<p>Coral diseases are a major threat to coral reef health and functioning worldwide. Little is known about how coral disease prevalence relates to multiple interacting changes in host densities, abiotic stressors, and levels of human impact. In particular, almost nothing is known about coral disease dynamics under changing abiotic conditions in the absence of direct anthropogenic stressors. Understanding how disease dynamics change relative to shifts in environmental conditions is crucial for the successful management and future survival of coral reefs. With the use of existing and novel field data and statistical modeling I examined the associations (abiotic and biotic) of multiple coral disease states across a variety of spatial scales encompassing a wide range of environmental conditions. Biomedical techniques were then used to relate these environmental associations to potential disease etiology. Study sites included areas with high levels of anthropogenic impact (e.g. Oahu, main Hawaiian Islands); to extremely remote quasi-pristine reefs removed from direct human influence (e.g. Palmyra Atoll National Wildlife Refuge). Over small spatial scales (100s m) at a marine reserve in the main Hawaiian Islands I modelled the spatial patterns of four coral diseases (Porites growth anomalies, Porites tissue loss, Porites trematodiasis and Montipora white syndrome). While Porites tissue loss and Montipora white syndrome were positively associated with poor environmental conditions (poor water quality, low coral cover), Porites growth anomalies and Porites trematodiasis were more prevalent in areas considered to be of superior quality (clearer water, increased host abundance, higher numbers of fish). At Palmyra Atoll, fatal tissue loss diseases were largely absent and although coral growth anomalies were present their prevalence was extremely low. Patterns of growth anomaly prevalence at Palmyra were positively associated with host abundance across four coral genera (Acropora, Astreopora, Montipora and Porites) and generally negatively associated with algal cover. Growth anomalies, although progressive and detrimental to the hosts, were most prevalent in the "healthiest" regions (the highest coral cover regions) of Palmyra. I hypothesised that differences seen in the types and prevalence of coral diseases between heavily populated parts of Hawaii and remote uninhabited locations such as Palmyra Atoll, could be a result of differing levels of either direct (e.g. pollution) or indirect (e.g. pollution leading to loss of key hosts) human stressors, in addition to natural changes in the environment. To begin disentangling the confounding effects of natural variability and human stressors on coral disease prevalence patterns I modelled two diseases (Acropora and Porites growth anomalies) across hundreds of sites throughout the Indo-Pacific Ocean (1000s km). Predictors included host densities, human population numbers, frequency of sea surface temperature anomalies, and input of ultra-violet radiation. Porites growth anomaly prevalence was positively associated with human population density (and to a lesser extent host density), while the prevalence of Acropora growth anomalies was strongly host density dependent. The positive association between the prevalence of Porites growth anomalies and human density suggests the presence and prevalence of the disease are related, directly or indirectly, to some environmental co-factor associated with increased human density at regional spatial scales. Although this association has been widely posited, this is one of the first wide scale studies unambiguously linking a coral disease with human population size. In summary, the types of coral diseases observed, their prevalence, and spatial patterns of distribution within reef systems are the result of multiple abiotic and biotic factors and stressors interacting, in some cases synergistically. Statistical modelling, in conjunction with biomedical techniques and field observations, proved essential to the understanding of coral disease ecology within single reefs and atolls to patterns across entire oceans.</p>
Case-control design identifies ecological drivers of endemic coral diseases