Corrigendum: Climate Driven Changes in Timing, Composition and Magnitude of the Baltic Sea Phytoplankton Spring Bloom

A Corrigendum on
Climate Driven Changes in Timing, Composition and Magnitude of the Baltic Sea Phytoplankton Spring Bloom

by Hjerne, O., Hajdu, S., Larsson, U., Downing, A. S., and Winder, M. (2019). Front. Mar. Sci. 6:482. doi: 10.3389/fmars.2019.00482

In the original article, there was a mistake in Figure 6A as published. This figure shows the regression analysis results of group composition at the offshore station BY31 and not at the coastal station B1 as stated in the text. The corrected Figure 6A appears below.

FIGURE 6

Figure 6. Regression analysis results of group composition (the relative contribution of diatoms, dinoflagellates and Mesodinium rubrum to the average biomass in March–May) at the coastal station (A, B1) and offshore station (B, BY31) in the northern Baltic Proper. The solid and dashed black lines are the fitted regression lines and the 95% confidence intervals of the average models, respectively. Red lines are the fitted regression lines of the individual models that contribute to the average model. If the slopes of a predictor are either all positive or negative in the individual models there is strong support for an effect of the predictor, while different signs or lack of slope indicate uncertainty of its effect. The black circles represent the partial residuals from the average model. p-values within the panels and the weighted R2 (wR2) values are based on the AICc weights. Model results should be interpreted with care when model validation plots indicate violation to the model assumptions: ⁽¹⁾ Possibly slightly non-normal distribution of residuals (long lower tail), ⁽²⁾ Just significant negative temporal autocorrelation at lag 3 and higher variance of residuals at low fitted values.

The authors apologize for this error and state that this does not change the scientific conclusions of the article in any way. The original article has been updated.