Studying coral reef patterns in UAE waters using panel data analysis and multinomial logit and probit models

摘要

Like coral reefs around the world, the reefs of the United Arab Emirates (UAE) are facing global climate change and associated threats. The coasts and islands that flank Abu Dhabi host an important number of corals that should be the focus of conservation actions. Well-designed conservation and management plans require efficient monitoring systems that include understanding coral reef patterns. To understand some of these patterns; coral cover data, satellite-derived and in-situ water quality parameters from nine key reef environments in the UAE from 2011 to 2014 to model coral patterns were used. The objectives were to model coral patterns and realistically predict coral damage intensity with changing environmental variables. Coral damage cover models were defined and estimated for the coral damage cover. Effects of environmental factors were estimated, and predictions of coral damage intensity were presented with changing factors. Main findings, based on the studied data, showed that nutrient enrichment, a proxy for anthropogenic pressure, and salinity are the most influential factors to induce coral damage in UAE waters. Furthermore, results demonstrated that the probability of severe damage increases with decreasing water oxygenation and with increasing temperature, light, salinity, acidity and nutrient levels. The defined and estimated predictions accounted for corals' behavioural aspects, across individual reefs and over time. This approach is more appropriate than estimation predictions that just account for historic trends. Nevertheless, there are, probably, many components within the model framework that can be expanded and/or improved as more information become available. An extended dataset will enable a means to independently validate the defined models and test other modelling approaches. Continually increasing the insitu and remote sensing data sizes, spatially and temporally, defines a long-term priority.