Nonphotochemical quenching (NPQ) is a response mechanism in plants and algae that allows them to process and dissipate excess excitation energy as heat safely. Collecting fluorescence data from these plants and algae in surface water environments can incur errors from NPQ, ultimately leading to inaccurate calculations of chlorophyll concentration for environmental and industrial water quality monitoring. Rensselaer inventors developed a novel approach to correcting NPQ-skewed fluorescence data by employing trained machine-learning modules that can be applied to fluorescence detection systems. Inventors' method eliminates the assumptions and guesswork of adjusting for NPQ by introducing trained machine modules along with input sensor data that are already standardized for an array of water depths. This standardization component, alongside the predictive capabilities of the module, allows for an improved and consistent overall method. The functionality of the method could be beneficial to consumer end-users as well as corporate partners. Advantages of this method include: operability as this can be implemented alongside current fluorescence probes on the market; reproducibility as mass production of the method is feasible; and novelty.