Articles | Volume 12, issue 5
https://doi.org/10.5194/cp-12-1215-2016
https://doi.org/10.5194/cp-12-1215-2016
Technical note
 | 
24 May 2016
Technical note |  | 24 May 2016

Technical note: Estimating unbiased transfer-function performances in spatially structured environments

Mathias Trachsel and Richard J. Telford

Related authors

Assessing performance and seasonal bias of pollen-based climate reconstructions in a perfect model world
Kira Rehfeld, Mathias Trachsel, Richard J. Telford, and Thomas Laepple
Clim. Past, 12, 2255–2270, https://doi.org/10.5194/cp-12-2255-2016,https://doi.org/10.5194/cp-12-2255-2016, 2016
Short summary
Modelling annual mass balances of eight Scandinavian glaciers using statistical models
M. Trachsel and A. Nesje
The Cryosphere, 9, 1401–1414, https://doi.org/10.5194/tc-9-1401-2015,https://doi.org/10.5194/tc-9-1401-2015, 2015
Short summary

Related subject area

Subject: Proxy Use-Development-Validation | Archive: Marine Archives | Timescale: Holocene
Evaluation of the distributions of hydroxylated glycerol dibiphytanyl glycerol tetraethers (GDGTs) in Holocene Baltic Sea sediments for reconstruction of sea surface temperature: the effect of changing salinity
Jaap S. Sinninghe Damsté, Lisa A. Warden, Carlo Berg, Klaus Jürgens, and Matthias Moros
Clim. Past, 18, 2271–2288, https://doi.org/10.5194/cp-18-2271-2022,https://doi.org/10.5194/cp-18-2271-2022, 2022
Short summary
Technical Note: Past and future warming – direct comparison on multi-century timescales
Darrell S. Kaufman and Nicholas P. McKay
Clim. Past, 18, 911–917, https://doi.org/10.5194/cp-18-911-2022,https://doi.org/10.5194/cp-18-911-2022, 2022
Short summary
Co-evolution of the terrestrial and aquatic ecosystem in the Holocene Baltic Sea
Gabriella M. Weiss, Julie Lattaud, Marcel T. J. van der Meer, and Timothy I. Eglinton
Clim. Past, 18, 233–248, https://doi.org/10.5194/cp-18-233-2022,https://doi.org/10.5194/cp-18-233-2022, 2022
Short summary
Holocene palaeoceanography of the Northeast Greenland shelf
Teodora Pados-Dibattista, Christof Pearce, Henrieka Detlef, Jørgen Bendtsen, and Marit-Solveig Seidenkrantz
Clim. Past, 18, 103–127, https://doi.org/10.5194/cp-18-103-2022,https://doi.org/10.5194/cp-18-103-2022, 2022
Short summary
A spectral approach to estimating the timescale-dependent uncertainty of paleoclimate records – Part 2: Application and interpretation
Andrew M. Dolman, Torben Kunz, Jeroen Groeneveld, and Thomas Laepple
Clim. Past, 17, 825–841, https://doi.org/10.5194/cp-17-825-2021,https://doi.org/10.5194/cp-17-825-2021, 2021
Short summary

Cited articles

Burman, P., Chow, E., and Nolan, D.: A cross-validatory method for dependent data, Biometrika, 81, 351–358, https://doi.org/10.1093/biomet/81.2.351, 1994.
Frechette, B., de Vernal, A., Guiot, J., Wolfe, A. P., Miller, G. H., Fredskild, B., Kerwin, M. W., and Richard, P. J. H.: Methodological basis for quantitative reconstruction of air temperature and sunshine from pollen assemblages in Arctic Canada and Greenland, Quaternary Sci. Rev., 27, 1197–1216, https://doi.org/10.1016/j.quascirev.2008.02.016, 2008.
Guiot, J. and de Vernal, A.: Is spatial autocorrelation introducing biases in the apparent accuracy of paleoclimatic reconstructions?, Quaternary Sci. Rev., 30, 1965–1972, https://doi.org/10.1016/j.quascirev.2011.04.022, 2011.
Hill, M. O. and Gauch, H. G.: Detrended Correspondence Analysis – an improved ordination technique, Vegetatio, 42, 47–58, https://doi.org/10.1007/BF00048870, 1980.
Juggins, S.: Quantitative reconstructions in palaeolimnology: new paradigm or sick science?, Quaternary Sci. Rev., 64, 20–32, https://doi.org/10.1016/j.quascirev.2012.12.014, 2013.
Download
Short summary
In spatially structured environments, conventional cross validation results in over-optimistic transfer function performance estimates. H-block cross validation, where all samples within h kilometres of the test samples are omitted is a method for obtaining unbiased transfer function performance estimates. We assess three methods for determining the optimal h using simulated data and published transfer functions. Some transfer functions perform notably worse when h-block cross validation is used.