Erratum to: Aquat Sci 69 (2007) 320–329 DOI 10.1007/s00027-007-0894-1

It has come to our attention that there was a consistent error in the calculation of the DMSP-lyase activity (DLA) values presented in Bell et al. (2007). We present the corrected values to enable accurate comparison with past and future studies. The relative patterns discussed in Bell et al. (2007) and the conclusions drawn are broadly unaffected. Figures 1, 2, 3, 4 (below) use the recalculated data and replace the plots shown in Bell et al. (2007). We also present the recalculated DLA data in a corrected version of Table 1 from Bell et al. (2007), which still demonstrates the same trends when comparing pre-dawn and late-morning CTD casts.

Fig. 1
figure 1

Latitudinal distribution of data for DLA (nmol L−1 min−1) as plotted in Fig. 2a from Bell et al. (2007) but using the corrected values (note that the units have changed). Open circles denote surface samples and filled squares denote chlorophyll maximum samples. In this and subsequent figures, sections IIV (see text) are labelled along the top of the figure and indicated by solid vertical lines

Fig. 2
figure 2

Latitudinal distribution of data for DLA:TChl a (nmol min−1 μg−1 TChl a) as plotted in Fig. 2c from Bell et al. (2007) but using the corrected values (note that the units have changed). Open circles (HPLC chlorophyll) and squares (acetone-extracted chlorophyll) denote surface samples. Filled squares denote chlorophyll maximum samples

Fig. 3
figure 3

Latitudinal distribution of data for DLA:DMSPp (min−1) as plotted in Fig. 3d from Bell et al. (2007) but using the corrected values (note that the units have changed). Open circles denote surface samples and filled squares denote chlorophyll maximum samples

Fig. 4
figure 4

Plot of phytoplankton carbon fixation (mmol C m−3 d−1) versus DLA:TChl a (nmol min−1 μg−1 TChl a) as plotted in Fig. 4 from Bell et al. (2007) but using the corrected values (note that the units have changed). Open circles (HPLC chlorophyll) and squares (acetone-extracted chlorophyll) denote surface samples. Filled squares denote chlorophyll maximum samples. The mean concentrations for carbon fixation (0.28 mmol C m−3 d−1) and DLA:TChl a (20.8 nmol min−1 μg−1 TChl a) are plotted as solid lines on the vertical and horizontal axes

Table 1 Surface DLA values from pre-dawn (PD) and late-morning (LM) CTD casts as shown in Table 1 from Bell et al. (2007) but using the corrected values (note that the units have changed). Each PD and LM sample pair was selected from the same day and was from the same section (sections described in original paper). The final column represents the difference between LM and PD. Sections I and III demonstrate elevated DLA in the LM cast relative to the same day’s PD cast, with the exception of two days in section III (20 May 2004 and 23 May 2004)

When considering the recalculated DLA data, two points should be noted. Firstly, the potential turnover times calculated in Bell et al. (2007) are incorrect. The corrected values are in closer agreement with the 5 min average turnover time presented in Harada et al. (2004), with recalculated potential DMSPp turnover times for our entire study ranging from 0.02 min (approx. 1 s) to 35 min. The recalculated average turnover times for sections I/III and II/IV are 3.8 and 5.7 min, respectively. Recalculating the turnover times using total DMSP generates average values of 7.0 min for sections I/III and 11.1 min for sections II/IV.

Secondly, whilst the 14C-derived phytoplankton carbon fixation and recalculated DLA:TChl a ratio data (Fig. 4) still demonstrate the same pattern as in Bell et al. (2007), the strength of correlation between the two has reduced slightly (Spearman’s ρ changes from –0.57 to –0.50), although this is still a significant result (p < 0.05, n = 17).