Using machine learning to correct for nonphotochemical quenching in high-frequency, in vivo fluorometer data

In vivo fluorometers use chlorophyllafluorescence (F-chl) as a proxy to monitor phytoplankton biomass. However, the fluorescence yield ofF(chl)is affected by photoprotection processes triggered by increased irradiance (nonphotochemical quenching; NPQ), creating diurnal reductions inF(chl)that may be mistaken for phytoplankton biomass reductions. Published correction methods are mostly designed for pelagic oceans and are ill suited for inland waters or for high-frequency data collection. A machine learning-based method was developed to correct vertical profiler data from an oligotrophic lake. NPQ was estimated as a percent reduction inF(chl)by comparing daytime values to mean, unquenched values from the previous night. A random forest regression was trained on sensor data collected coincident withF(chl); including solar radiation, water temperature, depth, and dissolved oxygen saturation. The accuracy of the model was assessed using a grouped 10-fold cross validation (mean absolute error [MAE]: 7.6%; root mean square error [RMSE]: 10.2%), which was then used to correctF(chl)profiles. The model also predicted NPQ and corrected unseenF(chl)profiles from a future period with excellent results (MAE: 9.0%; RMSE: 14.4%).F(chl)profiles were then correlated to laboratory results, allowing corrected profiles to be compared directly to collected samples. The correction reduced error (RMSE) due to NPQ from 0.67 mu g L(-1)to 0.33 mu g L(-1)when compared to uncorrectedF(chl)data. These results suggest that the use of machine learning models may be an effective way to correct for NPQ and may have universal applicability.