The salinity distribution in an inverse estuary (where the sea water is concentrated by the estuary water) possesses its own uniqueness due to excessive net freshwater loss and restricted circulation. The study of long-term balance of salinity can contribute to a better understanding of the mixing and transport properties in such a distinct type of water body (i.e., inverse estuary water), and can provide valuable information for sound water management and environmental assessment. The Persian Gulf is chosen to be the study region, as it is a typical large-scale inverse estuary with severe shortages of freshwater resources and has been of significant research interest during the past several decades. For basin-wide examinations of a large-scale inverse estuary, analytical solutions are typically unavailable and field measurements are expensive, so numerical modeling as well as validation with available data is the main focus in this thesis. Firstly, the salinity distribution in the Persian Gulf is simulated with 8 different schemes, and the obtained results are compared with the World Ocean Atlas 2013 (WOA13) data. The comparisons can validate the utilization of the numerical model in predicting the salinity distribution in a large-scale inverse estuary. Given that results are affected by the choice of the numerical scheme, a performance analysis of candidate schemes is performed. The most appropriate scheme for the Persian Gulf is figured out in this stage. Secondly, the validated scheme is used for the prediction with respect to the long-term salinity response of the Persian Gulf to the climate change and anthropogenic activities. The results show that without mitigation measures taken, the salinity in the Persian Gulf will continually increase with time. The long-term and basin-wide simulations that will be presented in this thesis are expected to be more useful than previous studies (which were generally limited in time from hours to a few months) in terms of inspecting long-term characteristics. The performance of various numerical schemes has been assessed for the first time through a practical case study, which can contribute to a better understanding of the applications and characteristics of these schemes. Besides, the long-term salinity variations in the Persian Gulf are predicted for the next half-century, and this is the first numerical prediction of the long-term salinity response of the Persian Gulf to climate changes and anthropogenic activities.

Osmoregulation is an essential process to maintain water and ionic balance and when euryhaline fish move between freshwater and seawater environments as part of their life cycle this presents additional osmoregulatory challenges. Migrating fish can be exposed in both environments to pollutants such as endocrine disrupting chemicals (EDCs) that include natural hormones (e.g. 17β-estradiol; E2), synthetic hormones (e.g. 17α-ethinylestradiol; EE2), and industrial chemicals (e.g. nonylphenol). The focus of this thesis was to study the effects of different categories of EDCs on the osmoregulatory functions of euryhaline fish such as three-spined sticklebacks (Gasterosteus aculeatus) and rainbow trout (Oncorhynchus mykiss). Osmoregulatory variables (such as osmolality, water and ionic content) were compared in plasma and tissues (white muscle and carcass) of rainbow trout. This validated the use of specific tissue parameters as a surrogate of plasma responses to various osmoregulatory challenges. Waterborne exposure to 17α-ethinylestradiol revealed differential sensitivity of vitellogenesis in the three-spined sticklebacks (no induction) and rainbow trout, but had a significant effect on calcium homeostasis in both species. Intraperitoneal implants of 17β-estradiol reduced CaCO3 production and apparent water absorption in the intestine and increased in tissue calcium stores of seawater-acclimated trout, but fish were able to compensate and showed no overall osmoregulatory disturbance. Waterborne exposure to nonylphenol in freshwater trout was also investigated, but no effects on osmoregulation were found up to 2 ng/l. Overall, estrogens can affect osmoregulation differentially in euryhaline fish species, and sometimes at EDC levels lower than the threshold for reproductive effects (i.e. vitellogenin induction).