Thesis (M.Sc. Biological Sciences)--University of Waikato, 2008. / Title from PDF cover (viewed March 11, 2009) Includes bibliographical references (p. 104-128)

To assess anthropogenic impacts of point-source and diffuse discharges on fish populations of the Waikato River, compare responses to different discharges and identify potential sentinel fish species, we sampled wild populations of brown bullhead catfish (Ameiurus nebulosus, (LeSueur, 1819)), shortfin eel (Anguilla australis Richardson, 1848), and common bully (Gobiomorphus cotidianus McDowall, 1975) in the Waikato River. Sites upstream and downstream of: geothermal; bleached kraft mill effluent (BKME); sewage and thermal point-source discharges were sampled. At each site, the population parameters, relative abundance, age structure and individual indices such as: condition factor; and organ (gonad, liver, and spleen) somatic weight ratios; and number and size of follicles per female were assessed. Indicators of fish residence and in some cases exposure to contaminants in discharges were analyzed. Bile chemistry of brown bullhead and shortfin eel was assayed, liver and muscle metal levels were analyzed for brown bullhead and shortfin eel respectively, and stable isotopes of C and N in common bully were measured. Bile, metal and isotopic signatures gave strong evidence that fish had been resident at sites for some time before sampling. Signatures of bile and metal contaminants showed contamination was localised to discharge areas. Gradients in stable isotopes in common bully showed evidence of changes in water sources and anthropogenic effects along the river. Biochemical variables, hepatic ethoxyresorufin-O-deethylase (EROD) and plasma steroids indicated exposure and response of brown bullhead and shortfin eel to pulp and paper contaminants at the BKME site. Physiological (blood) variables showed fish largely responded in a predictable way to elevated water temperatures at discharge sites at time of sampling, however total haemoglobin of brown bullhead and common bully blood failed to increase at the BKME site despite elevated temperatures and low dissolved oxygen. Growth rates, condition factor, age structure, and gonadosomatic index (GSI) suggest that discharges with significant heat or nutrients benefit brown bullhead despite physiological impairment at the BKME site. Shortfin eel individuals also benefited from heated water discharges. No consistent impacts on common bully health were obvious at individual discharge sites, or cumulatively along the river due to the gradual deterioration in water quality downstream. Common bully individuals also benefited from heat in discharges but lack of juveniles at sites where numerous juvenile brown bullhead were found, suggest that unlike brown bullhead populations, common bully populations were not responding with significant recruitment. Although I found little evidence of toxic effects of discharges on shortfin eel, caution is required in assessing the potential of contaminants to impact eel populations due to the life history of shortfin eel, and exploited nature of populations. For example, reproductive damage suffered by adult eels may not immediately manifest itself in the effected population due to temporal delays in gonadal maturation, and recruitment, and single panmictic populations supplementing recruitment of impacted populations. Distinct changes in population parameters at each of the paired sites and changes in individual variables showed that fish responded to discharges. The range of responses in species suggests different sensitivity to contaminants and amount of benefit which each species receives from heat in discharges. In these terms shortfin eel would be the most resistant, then brown bullhead and lastly common bully. Interpretation of population-level impacts at the geothermal and BKME discharge sites is made difficult due to benefits of additional heat. There is also the possibility that detection of sub-lethal or chronic effects on sensitive juvenile life-stages may be being hidden by compensatory density population responses. Responses and life history of common bully made them the preferred indicator species of the three species sampled, and supported overseas examples using small-bodied fish species as sentinels.

Previous research identified distinct genetic, life-history and reproductive differences between populations of common bully (Gobiomorphus cotidianus) upstream and downstream of a pulp and paper mill outfall on the Tarawera River in the Bay of Plenty, New Zealand. This thesis sought to investigate the distribution of amphidromous and non-amphidromous common bully in the Tarawera River by examining fish collected from upstream (37 km inland) and downstream (20 km inland) locations and comparing them to fish from similar inland locations (40 km and 17 km inland, respectively) in the nearby Rangitaiki River. Otolith microchemistry revealed life-history differences between upstream and downstream populations and stable isotope analysis ensured long-term site residency. Amphidromy dominated in the downstream river populations, while the disappearance of diadromous fish generally occurred with inland distance. A mixture of diadromous and non-diadromous fish were found in the upstream Rangitaiki, while a complete absence of diadromous recruits was found in the upstream Tarawera River. A reduction in oculoscapular canal structures also coincided with loss of diadromy in fish from both rivers. Temporal reproductive divergence was investigated through track annual trends in gonadosomatic index. The Tarawera River receives significant inputs from numerous industrial, municipal and natural sources, most notably from two pulp and paper mills. In the absence physical barriers in the Tarawera, it has been hypothesised that the lack of diadromous recruits in the upstream Tarawera River may be related to aquatic discharges in the downstream river. A behavioural study was performed to examine the hypothesis that pulp and paper mill effluent may be acting as a chemical barrier to fish migration within the river. A dual-choice chamber was employed to examine the responses of common bully exposed to a range of effluent concentrations (100, 50, 25, 12.5, 0% v/v). Fish exhibited significant avoidance responses when exposed to 100 and 50% effluent concentration, while no avoidance was observed at effluent concentrations below 50% This study demonstrated that common bully show a strong preference for river water when simultaneously exposed to effluent, albeit at environmentally unrealistic concentrations (i.e. greater than 15%), implicating potential for this effluent to act as a chemical barrier in the Tarawera River. Following the establishment of reproductive timing of common bully in the Tarawera River, a wild fish health assessment was undertaken to investigate the effects of long-term effluent exposure in situ. Adult common bully were sampled downstream of the mill influence and compared to an appropriate reference population from the downstream Rangitaiki River. Male and female fish from the Tarawera River demonstrated 6- to 9-fold greater ethoxyresorufin-O¬-deethylase (EROD) activity compared to reference fish, indicating exposure to organic contaminants in this river. Tarawera females showed some minor variation in hematological variables including decreased mean cell volume (MCV), mean cell haemoglobin (MCH) and increased total white blood cell count (WBCC) suggestive of an immune response. Slightly greater ovarian follicular steroid production in Tarawera fish potentially indicates some form of endocrine alteration. However, this response may also be related to differences in reproductive synchrony and gonadal development between the two fish populations.