divide the current database into two parts, to calibrate the TF
on the first part, to apply it to the second, called the validation sample, and to deduce the average validation error. If
this error remains reasonable, being inevitably greater than
the average calibration error, then the TF is considered to be
applicable. The same procedure is repeated after exchanging
the calibration and the validation subsets. If the two resulting
TFs obtained are acceptable, the TF can then be calibrated
on the full data set and used for fossil assemblies.
Other statistical techniques exist (bootstrap, jackknife,
permutations, etc.), but the basic paradigm is always to
verify the quality of the estimates on data independent from
those used for the calibration. Another important point is to
systematically provide confidence intervals associated with
climate reconstructions. This can be achieved for both the
TF and the AM. These confidence intervals mean that the
tolerance of the biological assemblies for a more or less wide
range of climatic conditions can be assessed and the
imperfection of the model to adjust all the current reference
points can be taken into account.
References
Guiot, J., & de Vernal, A. (2007). Transfer functions: Methods for
quantitative paleoceanography based on microfossils. In C.
Hillaire-Marcel & A. de Vernal (Eds.), Developments in marine
geology (Vol. 1, pp. 548–588). Dordrecht: Elsevier.
Imbrie, J., & Kipp, N. G. (1971). A new micropaleontological method
for quantitative paleoclimatology: Application to a Late Pleistocene
Caribbean core. In K. K. Turekian (Ed.), The Late Cenozoic glacial
ages (pp. 71–181). USA: Yale University Press.
Kucera, M., Weinelt, M., Kiefer, T., Pflaumann, U., Hayes, A.,
Weinelt, M., et al. (2005). Reconstruction of sea-surface temperatures
from
assemblages
of
planktonic
foraminifera:
Multi-technique approach based on geographically constrained
calibration data sets and its application to glacial atlantic and Pacific
Oceans. Quaternary Science Reviews, 24, 951–998.
Pfister, C. (1980). The climate of Switzerland in the last 450 years.
Geographica Helvetica (numéro spécial), 15–20.
144
V. Masson-Delmotte and J. Guiot
on the first part, to apply it to the second, called the validation sample, and to deduce the average validation error. If
this error remains reasonable, being inevitably greater than
the average calibration error, then the TF is considered to be
applicable. The same procedure is repeated after exchanging
the calibration and the validation subsets. If the two resulting
TFs obtained are acceptable, the TF can then be calibrated
on the full data set and used for fossil assemblies.
Other statistical techniques exist (bootstrap, jackknife,
permutations, etc.), but the basic paradigm is always to
verify the quality of the estimates on data independent from
those used for the calibration. Another important point is to
systematically provide confidence intervals associated with
climate reconstructions. This can be achieved for both the
TF and the AM. These confidence intervals mean that the
tolerance of the biological assemblies for a more or less wide
range of climatic conditions can be assessed and the
imperfection of the model to adjust all the current reference
points can be taken into account.
References
Guiot, J., & de Vernal, A. (2007). Transfer functions: Methods for
quantitative paleoceanography based on microfossils. In C.
Hillaire-Marcel & A. de Vernal (Eds.), Developments in marine
geology (Vol. 1, pp. 548–588). Dordrecht: Elsevier.
Imbrie, J., & Kipp, N. G. (1971). A new micropaleontological method
for quantitative paleoclimatology: Application to a Late Pleistocene
Caribbean core. In K. K. Turekian (Ed.), The Late Cenozoic glacial
ages (pp. 71–181). USA: Yale University Press.
Kucera, M., Weinelt, M., Kiefer, T., Pflaumann, U., Hayes, A.,
Weinelt, M., et al. (2005). Reconstruction of sea-surface temperatures
from
assemblages
of
planktonic
foraminifera:
Multi-technique approach based on geographically constrained
calibration data sets and its application to glacial atlantic and Pacific
Oceans. Quaternary Science Reviews, 24, 951–998.
Pfister, C. (1980). The climate of Switzerland in the last 450 years.
Geographica Helvetica (numéro spécial), 15–20.
144
V. Masson-Delmotte and J. Guiot