Two lakes were studied for two years to test for spatial early warning
statistics (EWS) prior to an experimentally induced algal bloom. In 2018,
both Peter and Paul lakes were unmanipulated and spatial measurements of
each lake were taken weekly from June 6th to August 21st to establish
baseline conditions and EWS values. In 2019, nutrients were added to Peter
Lake while Paul Lake remained an unmanipulated reference lake. Both lakes
were measured three times per week from May 29th to September 4th. More
details on nutrient additions (loading rates, N:P ratios) are provided in
Buelo et al. 2022 (Ecological Applications, link below).
The FLAMe sampling platform is fully described in Crawford et al. 2015
(Environmental Science and Technology, 49:442-450). Briefly, the platform is
a boat-mounted system that continuously pumps water from the lake through a
sensor array. Water quality measurements are collected every second (1 Hz)
and matched to concurrently taken GPS coordinates, as well as corrected for
the residence time of the sensor chamber and response time of the sensors
(“tau” corrected). Water quality measurements were taken by a YSI Inc. EXO2
multi-parameter sonde and include dissolved oxygen percent saturation, pH,
and phycocyanin/blue-green algae fluorescence (BGApc). Phycocyanin
concentrations (micrograms per liter) are based on manufacturer-provided
calibration curves but should be interpreted as relative measurements as we
did not calibrate the sensor to extracted pigment concentrations or
microscopy counts. The FLAMe system was mounted in a small, flat-bottomed
boat propelled by an electric motor at approximately 1.5 m/s, which was
driven in a grid pattern with approximately 15 m between tracks.
An R package utilizing these data for the evaluation of early warning
statistic methods to predict algal blooms is available on GitHub
(https://github.com/cbuelo/tvsews) and archived on Zenodo
(https://doi.org/10.5281/zenodo.5874868).