Copyright © 2022 ISEIS. All rights reserved
Long-term Water Quality Variations and Chlorophyll a Simulation with an Emphasis on Different Hydrological Periods in Lake Baiyangdian, Northern China
Eutrophication and water quality degradation comprise one of the most important environmental problems associated with protecting freshwater. Here, systematical analyses of trends, qualitative and quantitative analyses of water quality variables, and simulations of eutrophication were conducted to evaluate biochemical oxygen demand (BOD), total phosphorus (TP), total nitrogen (TN), dissolved oxygen (DO), chlorophyll a (Chl a), and Secchi disk data (SD) based on separate hydrological periods to enhance our understanding of lake ecosystem restoration. Long-term trends were identified using seasonal-trend decomposition with local error sum of squares, while non-supervised artificial neural networks were used to identify qualitative characteristics, and quantitative characteristics were measured using statistical analyses. Numerical simulation of Chl a by the hybrid evolutionary algorithm provided a theoretical solution for ecological warnings. The results were as follows: (1) declining trends in BOD, TP, TN, DO and Chl a were observed during long-term seasonal decomposition after December 2006, but SD increased after June 2003; (2) partitioned K-means maps revealed quantitative characteristics with heterogeneous changes during three hydrological periods, with BOD, TN, SD and Chl a showing the highest clustering quality; (3) BOD and DO showed clear relative hierarchies when compared with other parameters based on quantitative analysis; (4) Chl a simulation revealed heterogeneous changes in the three hydrological periods, and sensitivity analyses indicated that BOD was highly sensitive to Chl a, but TP was not. The sensitivities of other parameters changed during different hydrological periods. The methods described here can be used as preliminary management tools for degraded lakes.
Keywords: Artificial neural networks, chlorophyll a, hybrid evolutionary algorithm, hydrological period, Lake Baiyangdian, water quality variations
- There are currently no refbacks.