ECODOURO - Modelling the effect of freshwater reduction and pulse discharge on the water dynamics and processes of the Crestuma Reservoir

• Autor(a) principal: Bordalo, Adriano
• Data de Publicação: 2008
• Outros Autores: Duarte, Pedro, Wiebe, William
• Tipo de documento: Relatório
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10284/833

This report deals with the activities carried out during the project 72 months, from March 15, 2003 to September 15, 2006 under the framework of the ECODOURO - Modelling the effect of freshwater reduction and pulse discharge on the water dynamics and processes of the Crestuma Reservoir (POCTI/MGS/45533/2002). The report includes the fieldwork, data interpretation and mathematical modeling according to the seven working packages presented in the original proposal. Data is available to the public at http://ecodouro.icbas.up.pt. The main objectives of this goal oriented research project implemented on the Crestuma Reservoir (Douro River) are: i. To evaluate the long-term effects of flow reduction on key environmental variables and processes that occur in the Crestuma Reservoir; and ii. To evaluate the effect of high frequency oscillations (freshwater pulse discharges) on the ecosystem dynamics in terms of water column stratification, water temperature and turbidity, oxygen, nutrient availability, phytoplankton biomass and primary productivity. The original requested budget was reduced by 19%, and the evaluation commission recommended the non-acquisition of a second multiprobe CTD. Thus, the entire field survey program was changed accordingly. The seasonal sampling was conducted at two anchor stations located about 500 m and 2,500 m from the Crestuma dam crest. Surveys were performed on December 2003 (Fall), February 2004 (Winter), May 2004 (Spring) and August 2004 (Summer). An un-scheduled survey was carried out at the onset of the program (May 2003), in order to gather basic data since no departure data (short term series) were available contrarily to what was though. This approach proved to be very useful and the obtained results were applied in the following surveys. Moreover, since no workable bathymetrics were available to the research team (only depth contours in chart format restricted to the navigational channel) several dedicated surveys were organized in order to generate bottom contour maps for the last 10 km of the reservoir, covering the two-field stations location. As a result, a costume-made GIS was created and data used to run the model. All technical details of both the employed models (2D-vertically resolved model and the coupled biogeochemical model), including the complete mathematical formulae, are 2 presented in Annexes I and II. The rationale for selection of the modeled parameters are presented; the field data obtained from the 4+1 surveys were used to initialize the models. It was found that the different flow time-scales might influence water dynamics, biogeochemistry and primary productivity. For example, during the day, longitudinal flows results from upstream forcing, with river water entering into the reservoir. A clear longitudinal flow profile emergeed, disturbed in some points by depth changes that forces upwelling of bottom water. During the night, in the absence of river flow, at it happens frequently in Spring through Fall, convective flow dominated, with surface cooled water sinking to the bottom and forcing the ascent of bottom water. Vertical flows may be larger than horizontal ones. On the other hand, the highest phytoplankton production was observed in May and September, when microalgae were using more efficiently the available light. Furthermore, phytoplankton efficiency decreased from morning to dawn. Considering the objectives of this project, several model simulations were carried out to analyze the effects of flow variability on water column stratification and quality (cf. – Annexe III). These simulations were carried out with the same average flow magnitude and forcing conditions, in terms of water temperature, nutrients and chlorophyll concentrations, but different flow frequencies and amplitudes. Results obtained suggest that flow variability may not have a large effect on water temperature, nutrient and chlorophyll concentration and net primary production at annual time scales. However, when the time scale under analysis is reduced to seasonal and monthly, effects of flow variability become apparent, especially during the summer period and when flow is hold constant. Constant flow implies that extreme low and high values do not occur. Under this situation of “extreme” flow regulation, it appears that the synergies between river forcing and reservoir processes tend to reduce phytoplankton biomass. Therefore, from a management point of view, it is apparent that stabilizing the flow may prevent the development of phytoplankton blooms. On the other hand, results presented and discussed in Annex IV suggest that the Crestuma reservoir has undergone some important changes over the period 1999-2005, with significant increases in nitrogen and phosphorus and a decrease in the nitrogen: phosphorus ratio i.e. the Redfield ratio. One possible change, arising from the shifts in the nitrogen: phosphorus ratio, is the apparent autumn shift towards nitrate-nitrogen limitation of phytoplankton. Apart from increasing nutrient runoff within the watershed, the 3 noticeable nutrient increase trend found in the Crestuma resevoir may also be explained by the reduction of chlorophyll as a result of the massive development of the invasive clam Corbicula sp. Thus, the Crestuma reservoir is now a predominantly heterotrophic ecosystem that is the source of water to 20% of the Portuguese population and feeds the Douro estuary with most of the water that reaches the Atlantic Ocean. As a result of this project, a full functional calibrated with dedicated data model is available. During the 72 mo of the project, three young researchers were able to work and get training in the different areas covered by the program. Due to the enormous amount of data gathered, only two research papers were published. Thus, in the near future, two papers will be submitted (as a result of Annex III and Annex IV), and two additional papers are under preparation. Part of the obtained data has being used for teaching purposes at undergraduate and Master/PhD levels by the PI and the Cemas team.