Multiscale Assessment of Freshwater Water Quality Sensor Data of the St. Lawrence River

Session: 59. - Innovative Monitoring across the Great Lakes

El-Amine Mimouni, St. Lawrence River Institute for Environmental Sciences, [email protected]
Michael Twiss, Clarkson University, Dept. of Biology & Great Rivers Center, [email protected]
Joseph Skufca, Clarkson University, [email protected]
Jeff Ridal, St. Lawrence River Institute, of Environmental Sciences, [email protected]

Abstract

In order to develop better management policies and to confront forecasts with observed data, actual environmental data are essential. To that end, monitoring ecosystem across time for environmental data is an invaluable approach. More affordable sensors and increased computational capacity have made very large datasets more common. However, these datasets do not only bring a computational challenge by their size, but also a theoretical challenge by the scale of the processes they encompass. The issue of “scale” is a well-known concept, but one that is rarely integrated into analyses. We carried out a multiscale assessment of water quality sensor data collected from the REASON project. Using scale-dependent methods, we show that consideration of scale-dependent processes can lead to increased predictive power and a better understanding of ecosystems. Our results also suggest that multiscale methods are not only an alternative way of approaching long-term data but rather a necessity in order to avoid obtaining erroneous results.  Consequently, the concept of scale should be consistently integrated into long-term data studies.

1. Keyword
biomonitoring

2. Keyword
St. Lawrence River

3. Keyword
model testing

4. Additional Keyword
Multiscale analysis