Correction to “Response of planktic foraminiferal size to late Quaternary climate change”

[1] In the paper “Response of planktic foraminiferal size to late Quaternary climate change” by Daniela N. Schmidt, Sabrina Renaud, and Jörg Bollmann (Paleoceanography, 18(2), 1039, doi:10.1029/2002PA000831, 2003), the captions for Figures 2, 5, 6, and 7 were incorrect. The correct captions appear below. [2] Figure 2. Late Pleistocene sizeassemblage5 changes compared to paleotemperature (Wefer et al. [1996], winter temperature; Brathauer and Abelmann [1999], summer temperature; Mix et al. [1999], winter temperature) and paleoproductivity [Wefer et al., 1996; Mackensen et al., 2001; P. Müller, unpublished data, 1999] at the three core locations. Numbers represent the marine isotope stages, and shaded areas indicate glacial stages. Error bars on size assemblage5 represent the accuracy of the 95th percentile measurement. [3] Figure 5. Variability of size assemblage5 (μm) (black circles), the ratio of tropical to upwelling planktic foraminifera species (gray triangles), and the ratio of symbiont bearing to nonsymbiont bearing species (gray squares) during the past 250 ka in core GeoB1105. Ratios were estimated from data published by Wefer et al. [1996]. Error bars on size assemblage5 represent the accuracy of the 95th percentile measurement. Shaded areas indicate glacial stages. [4] Figure 6. Comparisons of Holocene and late Quaternary size and temperature relationships. (a) Maximum size (μm) of G. truncatulinoides, G. inflata, and G. crassaformis in Holocene and late Quaternary samples versus paleotemperature. Open symbols represent Holocene data (Schmidt et al., submitted manuscript, 2003); filled symbols represent late Quaternary samples from isotope stages 2 to 8. (b) Maximum sizes for investigated species in the Holocene compared to maximum sizes in the late Quaternary. Species from top to bottom are: Globorotalia menardii, G. tumida, Globigerinoides sacculifer, Orbulina universa, Globorotalia truncatulinoides, Globigerinoides ruber, Globigerina bulloides, Globorotalia inflata, Neogloboquadrina pachyderma. The line indicates 1:1 relationship. (c) Optimum sea surface temperatures for Holocene species versus derived optimum (paleo) sea surface temperatures in isotope stages 2 to 8 for species investigated (for species names, see Figure 6b). Ecological optimum temperatures were derived from samples where a species reaches its largest test size and its maximum abundance (for references of temperature estimates, see Figure 2). Note that instead of using the temperature estimates as in Figure 2, the summer temperature estimates were chosen for all sites as optimum paleotemperatures in order to obtain consistency of temperature estimates for all three cores. The line indicates 1:1 relationship. [5] Figure 7. Variability of sizeassemblage5 (μm) and sizetrunc (μm) during the last 130 kyrs in cores PS2498, GeoB1413, and GeoB1105. Shaded areas represent glacial stages and substages. Numbers represent the marine isotope stages. Error bars on sizeassemblage5 represent the accuracy of the measurements and error bars on the average size of G. truncatulinoides represent the standard deviation for the mean.