Understanding Water Quality Dynamics of the Lake Water Column using Modular Compositional Learning

Water quality in lakes is variable throughout the water column. The drivers of this variability are challenging to measure due to the high-frequency measurements required, but we can gain a greater understanding of this variability through modeling. Modular Compositional Learning (MCL) is a framework within the Ecology Knowledge-Guided Machine Learning (Eco-KGML) paradigm that allows ecosystem processes to be segmented and individually modeled using either process-based models or machine learning. Water quality emerges as a result of multiple processes including aquatic metabolism and physics. MCL has the potential to be an effective modeling approach for water quality due to its multi-process nature. We have constructed a one-dimensional lake model using MCL to investigate the relationship between surface and subsurface water quality dynamics to gain a greater understanding of the drivers of variability along the water column.