UHF: USING ULTRA-HIGH FREQUENCY SENSOR DATA PROVIDES A LEADING INDICATOR OF HARMFUL ALGAL BLOOMS IN AN OLIGOTROPHIC LAKE

Harmful algal blooms (HABs) threaten the quality of many water bodies, even oligotrophic lakes once thought to be immune from algal blooms. Episodic HABs are best characterized as aggregations of cyanobacteria at the water surface and represent a rapid concentration of deep-water colonies previously dispersed throughout the water column. Predicting the onset and duration of HABs has proven challenging, as the dynamic nature of these events can be difficult to characterize using traditional high-frequency sensor technology. While dense surface aggregations may be easier to identify using traditional sensor methods, the same algal colonies can be much more difficult to detect when dispersed throughout the mid-water column. This is especially true in oligotrophic lakes where phytoplankton biomass is low to begin with. While monitoring mid-column phytoplankton can be challenging in oligotrophic lakes, identifying the sometimes-subtle dynamics of deep water cyanobacteria assemblages is essential to further understanding the development of HABs, and may provide a useful leading indicator of surface HABs. Here, we use advanced vertical profiling sensor platforms deployed in an oligotrophic lake (Lake George, NY, USA) to study HAB dynamics. Our results show that a novel use of ultra-high frequency (<1 second) data can be used to detect mid-column cyanobacterial dynamics that precede surface HABs. These approaches enable the possibility of characterizing mid-column cyanobacterial development and provide important insight for both ultimate and proximal drivers of surface HABs.  

Primary Presenter: Vincent Moriarty, Rensselaer Polytechnic Institute/ IBM Research (vincent.w.moriarty@ibm.com)

Authors:

Vincent Moriarty, Rensselaer Polytechnic Institute/ IBM Research  (vincent.w.moriarty@ibm.com)

Michael Kelly, Rensselaer Polytechnic Institute/ IBM Research  (mrkelly@us.ibm.com)

Guillaume Auger, IBM Research  (Guillaume.Auger@ibm.com)

Mark Lucius, Rensselaer Polytechnic Institute  (lucium@rpi.edu)

Jonathan Borrelli, Rensselaer Polytechnic Institute  (borrej4@rpi.edu)

Harry Kolar, IBM Research  (kolar@us.ibm.com)

Kevin Rose, Rensselaer Polytechnic Institute  (rosek4@rpi.edu)