direkt zum Inhalt springen

direkt zum Hauptnavigationsmenü

Sie sind hier

TU Berlin

Inhalt des Dokuments

INNOVATE Research Module 5.4: High resolution multi-dimensional modelling for hydrodynamics and water quality in Luiz Gonzaga reservoir

Project head
Prof. Dr.-Ing. R. Hinkelmann [1]
Research associate(s)
E. Matta, M. Sc. [2]
Scientific assistant(s)
Richard Lorenz
Project period
April 2012 - December 2016
Federal Ministry of Education and Research / Research for Sustainability / Sustainable Land Management

Project partner
University of Hohenheim, Leibniz Institute of Freshwater Ecology and Inland Fisheries, Potsdam Institute for Climate Impact Research, University of Applied Sciences Dresden,  Senckenberg Natural History Collections Dresden, Universidade Federal de Pernambuco, Universidade Federal Rural de Pernambuco, Agricultural Institute of Pernambuco state, Technology Institute of Pernambuco, soil section of the Brazilian Agricultural Research Corporation and the National Institute of the Semi-Arid Region
$this->_build_link_list($this->linkCount++, "http://www.innovate.tu-berlin.de", "www.innovate.tu-berlin.de [3]")


The Luiz Gonzaga dam (former Itaparica dam) was built between 1979 and 1988 to dam up the São Francisco river in Northeast Brasil. The resulting Itaparica reservoir is facing several environmental and socio-economical problems.

The INNOVATE (Interplay among multiple uses of water reservoirs via innovative coupling of substance cycles in aquatic and terrestrial ecosystems) project aims at an interdisciplinary approach to develop a sustainable watershed management. The sub-project 5: Modelling focuses on the development of models of the basin of São Francisco to study the impact of climate and land use changes on the water resources.

Research Module 5.4, which is part of sub-project 5, investigates the current situation in the reservoir and the impacts of the above mentioned changes on the water quantity and quality, using two- and three-dimensional high resolution hydro-numerical models. Based on the results, critical conditions can be identified and adaption measures will be developed.

Location of different models
Lupe [4]

Project structure

Constellation and collaborations within INNOVATE
Lupe [5]

Large scale 2D flow and transport model of the Itaparica reservoir with the Hydroinformatics Modelling System

A large scale two-dimensional model for hydrodynamics and tracer transport was set up in the framework of HMS. First simulation runs were carried out on a simplified topography. The reservoir was modeled as a rectangular channel with a constant inclination from the inflow to the outflow.

Results show that for mean flow conditions, there are so-called dead water zones in the reservoir where the water is almost standing. These areas are not influenced by the main stream in the São Francisco river. This is an important insight in the behaviour of the reservoir in regard to future investigations concerning water quality and contamination processes.

Flow velocities for mean flow conditions in Itaparica reservoir
Lupe [6]

Future research will take into account a more detailed topography of the reservoir. Additionally, wind-induced flow and tracer transport will be investigated. Further, boundary conditions provided by the hydrological model SWIM of Potsdam Institute for Climate Impact Research will be used.

The results of the simulations of the large scale model will be used to drive the small scale two-dimensional model of Icó-Mandantes bay.

Small scale 2D flow and transport model of the Icó-Mandantes bay with the Hydroinformatic Modelling System

Icó-Mandantes bay is one research area of the INNOVATE project. The bay faces a serious eutrophication problem caused by the intensive agriculture around the bay as well as the waste water inflow from surrounding cities.

A two-dimensional model for hydrodynamics and transport was set up in the framework of Hydroinformatics Modelling System (HMS), a modeling framework developed at the chair. Aim of the first research phase was to simulate the hydraulic interaction between the bay and the São Francisco river, to get an understanding of the whole water system. The results show that the interaction between bay and river is negligibly small.

Flow velocity field in Icó Mandantes bay
Lupe [7]

Successively, wind-induced flow and transport in Icò-Mandantes bay were simulated, to analyze the effects of the wind on the water body and on the interaction between the São Francisco river and the bay, for different scenarios and different wind conditions.  The velocities in the bay remained relatively small (max. cm/s), but they were significantly increased by the wind shear stress and no longer stagnant. Additionally, a tracer was injected in the area to simulate the spreading of a contaminant. An increase of tracer exchange between the river and the bay was observed.

Surface flow simulation for mean flow conditions and mean wind
Lupe [8]
Transport simulation for mean flow conditions and extreme wind after ca. 1 day
Lupe [9]

Large scale 2D flow and transport model of the Itaparica reservoir with Telemac

A new large scale 2D model of the Itaparica reservoir was built, using bathymetry field data received by the SP1-1, assuming suitable and more complicated and realistical channel profiles and interpolating between these profiles. 

Shallow water flow and transport was simulated in the mentioned model. Different scenarios were investigated, as the influence of the roughness of the soil and the turbulent diffusion, for mean flow conditions as well as for a drought and a flood case, using the Telemac system. The simulations indicated that the stream in the main river dominates the flow field for mean flow as well as for the drought, while in the bays the water was almost stagnant and a tracer injected at the inflow boundary was mainly transported downstream without entering the bays. For the flood case, the flow velocities considerably increased in the whole reservoir. Therefore, a tracer injected at the inflow boundary entered parts of bays. When a tracer was injected punctually inside Icò-Mandantes bay, the spreading was extremely slow. After ca. 200 days of simulation the tracer was still spreding inside the bay: the concentration of the tracer was lower than 1 % of the initial value.

Large scale 2D model of the Itaparica reservoir: bottom elevation
Lupe [10]
Transport simulation on the large scale 2D model of the Itaparica reservoir: punctual injection inside Icò Mandantes bay, at t = 200 days
Lupe [11]

Small scale 2D model of the Icò-Mandantes bay in Telemac

New bathymetry data of Icò-Mandantes bay were collected from SP1-1 during the last field trip in Brazil (December 2013) and received in January 2014. A new small scale 2D model of Icò-Mandantes bay was built, to refine the previous bathymetry of the bay. The model was tested and validated using Telemac 2D. SP1-1 provided also data about uptakes and tributaries along the bay, some hydrological data, wind data and nutrient concentrations in the discharge, as N, P, Chl.a.

Low and high water level scenarios are planned to be investigated, considering different inflows and outflows, as tributaries, uptakes, wind effects, impact of the vegetation, influence of nitrogen, phosphorus and macrophytes. Successively, the 3D model of the Icò Mandantes bay will be set up also with Telemac, extending the developed model to observe three dimensional effects. The model will be validated, simulating different scenarios, considering heat exchange with atmosphere, density-induced flow, wind effects, day-nights cycles, vegetation and the existing tributaries and uptakes. Also the set up of a water quality model is planned.

The models implemented could help the other sub-project partners of the INNOVATE (e.g. SP1-1, SP5-1, SP5-2), the stakeholders (e.g. CHESF or ANA), the Brazilian partners and others to better understand the behaviour of Icó- Mandantes bay and the interaction with the São Francisco river. In future work climate and land use change as well as stakeholder-defined scenarios will be investigated on the 3D and water quality models.

Small scale 2D model of the Icò-Mandantes bay: bottom elevation (m)
Lupe [12]

Summary of main results and management guidelines

Studying various project partners- and stakeholder-oriented scenarios in Icó-Mandantes bay and Itaparica reservoir, we gained a wider knowledge about the systems and their interactions. Water is used for different purposes and at the same time is affected by water quality problems, due to e.g. high water level variation (HPP, evaporation), not adequate waste water system. In such bays, water exchange with the main stream of Itaparica hardly occurs and, therefore, we incur in higher risks of withdrawing low quality water or contribute to eutrophication in the bay, if nutrients increase by net cages emissions or by flash floods from the intermittent tributary Riacho dos Mandantes.

It is extremely important to adapt the management to the climate change scenarios and anthropogenic uses. Indeed, considering a flood event from the small effluent Riacho dos Mandantes due to heavy rain, we saw how it influenced water quantity mainly in the bay itself, in particular in the near field of the river inflow, and how the hypothetical material transported interacts with the different uses. The intakes for irrigation agriculture and Transposition project, which are located in the near field of the tributary, will be negatively affected by it (on the longer term assuming a wet scenario). Residence times are in fact much higher (> 1 year) in Icó-Mandantes bay, compared to the ones calculated for Itaparica reservoir (ca. 2 months).

Regarding aquaculture, we hypothesized to install a new net cage system producing 130 t y-1 of Tilapia inside the bay and observed the consequent increase of total nitrogen and phosphorus, compared to the current mean values and maximum concentrations allowed (SP1-1; CONAMA 357/05). The results showed that nutrient emissions due to a small production would lead the bay to have an increase of phosphorus concentration of ca. 8 µg L-1 after one week of simulation under drought conditions. Therefore, it should be avoided to install new aquaculture systems in such bays and, otherwise, to ensure enough water column beneath the cages (min. 10 m, SP1-1).  

In our studies, we could also observe that the eastern channel of the Transposition project influenced not relevantly water quantity at the local scale of our models, except for the increase of flow velocities in the near-field of the intake, but water quality control measures need to be applied, because of the critical location of the withdrawal. Nevertheless, the simulations are based on the assumed withdrawals declared by ANA (411/2005), since the axis is not yet operating.

Finally, we focused in these years on the important role of the wind in our research area, studying wind-induced flow in 2D and 3D, considering various scenarios. Wind is one of the main drivers of hydrodynamics and transport and 3D analyses allow us to observe the consequent circulation patterns along the vertical direction (z) and in this way increase our knowledge of this complex system. The final simulations of the upcoming year (2016-2017) deal with 3D modeling in regards of stratification in the bay, due to temperature gradients.

Study area and computational domain: Itaparica reservoir (left); computational domain of Icó-Mandantes bay, where multiple uses of water are shown (right)
Lupe [13]
copyright XXXXX
Velocity field and tracer concentrations, one week after a 3-days-flood-event from the effluent Riacho dos Mandantes, for a dry (left) and a wet scenario (right)
Lupe [14]
Mass-conservative passive tracer set as initial conditions, with an initial concentration equal to 10 [-], after 6 months-simulations for a dry (left) and a wet scenario (right)
Lupe [15]
An example of 3D results, analyzing mean wind and mean flow conditions in the domain, before a wind storm event. Zoom of the flow field on the surface layer in an intermediate part between the bay and the reservoir main stream (left); vertical section of flow velocities (right)
Lupe [16]

Associated theses


  • Simshäuser, M. N. Impacts of a tributary on flow and transport in Icó Mandantes Bay. Bachelorarbeit im Studiengang Bauingenieurwesen

  • Wirth, S. Auswirkungen eines Wasserüberleitungskanals auf die Strömungs- und Wasserqualitätsverhältnisse in der Icó-Mandantes Bucht. Bachelorarbeit im Studiengang Bauingenieurwesen


  • Broecker, T. Simulation of two-dimensional flow and transport processes in a Brazilian reservoir. Master thesis (Civil engineering).


  • Matta, E. Simulation of wind-induced flow in Icó Mandanted bay, Brazil. Master thesis (Civil engineering).
  • Nasser, S. Numerische Simulation der Strömungsprozesse im Reservoir des Luiz Gonzaga Staudammes, Brasilien. Bachelor thesis (Civil engineering).
  • Storck, T. Erstellung und Optimierung eines zweidimensionalen Gitternetzes für das Itaparica Reservoir, Brasilien. Student research project (Civil engineering).


<< previous [17]

Matta, E., Özgen, I., Cabral, J., Candeias, A.L. & Hinkelmann, R. (2014). Simulation of Wind-Induced Flow and Transport in a Brazilian Bay [22].

Matta,E., Broecker,T., Özgen,I. & Hinkelmann,R. (2014). SP5-4: High resolution reservoir model. [23].

<< previous [24]
------ Links: ------

Zusatzinformationen / Extras

Quick Access:

Schnellnavigation zur Seite über Nummerneingabe

Auxiliary Functions

Copyright TU Berlin 2008