REAL-TIME, ADAPTIVE, SELF-LEARNING MANAGEMENT OF LAKES
|dc.identifier.citation||Imberger, J. and Marti, C. and Dallimore, C. and Hamilton, D. and Escriba, J. and Valerio, G. 2018. REAL-TIME, ADAPTIVE, SELF-LEARNING MANAGEMENT OF LAKES, 37th IAHR World Congress.|
Lakes and reservoirs are increasingly threatened by anthropogenic activities, with serious environmental and financial consequences. In particular, nutrient loadings are increasing due to expanding human populations and food demand, and thermal stratification is increasing due to global warming. For deep lakes, this is leading to an increase of the seasonal water column stability, extending the duration of seasonal deoxygenation of hypolimnetic waters, as well as increasing the nutrient inventory within the bottom sediments. The large volume of deoxygenated hypolimnetic water and nutrient-enriched bottom sediments increase nutrient releases. This overturn brings the deoxygenated, nutrient rich hypolimnetic water into the euphotic zone, leading to increased primary productivity, organic enrichment, and providing a feedback mechanism for further degradation of the fauna and flora living in the epilimnion, potentially culminating in total death of aerobic organisms by asphyxiation. Reservoirs and shallow lakes are increasingly subjected to toxic algal blooms in response to changing patterns of stratification in combination with the increasing nutrient loadings. We use two examples to illustrate the problems presently encountered, the range of control strategies available to manage the consequences and then show how adaptive, real-time, self-learning technologies may be used to dynamically optimize the ecosystem health, as both the impacts and the system change with time. The first example is deep Lake Iseo in Italy. The period between overturns has increased from around every ten years to twenty years in Lake Iseo over the last 50 years. We show that the water column stratification may be controlled with solar-powered impellers, allowing the frequency of overturning to be regulated so as to prevent the buildup of large volumes of low-oxygen hypolimnetic waters. The second is shallow Lake Ypacarai in Paraguay. It has undergone rapid eutrophication, resulting in severe toxic algal blooms that are having a devastating impact on the economy of Paraguay. Using the numerical simulations of the lake ecosystem, we carried out a sensitivity analysis of the available controls for the mitigation of the algal blooms. The simulations demonstrate how a decrease in the nutrient loadings, decrease in the water levels, flushing of the lake with bore or river water and increase in water opacity provide multiple potential bloom control mechanisms. The results for both lakes indicate that a real-time adaptive management system using model forecasts could optimize environmental controls on attributes (e.g. cyanobacteria and dissolved oxygen) that would otherwise seriously deteriorate and impact the ecological processes throughout the lake ecosystems.
|dc.title||REAL-TIME, ADAPTIVE, SELF-LEARNING MANAGEMENT OF LAKES|
|dcterms.source.title||Proceedings of the 37th IAHR World Congress|
|dcterms.source.series||Proceedings of the 37th IAHR World Congress|
|dcterms.source.conference||37th IAHR World Congress|
|curtin.department||Sustainable Engineering Group|
|curtin.accessStatus||Fulltext not available|
Files in this item
There are no files associated with this item.