Abstract

Increasing energy demand driven by rapid population and economic growth, the need for climate change mitigation, and the depletion of fossil fuels is stimulating the search for renewable, climate neutral energy sources. Hydropower provides an efficient, low maintenance and flexible form of energy, which can provide ancillary benefits such as flood control, water storage and job creation. Yet, the construction of dams for hydropower production has been recognised by scientists as one of the major threats to the ecological integrity of river systems. For instance, the fragmentation of river systems alters the flow, thermal, and sediment regimes of rivers, and restricts the free movement of aquatic organisms. Disruption to the natural flow regime results in the degradation of physical habitat features which generate acoustic stimuli that are relevant to organisms. In addition, initial flooding of terrestrial habitats results in the rapid decay of organic matter, which releases greenhouse gases (GHG) into the atmosphere. Conservation and management of river systems therefore requires a greater understanding of the processes and mechanisms which underpin the ecohydrological impacts of hydropower. In this context, this doctoral thesis aims to investigate: (i), the ramifications of a global boom in hydropower construction, (ii) the prediction of GHG emissions from hydropower reservoirs, and (iii) the temporal and spatial changes in underwater river soundscapes affected by hydropower. Researchers have investigated the social, economic and ecological consequences of reservoir construction for decades. However, the lack of coordinated, georeferenced databases has hindered catchment decision making, and limited the development of regional and global research in particular. In Chapter 1, the primary objectives were to create a high resolution, georeferenced database of hydropower dams under construction or planned to assess the dimension and spatial distribution of hydropower developments, their density relative to available catchment water resources and the future impact on river fragmentation. Data were collected on hydropower schemes under construction or planned with a capacity of 1 MW or above, from government and non-government databases, grey literature and news reports. Spatial analyses were conducted in a geographical information system (GIS) on the extent of global development, impact per water availability and potential consequences for existing status of river fragmentation. The relative contribution of hydropower reservoirs to the global GHG budget, particularly in sub-tropical and tropical regions, remains the subject of intense critical debate. The initial objective of the second study was therefore, to identify principal parameters and underlying processes that drive GHG emissions from reservoirs. The second step was to review global reservoir emission measurements and their source pathways in hydropower systems. Meteorological and landscape derived parameters were then correlated with the GHG measurements in order to assess if and which selected parameters might explain variations in GHG emission data. Similarly, existing empirical models were applied to the measured data to assess their suitability in predictive modelling. Finally, a newly developed process based model (FAQ-DNDC v1.0) was used to simulate ‘net’ CO2 emissions from a newly flooded tropical reservoir and compared to the measured results. The final study (Chapter 3) examined the influence of hydropower systems on the underwater acoustic properties of river habitats. Using recently developed acoustic sensors in addition to traditional hydrophones, the study characterised the temporal and spatial changes in river soundscapes affected by hydropeaking, compared their frequency composition to unaffected river soundscapes, and critically appraised the ecological implications. The results of Chapter 1 indicate that we are now experiencing an unprecedented growth in global hydropower construction. Over 3,700 dams are planned or under construction, primarily in Africa, South America and East Asia. The expansion in dam building will reduce the number of free flowing rivers on a global scale by approximately 21%. The results of Chapter 2 show that variation in measured emissions due to the inherent heterogeneity of the underlying processes, in addition to methodological limitations, impede the prediction of GHG emissions. Source pathways of CO2 are similar for the majority of systems, however, pathways of CH4 emissions are highly variable and dependent on local operating conditions and the configuration of the given hydropower system. A newly developed process based model (FAQ-DNDC v1.0) shows that a mechanistic approach may provide the basis for the ‘net’ assessment of future hydropower reservoirs. Chapter 3 reveals that distinct river soundscapes undergo changes which are highly correlated to hydropower operations, and thus rapid sub-daily changes occur at timescales not often found in natural systems. These changes occur mostly in low frequency bands, which are within the range of highest acoustic sensitivity for fish. In pool habitats affected by hydropeaking, sound pressure levels in the lower frequencies (~0.0315 kHz) may increase by up to 30 decibels. Similarly, sound pressure levels of riffles increase by up to 16 decibels in the low to mid frequencies (~0.250 kHz). Overall, the findings of this thesis have a number of implications for river catchment management. Hydropower construction is taking place in some of the most ecologically sensitive areas of the globe, thus, this research provides a timely contribution to: (i) Provide a foundation for future research at catchment, regional and global scales. For instance, systematic conservation based planning is required to designate ‘no go’ areas to promote the long-term survival of biodiversity. Strategic positioning of future dams or reconfiguration of existing hydropower systems may reduce the combined impacts on biodiversity and GHG emissions without losing power capacity. (ii) Assess driving parameters of GHG emissions, critically appraise current predictive GHG emission models and use a process based approach to simulate ‘net’ emissions from a sub-tropical reservoir. Future reservoirs will sequester, mineralise and emit an increasing quantity of carbon to the atmosphere, and subsequently, will take a greater role in the global GHG budget. This research concludes that, in some cases empirical models may not be suitable for making robust estimations of future GHG’s from hydropower reservoirs. Combining the underlying carbon cycling processes within a process-based model allows the estimation of ‘net’ CO2 emissions from hydropower reservoirs. This approach may be integrated by catchment planners into the future lifecycle assessment of hydropower reservoirs. (iii) Characterise acoustic changes in underwater sound in rivers affected by hydropeaking. The findings emphasise that flow regulation by hydropower results in rapid changes to the amplitude and frequency spectrum of the riverine acoustic environment. These changes persist for longer periods than other forms of anthropogenic sound and may have implications for the whole biota. Thus, future studies should focus on measuring the behavioural and physiological impact on riverine organisms in order to develop guidelines for hydropower licensing.