Air-water flow is an undesired condition in many systems for the transportation of water or wastewater. Air in storm water tunnels may get trapped and negatively affect the system. Air pockets in hydropower tunnels or sewers may cause blow-back events and inadmissible pressure spikes. Water pipes and wastewater pressure mains in particular are subject to air pocket formation in downward-sloping reaches, such as inverted siphons or terrain slopes. Air pocket accumulation causes energy losses and an associated capacity reduction. Whereas in horizontal and in upward inclined pipes all entrained air is transported with the water flow, the air in downward sloping pipes can move in both directions. Knowledge on air pocket motion in downward sloping pipes is essential for the proper venting of pressurized pipes and for the prevention of severe blow-back events. The motion of air pockets in downward sloping pipes, is closely related to liquid slugs in inclined pipes carrying gas with a small fraction of liquids (i.e. water, oil and gas condensate). The bubble-shaped interface and gas entrainment at the slug front are two features that are similar with air pockets in downward sloping pipes. Existing two-phase flow models have been validated mainly on data in horizontal and vertical pipes in which the gas phase drives the liquid phase. The performance of these models in inclined pipes, in which the liquid phase drives the gas phase, is not yet known. Despite its practical relevance in a variety of engineering fields, the literature on air-water flows in downward sloping pipes is scarce. The fundamental momentum balance that predicts when an elongated air pocket becomes stagnant in a downward inclined pipe, is yet to be developed. Lubbers was the first to systematically investigate the co-current flow of air and water in downward sloping pipes over the complete range of possible air accumulations. Like this thesis, Lubbers’ experimental research was part of the CAPWAT project on capacity losses in pressurised wastewater mains. The main research question, addressed in this thesis, is the development and validation of a total air transport model by flowing water, including the influence of pipe angle, length of sloping section, pipe diameter, surface tension, absolute pressure, pipe friction factor and viscosity. Furthermore, the air discharge by flowing water and the gas pocket head loss in wastewater will be compared with those in clean water. In order to quantify scale effects new measurements have been performed in laboratory facilities with internal pipe diameters of 0.08 m and 0.15 m and in a large-scale facility at a wastewater treatment plant with internal pipe diameter D = 0.192 m, a downward sloping length of L = 40 m (L/D = 209) and a downward pipe angle of 10°. Three series of experiments on co-current air-water flow have been conducted in the large-scale facility, each with its own specific objective in addition to the purpose of model validation: 1 Experiments with clean water, which provided quantitative information on the influence of the length of the downward sloping reach on the air pocket head loss and net air discharge. 2 Experiments with surfactant-added water for the assessment of the influence of surface tension on the air pocket head loss and net air discharge. 3 Experiments with untreated wastewater in order to determine the air pocket head loss and net air discharge in pipelines carrying wastewater. Obviously, these results have been compared with the first experimental series on clean water. The following main conclusions are drawn from this thesis: 1 A physically-based predictive model has been developed for the net air discharge by flowing water in downward sloping pipes. The model parameters include the length of the downward sloping reach and total length of the air pockets, pipe angle, pipe diameter, water (or liquid) discharge, viscosity, surface tension and pipe friction factor. 2 The model has been calibrated to a unique dataset of co-current air-water flows in downward sloping pipes. 3 The composition of wastewater, i.e. lower surface tension and solids content, does not enhance the air transport in comparison with the air transport in clean water. 4 A new velocity criterion for the occurrence of multiple air pockets in a downward sloping reach has been developed. This criterion defines whether the maximum gas pocket head loss may occur in practice. 5 A new momentum balance for elongated air pockets in downward sloping pipes has been developed. This momentum balance defines the clearing flow number. It is useful in practice to predict the direction and velocity of an elongated air pocket in a downward sloping pipe. The momentum balance and velocity criterion support the design of storm water storage tunnels and bottom outlets of hydropower stations for the proper venting of pipes and tunnels and for the prevention of severe blow-back events. Furthermore, two-phase flow models for the prediction of the transition to slug flow and its properties may benefit from these developments. 6 The required water velocity to start the transport of an elongated gas pocket to the bottom of a downward sloping pipe reach is 0.9?(gD)^1/2 (or Fw = 0.9) over a wide range of pipe angles (5° – 20°). This statement has been substantiated with experimental data at D > 0.19 m and the derived momentum balance. 7 A gas pocket detection method for the prediction of a gas pocket location has been extended with a total gas volume prediction. The detection method has been tested successfully in a field experiment.