Phytoplankton blooms in Lake Erie impacted by both long-term and springtime phosphorus loading

J.C. Ho and A.M. Michalak

Understanding the drivers behind harmful algal blooms is important for setting robust nutrient targets for reducing bloom severity. In Lake Erie, where increasing bloom severity prompted new phosphorus targets, these targets were based on models that are limited in their ability to understand underlying processes operating at longer timescales. In this paper, we explore long-term drivers of phytoplankton blooms in Lake Erie using a new data set spanning thirty-two years (1984-2015), finding that springtime and long-term phosphorus together explain 75% of the variability in bloom size over this period. The fact that long-term phosphorus loading explains bloom severity suggests that recycling of phosphorus trapped in lake sediments is an important factor in explaining blooms. These results indicate that the new phosphorus targets potentially underestimate the time necessary to achieve bloom reductions. The potential for delays in lake recovery to reduced nutrient inputs may require patience for water managers moving forward.

Figure: Statistical model predictions compared to historical bloom size from multiple space-based sensors for Lake Erie. The statistical model (sum of blue and orange regions) based on April to July Dissolved Reactive Phosphorus (blue) and long-term 9-year Cumulative Dissolved Reactive Phosphorus (orange) explains 75% of the interannual variability in maximum summertime bloom extent (green and grey bars).


Harmful algal blooms in Lake Erie have been increasing in severity over the past two decades, prompting new phosphorus loading target recommendations. We explore long-term drivers of phytoplankton blooms by leveraging new estimates of historical bloom extent from Landsat 5 covering 1984-2001 together with existing data covering 2002-2015. We find that a linear combination of springtime and long-term cumulative dissolved reactive phosphorus (DRP) loading explains a high proportion of interannual variability in maximum summertime bloom extent for 1984-2015 (R2 = 0.75). This finding suggests that the impacts of internal loading are potentially greater than previously understood, and that the hypothesized recent increased susceptibility to blooms may be attributable to high decadal-scale cumulative loading. Based on this combined loading model, achieving mild bloom conditions in Lake Erie (defined in recent studies as bloom areas below 600 km2 nine years out of ten) would require DRP loads to be reduced by 58% relative to the 2001–2015 average (equivalent to annual DRP loading of 240 MT and April to July DRP loading of 78 MT). Reaping the full benefits of load reductions may therefore take up to a decade due to the effects of historical loading.

Ho, J.C., A.M. Michalak (2017) "Phytoplankton blooms in Lake Erie impacted by both long-term and springtime phosphorus loading", Journal of Great Lakes Research, 43 (3), 221-228, doi:10.1016/j.jglr.2017.04.001.