Predicting impact of eutrophication on lake ecosystem services
Scientists in the PRAGMA Lake Expedition are working to better understand and predict the globally pervasive degradation of lake and reservoir water quality. Meaningful advancements require new understanding of how complex biophysical interactions result in extreme growth of phytoplankton (harmful algal blooms) and ensuing production of noxious substances that make water unsuitable for use. While algae blooms can be detected by sensor networks, the real goal is to predict the circumstances in which blooms may occur, thereby providing advanced warning for water managers and the public who may rely on the services provided by lakes and reservoirs.
Technical advancements that span disciplines and national boundaries are helping lake ecologists improve predictions of algal blooms. Large data sets from sensors embedded in lakes and reservoirs, combined with new simulation models and expert knowledge from the global community show promise for making predictions of the future based on patterns from the past.
Many challenges remain due to:
- their nature, extreme events are difficult to predict and tend to be evident only in high-frequency, large data sets, which have an abundance of pattern, as well as noise, at multiple time scales;
- complex models of ecosystems require thousands of model runs and advanced parameterization techniques for proper calibration;
- validation of predicted bloom events requires pattern matching between simulations and the surprises that occur in the data, and;
- the cyber-infrastructure and compute resources required to support these activities are out of reach for most lake ecologists.
creative application of pattern detection and calssification algorithms adapted from computer science;
easy user access to cyberinfrastructure that brings much needed compute resources to the desktops of ecologists, and;
a community approach to technology systems development that embeds scientists from multiple disciplines in the approach.