Jiacong Huang

Modeling the effects of environmental variables on short-term spatial changes in phytoplankton biomass in a large shallow lake, Lake Taihu

Understanding the short-term response of phytoplankton biomass on environmental variables is needed for issuing early warnings of harmful algal blooms in aquatic ecosystems. Predicting harmful algal blooms are particularly challenging in large shallow lakes due to their complex mixing patterns. This study used a two-dimensional hydrodynamic–phytoplankton model to evaluate the effects of environmental variables on short-term changes in the horizontal distribution of phytoplankton biomass in a large shallow lake, Lake Taihu, China. Two simulations were performed using daily and hourly average wind condition and water temperature data collected in 2009. Other model inputs were identical for these two simulations. The response of phytoplankton to wind conditions, light intensity, water temperature, and total dissolved phosphorus and nitrogen concentrations were examined based on a sensitivity analysis using the hourly data. Hourly simulation achieved a more realistic distribution of phytoplankton biomass than the daily simulation. This finding implies that data with a higher temporal resolution are more useful for short-term prediction of phytoplankton biomass in this lake. Sensitivity analysis indicated that water temperature and light intensity dominate short-term changes in phytoplankton biomass in this lake. Wind conditions also affect phytoplankton biomass distribution by causing advective water movement.

Modeling daily chlorophyll a dynamics in a German lowland river using artificial neural networks and multiple linear regression approaches

Phytoplankton biomass is an important indicator for water quality, and predicting its dynamics is thus regarded as one of the important issues in the domain of river ecology and management. However, the vast majority of models in river systems have focused mostly on flow prediction and water quality with very few applications to biotic parameters such as chlorophyll a (Chl a). Based on a 1.5-year measured dataset of Chl a and environmental variables, we developed two modeling approaches [artificial neural networks (ANN) and multiple linear regression (MLR)] to simulate the daily Chl a dynamics in a German lowland river. In general, the developed ANN and MLR models achieved satisfactory accuracy in predicting daily dynamics of Chl a concentrations. Although some peaks and lows were not predicted, the predicted and the observed data matched closely by the MLR model with the coefficient of determination (R2), Nash–Sutcliffe efficiency (NS), and the root mean square error (RMSE) of 0.53, 0.53, and 2.75 for the calibration period and 0.63, 0.62, and 1.94 for the validation period, respectively. Likewise, the results of the ANN model also illustrated a good agreement between observed and predicted data during calibration and validation periods, which was demonstrated by R2, NS, and RMSE values (0.68, 0.68, and 2.27 for the calibration period, 0.55, 0.66 and 2.12 for the validation period, respectively). According to the sensitivity analysis, Chl a concentration was highly sensitive to dissolved inorganic nitrogen, nitrate–nitrogen, autoregressive Chl a, chloride, sulfate, and total phosphorus. We concluded that it was possible to predict the daily Chl a dynamics in the German lowland river based on relevant environmental factors using either ANN or MLR models. The ANN model is well suited for solving non-linear and complex problems, while the MLR model can explicitly explore the coefficients between independent and dependent variables. Further studies are still needed to improve the accuracy of the developed models.

Towards better environmental software for spatio-temporal ecological models: Lessons from developing an intelligent system supporting phytoplankton prediction in lakes ☆

Abstract Implementing a case study using existing spatio-temporal ecological models could be time-consuming and error-prone. To alleviate this problem, several strategies, aiming to achieve a robust but easy-to-use environmental software, were used to develop an intelligent system supporting phytoplankton prediction in Lakes (iLake). This environmental software coupled three modules (a two-dimensional hydrodynamic module, a mass-transport module and a phytoplankton kinetics module) together to predict the time dynamics of phytoplankton distribution in a lake. A case study of phytoplankton prediction in Lake Taihu using iLake demonstrated its high potential, but low learning curve, for lake modeling.

Combination of artificial neural network and clustering techniques for predicting phytoplankton biomass of Lake Poyang, China

A single artificial neural network (ANN) model is inadequate for predicting phytoplankton biomass in a large lake due to its high spatial heterogeneity. In this study, ANN was combined with a clustering technique to simulate phytoplankton biomass in a large lake (Lake Poyang) using a 7-year dataset. Two ANN models (named ANN_Downstream and ANN_Upstream) were developed for the downstream and upstream areas based on the k-means clustering results of 17 sampling sites at Lake Poyang, China. They performed better than ANN_Poyang (an ANN model for the whole lake), indicating the success of the clustering technique in improving ANN models for predicting phytoplankton biomass in different sub-regions of the large lake. A sensitivity analysis based on ANN_Downstream and ANN_Upstream showed that phytoplankton dynamics responded differently to environmental variables in different sub-regions of Lake Poyang. This case study demonstrated the good performance of ANN models in describing phytoplankton dynamics, and the potential of coupling ANN with a clustering technique to describe the spatial heterogeneity of natural ecosystems.

An ensemble simulation approach for artificial neural network: An example from chlorophyll a simulation in Lake Poyang, China

Artificial neural network (ANN) models have been widely used in environmental modeling with considerable success. To improve the reliability of ANN models, ensemble simulations were applied in this study to develop four ANN ensemble models for chlorophyll a simulation in the largest freshwater lake (Lake Poyang) in China. Reliability (evaluated by model fit and stability) of these ANN ensemble models was compared with that of single ANN models from ensemble members. The model fit of these single ANN models varied significantly over repeated runs, indicating the unstable performance of the single ANN models. Comparing with the single ANN models, the ANN ensemble models showed a better model fit and stability, implying the potential of ensemble simulation in achieving a more reliable model. An ensemble size of 30 was adequate for the ANN ensemble models to achieve a good model fit, while an ensemble size of 50 was adequate to achieve good stability. This case study highlighted both the necessity and potential of the ensemble simulation approach to achieve a reliable ANN model with good model fit and stability.

Modeling the effects of the streamflow changes of Xinjiang Basin in future climate scenarios on the hydrodynamic conditions in Lake Poyang, China

The responses of lake hydrodynamics to the hydrological processes in watersheds have been associated with the ecological evolution of and the biochemical processes in aquatic ecosystems. This paper investigates how the changes in the streamflow of Xinjiang Basin in different future climate scenarios could affect the hydrodynamic conditions in Lake Poyang. First, the hydrodynamic processes in Lake Poyang (i.e., lake level and water velocity) were simulated based on the environmental fluid dynamics code (EFDC). Error statistics indicated that the hydrodynamic model reasonably reflected the hydrodynamics in Lake Poyang. Second, the future streamflow of Xinjiang Basin from 2016 to 2050, which was projected in two future climate scenarios (Sim_RCP4.5 and Sim_RCP8.5) based on the Xin’anjiang model, was applied to hydrodynamic modeling to investigate the relationship between future discharge and hydrodynamics. Results showed that the streamflow changes of Xinjiang Basin in future climate scenarios considerably affected the seasonal distribution of the flow field in Lake Poyang. The hydrodynamic change region that exceeded the threshold values under these two climate scenarios both demonstrated a fluctuating trend and nearly covered the entire lake until April. In Sim_RCP8.5 a slightly larger area was influenced than in Sim_RCP4.5, except in January, and the eastern channel was always significantly affected by streamflow change. These analyses can enhance the present understanding of the relationships between the hydrodynamics in lakes and the hydrological processes of sub-basins.