Modelling the effects of large dams on water quality in tropical rivers

Abstract

River damming is a common way to use river systems to generate hydroelectric power, provide water for irrigation and supply drinking water, and it has been practised for millennia. The number of completed dam projects peaked in North America and Western Europe in the 1960s and 1970s. In the last decades, instead, the hydropower industry moved to build dams in the global south in order to serve growing industries and urban populations. The ongoing growth of the hydropower sector at low latitudes calls for an examination of the political, socio-economic and environmental effects of tropical dams. Despite the many services provided by dams, they affect the river ecosystem in many different ways. Dams disrupt the continuum of rivers by altering their natural hydrological regimes and also create new lentic systems by increasing water residence time. This has cascading effects on the morphology, biogeochemistry and ecology of downstream river environments. Concerning biogeochemistry, dams interrupt the flow of organic carbon, change the nutrient balance and alter oxygen and thermal conditions. Thus, they alter river water quality. Large reservoirs are also potential hotspots for mineralization processes. Thus, reservoirs, especially in the tropics, may be responsible for substantial amounts of greenhouse gas emissions. In recent years, the scientific community started analysing the environmental impacts of large dams in a more holistic manner to better inform stakeholders and decision-makers to find a balance between tapping hydropower potential and sustaining key natural resources. This project investigates the effects of damming on water quality at low latitudes with a specific focus on three water quality parameters: water temperature, dissolved oxygen and carbon dioxide (CO2). Water temperature and dissolved oxygen are key parameters for the survival and reproduction of aquatic species. Water temperature alterations can affect community composition and even trigger the local extinction of species. Low oxygen concentrations alter lifecycle performance, growth capacity, reproductive success and disease vulnerability of fish, whilst hypoxia leads to higher fish mortality. Carbon dioxide, instead, is a greenhouse gas and most of the carbon mineralized in inland waters is released as CO2 to the atmosphere. Major uncertainties remain regarding the consequences of anthropogenic hydrological alterations, especially those stemming from large dams, on carbon emissions. This thesis considers the Zambezi River Basin (southeastern Africa) as a specific case study. The Zambezi River Basin is one of the most dammed African river basins, and many additional dams are already planned or under construction. Among others, the Zambezi River Basin hosts Kariba Dam, which forms the largest artificial lake in the world by volume. Kariba Dam and its hydropower plant are transboundary structures, with management shared between Zambia and Zimbabwe. In general, the transboundary character of water infrastructures complicates the water resources management and thus, serious omissions in the discussion of downstream water quality effects often occur. This specific case study in the Zambezi Basin serves as a starting point to shed some light on the general effects of large dams on water quality at low latitudes. The first study of this thesis presents a global review and synthesis of the effects of river damming on water quality with a special focus on low latitudes. Two physical processes were identified as drivers of most water quality changes: the trapping of sediments and nutrients and the thermal stratification in reservoirs. Analysing the mixing behaviour of the 54 largest low-latitude reservoirs revealed that most, if not all, large low-latitude reservoirs stratify on at least a seasonal basis. Stratification creates density and temperature gradients within the lake water column, facilitating the development of low-oxygen conditions in the deep colder waters. By releasing such water, low-latitude large dams have the potential to impact downstream ecosystems by altering thermal regimes or causing hypoxic stress. The second study of this thesis shows how a detailed statistical analysis of vertical profiles in reservoirs allows generating an assessment tool for water quality alterations downstream of large dams. This finding suggests that designing and maintaining an efficient water quality monitoring of reservoirs is key for their sustainable management. Due to the spatial heterogeneity of water quality in large reservoirs, water quality monitoring should be designed for capturing the temporal dynamics close to outlets of dams in order to predict downstream water quality. In the third study, the alterations of the thermal and oxygen regimes of the Zambezi River downstream of Kariba Dam were quantified by means of a one-dimensional numerical lake model. Results suggest that these alterations depend on the stratification and the water level of the reservoir but also on the management of water withdrawal, thus on the transboundary policies of the dam. Scenarios show that cooperative management of the existing infrastructure of Kariba Dam has the potential to partially mitigate the actual downstream water quality alterations. These results reveal that transboundary dams may offer additional opportunities for optimized management. Moreover, outcomes show that biogeochemical lake models are effective tools to test the effectiveness of such transboundary management scenarios to mitigate downstream water quality alterations. Finally, the last part of the thesis addresses the effects of large dams on the carbon dioxide emission dynamics of inland waters. Monitoring the seasonal and sub-daily fluctuations of water quality properties downstream of Kariba Dam revealed that atmospheric CO2 emissions from the Zambezi River surface downstream of Kariba fluctuate strongly over different timescales. Seasonal changes were driven by reservoir stratification and the accumulation of carbon dioxide in hypolimnetic waters. Sub-daily variability of CO2 emissions, instead, was linked to the hydropeaking resulting from the daily variability in electricity production. Failing to account for these fluctuations in downstream CO2 emissions could lead to errors in the carbon budgeting of hydroelectric reservoirs. Thus, it is critically important to include both limnological seasonality and dam operation at sub-daily time steps in our assessment of carbon budgeting of reservoirs and carbon cycling along the aquatic continuum. This thesis underlines potential environmental drawbacks associated with hydropower, nevertheless recognizing the many benefits of such energy source to societies worldwide. Thus, it aims at inspiring innovative strategies for more sustainable design and management of dams.