Phytoplankton production in relation to simulated hydro- and thermodynamics during a hydrological wet year – Goczałkowice reservoir (Poland) case study

Abstract

Phytoplankton is one of the crucial components of water body ecosystems. Its presence and development depend on biological, physical and chemical factors and in consequence it is an important indicator of ecosystem condition. Monitoring of phytoplankton production, measured as chlorophyll a concentration, is a useful tool for assessing the status of dam reservoirs. Modeled chlorophyll a concentrations are used as water quality indicators in locations not included in monitoring systems, in situations when the temporal resolution of the monitoring is not enough, and in assessments of the impacts of future activities. Therefore, the aim of this study was to find correlations between hydro- and thermodynamics and the chlorophyll a concentration for possible application in reservoir monitoring and management, using an ELCOM-CAEDYM model. The analysis included summer and fall which are most prone to algal blooms, and four phytoplankton groups identified as dominant in the reservoir based on periodic observations. Comparisons of simulated water temperature and both observed and simulated chlorophyll a concentrations confirmed that these variables are significantly correlated (correlation of hourly chlorophyll a and water temperature was 0.70, ranging from 0.55 to 0.81 in the bottom and surface water layers, respectively, while for daily outputs it was 0.74, ranging from 0.60 to 0.83). This relation was stronger than that of chlorophyll a to nutrient (N, P and Si) concentrations. What is more, the method used allowed the assessment of a much more detailed spatial and temporal distribution of phytoplankton groups compared with conventional monitoring techniques. The study indicated that the phytoplankton community was dominated by Chlorophytes and Diatoms with a larger share of Chlorophytes in shallow parts of the reservoir. This domination was weaker after short water mixing events in summer and especially after the fall turnover. The increase in phytoplankton diversity was estimated to occur mainly near the surface and in shallow parts of the reservoir. Most of the observed concentrations of individual phytoplankton groups differed from simulation results by less than 25% and the model reflected accurately 74% of observed trends in concentrations. Calculated chlorophyll a concentration was well matched to hourly monitoring data (mean squared error = 5.6, Nash–Sutcliffe model efficiency coefficient = 0.51, Pearson correlation coefficient = 0.72 and p-value = 0.0007). High compatibility of the model to the values measured in the reservoir make it a promising tool for the prediction and planning of actions aimed at maintaining good functioning of the reservoir.