A statistical model for computing causal relationships to assess changes in a marine environment

INTRODUCTION Large-scale coastal development for land reclamation has been carried out on the Saemangeum coast of the southwestern part of the Korean peninsula by constructing a 33 km-long sea dike since 1991. This type of construction has an effect upon both the coastal morphology and tidal systems. After construction of the dike, changes in both tides and tidal currents have occured. The weakening of the tidal currents subsequently led to abrupt changes in the marine environments, which can be seen in the decrease in the suspended sediment concentration, and in plankton blooms that have occasionally occurred (Lie et al., 2008; Lee et al., 2008). For example, the construction of the Isahaya dike in Japan influenced the tidal system of the Ariake Sea (Tsukamoto and Yanagi, 2002; Unoki, 2003). Changes in the marine environment in response to the disturbance of the natural coastal system are inevitable, and there are causal relationships among these changes. To achieve sustainable development that is accompanied by environmental preservation and the protection of vulnerable environmental features, it is necessary to understand the entire complex process of change, and to be able predict changes in advance in order to prepare appropriate countermeasures or alternatives. In previous research, a number of statistical methods such as time-series analysis and regression analysis were employed (Kim and Park, 2012; Lee et al., 2011). These studies, however, only produced a fragmentary analysis, because they employed univariate analysis or only investigated a one-to-one correspondence between variables. In this paper, we establish a statistical model and perform a multivariate statistical analysis in order to identify the causal relationships of changes in the marine environment, and we make quantitative estimates of the degrees of relational influence, using observed data acquired over the past 10 years. As a result, a valid statistical model was derived that can explain the processes taking place and assess the changes in the marine environment, as well as confirm the theoretical processes involved. Furthermore, we also assessed the difference of changes that have taken place between the inner reclaimed area and the outer sea area, which are geographically separated by the constructed dike. Moreover, by ABSTRACT