This thesis examines the dynamics of estuarine macroalgae blooms of the genus Ulva and the environmental factors controlling them, using process-based mathematical models as well as observation data from Tauranga Harbour (North Island, New Zealand). Based on long-term monitoring datasets of water column nutrients, temperature and irradiance, the Ulva tissue nitrogen concentration is modelled using an algebraic short-term equilibrium solution to a classic set of ordinary differential equations describing tissue nutrient dynamics. A detailed sensitivity analysis of this equilibrium solution shows that although the variability in ambient concentration of inorganic nitrogen explains 60% of model output variance, the uncertainty in two of the most influential physiological parameters has a similar magnitude of influence (32% of variance). Repeated calibrations using an algorithm with random starting points and evolutionary adaptions lead to broad and in some cases multimodal distributions of physiological parameters. Although the model performs well in reproducing observations, calibrated parameter values from individual calibrations should therefore not be interpreted as reliable estimates of physiological properties. Using a zero-dimensional simulation model of Ulva tissue nitrogen and phosphorus concentrations and biomass, seasonal and long-term population dynamics of Ulva are examined. Calibrating against a combination of all three state variables, long-term tissue nitrogen and biomass dynamics are reproduced well, while the seasonal amplitude of tissue phosphorus variability is underestimated. From the long-term observation datasets, seasonal scenarios are derived based on annual cosine fits to monthly percentiles of the observed forcing data. These scenarios represent above and below average seasons in the environmental conditions of ammonia, nitrate and phosphate concentrations in the water as well as temperature and irradiance. Comparison of the different scenarios points to the inorganic nitrogen concentrations (both ammonia and nitrate) as most influential in determining Ulva seasonal peak biomass and timing. To improve upon the existing long-term observations of water column nutrients available for Tauranga Harbour, a sampling programme was designed taking into account the challenges of accurately measuring nutrients in a tidally dominated estuary with complex geometry. Samples were taken at consecutive ebb and flood tides at the same local tidal phase at sites representing distinct subregions within the estuary. Bimonthly samples at eight sites and fortnightly samples at two sites over one year show that the statistical distributions of nutrient fractions vary between sites and between tides. At the estuary mouth, phosphate is exported, while at site further upstream, ammonia and nitrate are exported. Modelling the passive transport of nutrients using a numerical hydrodynamical model shows differences in the connectivity between subregions, and in some cases mid-term upstream transport. The contribution of upstream subregions to individual sites changes rapidly during the tidal cycle, emphasising the importance of the local tidal phase. The data collected may help to improve the sampling design of the long-term monitoring programmes in Tauranga Harbour and provide a more accurate basis for future modelling studies of Ulva population dynamics.