In this work, we address the formulation of an agroecohydrological dynamic optimization model that integrates AI tools to first principle models for the management of lake basins, taking into account environmental impact of extreme events. The mitigation of the consequences of extreme environmental events, such as floods and droughts, has been addressed as an optimal control problem (Siniscalchi et al., 2022). An extension of the optimization model is proposed for the sustainable management of ecohydrological processes, agriculture and livestock within lake basins. The case study is a salt lake basin. There are several goals to fulfil within management and mitigation strategies: (i) to prevent flooding in a nearby village and its touristic areas during wet periods by diverting part of the flow from its tributary into an artificial reservoir (the diversion flowrate is a control variable); (ii) to optimize management of the artificial reservoir to keep lake salinity within desired levels for silverside fish during drought periods (the stream flowrates from this artificial reservoir to the salt lake and to irrigation purposes are control variables); (iii) to include restoration strategies for native specieof trees, currently existing in the salt lake basin, combining plantation of
Prosopis flexuosa and
P. caldenia with drought resistant crops (
Chenopodium quinoa) and pasture, and (iv) to provide drinking water and shade to cattle. Another objective of this work is to perform life cycle assessment (LCA) for the agricultural and livestock system under study. AI/ML tools have been used to obtain climatologic variable profiles (wind, temperature, relative humidity, heliophany, solar radiation, etc.). Dynamic mass balances have been formulated for a salt lake and an associated artificial freshwater lake. The objective function is an integral one that aims to keep the salt lake volume (and its associated salinity, as it is an endorheic basin), at a desired value, to avoid flooding of the nearby village and to keep salinity at optimal values for reproduction of a valuable fish species. The resulting optimal control problem is constrained with a DAE system of equations representing the above-mentioned processes. Energy calculations are performed for pumps used in drip irrigation for crop and pasture. Numerical results show that the proposed management strategies not only mitigate extreme events, but provide economical and natural decarbonization within a currently unproductive area. Negative environmental impacts due to tree plantations, drought resistant crops and pasture are slightly larger than positive impacts associated to cattle CO2-eq emissions (gases and manure), aiming at the net zero-carbon goal. The model has proven to be an efficient computational tool for the sustainable management in the context of the water-food-energy-carbon nexus of a salt lake and its basin.
References
Siniscalchi, A, Diaz, M.S, Lara, R.J (2022). Sustainable long-term mitigation of floods and droughts in semiarid regions:Integrated optimal management strategies for a salt lake basin.Ecohydrology. 2022;15:e2396
