct glacial to interglacial changes in pCO 2 with good quantitative accuracy using the same technique ( Palmer and Pearson 2003 ). Clearly , however , the record needs replacing by newer and better data and some contributions toward that are considered below .
ii . Pagani et al . ( 1999 and 2005 )
Fig . 2 . Record of Pagani et al . ( 2005 ) from alkenones from 0-50 Ma . Note there are no data from the last 5 m . yr . Red line added .
The record of Pagani et al . ( 2005 ) ( the Miocene part of which is from Pagani et al . 1999 ) utilises the alkenone proxy . It can be seen from this record that the last time there were values > 400 ppm was in the Oligocene epoch ( or just maybe in the early Miocene ). There have been some recent revisions and additions to this data set since , for example , attempts to account for cell size effects , to deal with areas of high productivity , and to provide better estimates of temperature . However in general the record is similar to Pearson and Palmer ( 2000 ) in showing the relatively low pCO 2 values from the Miocene onward ( which is , incidentally , something neither research group expected to find when the work started ).