[1] https://www.sciencedirect.com/science/article/pii/S007966111...
>This work builds upon an earlier analysis (Boyce et al., 2010) by taking published criticisms into account, and by using recalibrated data, and novel analysis methods.
Having looked at both papers now, I think that the 2014 followup confirmed the direction of the trend (downward), but it looks like it also significantly reduced the magnitude of the trend.
Look at figure 2b in the 2010 paper and figure 1a in the 2014 paper. They both show "Rate of Chl change (mg m–3 yr–1)" on a grid of cells covering the world's oceans. But the scale in the 2010 paper went from -0.40 to 0.40. In the 2014 paper the extremes of the scale are only -0.10 and 0.10. It looks like re-analysis indicated less dramatic rates of change both in the cells showing increase and those showing decrease.