It has been established in New Zealand and internationally that morbidity and mortality tends to rise during colder winter months, with a typical 10-20% excess compared to the rest of the year. This study sought to investigate the spatial, temporal, climatic and demographic patterns and interactions of excess winter morbidity (EWMb) among the elderly in New Zealand. This was achieved through analysis of acute hospital admissions in New Zealand between 1996 and 2013 for all patients over the age of 60 with an element of circulatory or respiratory disease (N=1,704,317) including a primary diagnosis of circulatory (N=166,938) or respiratory (N=62,495) disease. A quantitative approach included ordinary least squares and negative binomial regression, graphical analysis and age standardisation processes. Admission rates and durations were regressed against a set of 16 cold spell indicators at a national and regional scale, finding significant spatial variation in the magnitude of EWMb. EWMb was ubiquitous across New Zealand despite climatic variation between regions, with an average winter excess of 15%, and an excess of 51% for chronic obstructive pulmonary disease (COPD). Statistically significant relationships were found between hospital admission durations and cold spells up to 28 days prior; however the magnitude would not be expected to have a significant impact on hospital resources. Nonetheless, there is potential for preventative public health strategies to mitigate less severe morbidity associated with cold spells. Patients over the age of 80 were particularly vulnerable to EWMb; however socioeconomic deprivation and ethnicity did not affect vulnerability. Patients residing in areas of high socioeconomic deprivation or identifying with Maori or Pacific Island ethnicity experienced significantly shorter admissions than other groups, and this warrants further investigation. Further investigation into winter COPD exacerbations and non-climatic factors associated with the EWMb are recommended. A comprehensive understanding of EWMb will enable preventative measures that can improve quality of life, particularly for the elderly population.

Ongoing climate change is predicted to increase water temperatures and export of terrestrial dissolved matter (TDOM) to aquatic ecosystems influencing ecosystem productivity, food web dynamics and production of top consumers. Ecosystem productivity is mainly determined by the rates of primary production (GPP) in turn controlled by nutrients, light availability and temperature, while temperature alone affect vital rates like consumption and metabolic rates and maintenance requirements of consumers. Increased level of TDOM causes brownification of water which may cause light limitation in algae and decrease GPP and especially so in the benthic habitat. Temperature increase has a been suggested to increase metabolic rates of consumers to larger extent than the corresponding effect on GPP, which suggest reduced top consumer biomass and production with warming. The aim of this thesis was to experimentally study the effects of increased temperature and TDOM on habitat specific and whole ecosystem GPP and fish densities and production. In a replicated large-scale pond experiment encompassing natural food webs of lotic ecosystems I studied population level responses to warming and brownification in the three- spined stickleback (Gasterosteus aculeatus). Results showed overall that warming had no effect on whole ecosystem GPP, likely due to nutrient limitation, while TDOM input decreased benthic GPP but stimulated pelagic GPP. In fish, results first of all suggested that recruitment in sticklebacks over summer was negatively affected by warming as maintenance requirements in relation to GPP increased and thereby increased starvation mortality of young-of-the-year (YOY) sticklebacks. Secondly, brownification increased mortality over winter in YOY as the negative effect on light conditions likely decreased search efficiency and caused lower consumption rates and starvation over winter in sticklebacks. Third, seasonal production of YOY, older, and total stickleback production was negatively affected by warming, while increased TDOM caused decreased YOY and total fish production. The combined effect of the two was intermediate but still negative. Temperature effects on fish production were likely a result of increased energy requirements of fish in relation to resource production and intake rates whereas the negative effect of TDOM likely was a result of decreased benthic resource production. Finally, effects of warming over a three-year period caused total fish density and biomass and abundance of both mature and old fish to decrease, while proportion of young fish increased. The main cause behind the strong negative effects of warming on fish population biomass and changes in population demographic parameters were likely the temperature driven increased energy requirements relative to resource production and cohort competition. The results from this thesis suggest that predicted climate change impacts on lentic aquatic ecosystems will decrease future densities and biomass of fish and negatively affect fish production and especially so in systems dominated by benthic resource production.