IWRA Proceedings

AbstractIn the framework of AquaStress (an EU funded integrated project), a small reservoir (Enxoé) located in the South of Portugal was studied. The reservoir was built to supply water for human consumption, but it shows several water quality problems namely cyanobacteria blooms. In the last years and as a consequence of these problems, it was not possible to use it to supply water during the summer. In order to try to understand the possible causes of the problems and, hopefully, find proper solutions to solve them, an integrated study involving the catchment, the data available concerning the nutrient loads, the meteorology and the water quality and mathematical models was set up. This approach allowed the suggestion of some possible actions that may lead to improve the conditions within the reservoir. Some major conclusions of data analysis and first model implementation put in evidence that there was a sudden change after the winter 2000/2001 floods, namely a rapid phosphorous enrichment and a rapid decrease of TN:TP ratio, bottom anoxia and presumably phosphorous release from sediments and permanent cyanobacteria dominance since the 2001/2002 floods. The first modelling results also put in evidence that the model was not reproducing very well the behaviour of the reservoir. The main cause for this is probably due to errors in the methods of the loads quantification, although a standard approach based on OSPAR guidelines that showed to be successful in other applications made in the north of Portugal is being used. The cyanobacteria dominance started only after the winter of 2000/2001, indicating that the big floods that occurred on this year and associated erosion, may be responsible for a major source of P for the reservoir. This source is possibly misrepresented by any estimate of loads. In order to try to clarify these aspects it was decided to find similar data from other reservoirs in the neighbourhood (e.g. Monte Novo and Roxo) to validate the thesis that floods might be a triggering mechanism for cyanobacteria dominance. Also to solve the problem of boundary conditions for the reservoir model, a simple inverse model that computes the loads as a function of measurements of Phosphorus in the reservoir and exchanges between the water column and the sediments was used. In this paper a discussion of the influence of the floods in the nutrient loads that are transported to reservoirs in such semi-arid regions will be made based on the data available for Enxoé and Monte Novo reservoirs and an improved erosion/retention model for the catchment able to reproduce the nutrient loads in a more proper way will be proposed.