(254bm) Ecological Study, Parameter Estimation, Validation and Lake Restoration Planning with Dynamic Optimization Strategies
Cyanobacterial algal blooms development in water bodies is mainly associated to anthropogenic related pollution that includes nitrogen and phosphorus downloads from point and nonpoint sources, such as industrial and domestic wastewater and agricultural activities, respectively. During the last decades, freshwater scientists have argued, based on the ability of some species to fix atmospheric nitrogen, that phosphorus is the main factor that controls cyanobacteria blooms. However, recently published experimental evidence shows that nitrogen plays a major role together with phosphorus in the control of algae blooms (Paerl et al., 2011a; Paerl et al., 2011b) and even more with iron dynamics (Molot et al., 2014). The US Environmental Pollution Agency has released a new guidance document (EPA, 2015) that gives scientific support to the dual nutrient criteria (nitrogen and phosphorus). Bloom conditions prediction and control methods constitute a priority in freshwater sources. Mechanistic models are powerful tools to describe freshwater ecosystems and to evaluate and planning water restoration strategies. In previous work, Estrada et al. (2011, 2015) develop a mechanistic ecological model that includes dynamic mass balances for phytoplankton groups; zooplankton groups and local zooplanktivorous fish, as well as dissolved oxygen and main nutrients. Algebraic equations represent forcing functions profiles, such as temperature, solar radiation, river inflows and concentrations, etc. Even though a set of experimental data (2004) was used to calibrate the model, no data were available for zooplankton and fish dynamics.
In this work, we address monitoring, parameter estimation, validation and restoration planning for Paso de las Piedras Reservoir, a non-stratified lake which is the drinking water source for two cities in Argentina. Field data is being collected throughout 2014-2015 and include nutrient, phytoplankton, dissolved oxygen, macro and microzooplankton, as well as zooplanktivorous fish biomass concentrations. Phytoplankton and zooplankton data are collected weekly and twice a month, respectively, while fish data are collected every three months during the first year and bimonthly in the second year. The study of phytoplankton community allows the identification of 189 taxa, three of six groups were determined as the most important in terms of number of species and biomass. Zooplankton is represented by cladocera and copepoda groups. In the case of fish community, experimental data show that it is represented by ten taxa, from which the most representative in terms of biomass are Odontesthes bonariensis and Oligosarcus hepsetus. Based on ecological study results we incorporate the dynamic of Oligosarcus hepsetus population, which is a predator of Odontesthes bonariensis. This is the first time an integral monitoring project has been carried out, including fish and zooplankton communities. The inclusion of main fish dynamics is relevant to evaluate the application of one of the most widely used restoration strategies, biomanipulation (SÃ¸ndergaard et al., 2007), together with other strategies like wetlands. Collected data during 2014 have been used to calibrate the ecological model and data from 2015 have been used for validation. Both parameter estimation and dynamic optimization problems have been formulated within an equation oriented optimization framework in gPROMS (PSEnterprise, 2015). The re-calibrated model provides a reliable tool for restoration planning.
Di Maggio, J., Estrada V., M.S. Diaz, Water Resources Management with Dynamic Optimization Strategies and Integrated Models of Lakes and Artificial Wetlands, PSE2015 â?? ESCAPE25, 12th International Symposium on Process Systems Engineering and 25th European Symposium on Computer Aided Process Engineering , 31 May - =4 June 2015, Copenhagen, Denmark
Estrada V., J. Di Maggio, M.S. Diaz (2011), Water Sustainability: A Process Systems Engineering approach to the restoration of eutrophic lakes, Computers and Chemical Engineering, 35, 8, 1598-1613
Jeppesen, E., SÃ¸ndergaard, M., Lauridsen, T., Davidson, T., Liu, Z., Mazzeo, N., Trochine, C., Ã?zkan, K., Jensen, H., Tolle, D., Starling, F, Lazzaro, X., Johansson, L., Bjerring, R., Liboriussen, L., Larsen, S., Landkildehus, F., Egemose, S. & Meerhoff, M. (2012). Biomanipulation as a Restoration Tool to Combat Eutrophication: Recent Advances and Future Challenges. Advances in Ecological Research, 47.
Paerl, H. and Otten, T. (2013) Harmful Cyanobacterial Blooms: Causes, Consequences, and Controls. Environmental microbiology, 65, 995-1010.
Process Systems Enterprise (2013), gPROMS, www.psenterprise.com/gproms
SÃ¸ndergaard, M., Jeppesen, E., Lauridsen, T. L., Skov, C., van Nes, E. H., Roijackers, R., Lammens, E. & Portielje, R. (2007). Lake restoration: successes, failures and long-term effects. Journal of Applied Ecology, 44, 1095â??1105.