Wave exposure plays a major role in shaping the ecological structure of nearshore communities, with different community types able to survive and/or thrive when typically exposed to different levels of wave energy. This can be quantified by taking direct field measurements with wave buoys over time and then manipulating the data to derive typical conditions. However, taking these measurements is only feasible for very limited areas due to logistical constraints, and generating them with numerical wave models can also be expensive to run and may require data inputs that are either lacking or are highly uncertain. Instead, the relative differences in wave exposure between places (relative wave exposure) may be sufficient to distinguish between different community types. It is possible to approximate relative wave exposure using a cartographic approach. Typically this involves measuring the relative shelter or openness of a particular location based on the distances from it to the nearest potential wave blocking obstacle in all directions with provides an approximation of fetch. Given that dominant wind speed and direction data is available for a particular site, these fetch distances can be manipulated to estimate the potential wave climate at that site, with some models going as far as to link this to linear wave theory in order to calculate wave power. This works because the extent to which large waves can form, and to which seas are ‘fully developed’, is constrained by wind velocity, time and fetch. Mapping relative wave exposure in this relatively simple way could be used to predict the spatial distribution of broad categories of ecological community types, especially where this information is difficult to collect using more direct methods.
Despite its relative efficiency and simplicity, running a cartographic-based relative exposure model for more than a local study area quickly becomes computationally intensive, which drives the need to set up the model to run as quickly as possible while minimizing the risk of not detecting potential wave blocking obstacles, and thus underestimating the wave exposure. Yet surprisingly, no studies have tested the sensitivity of the relative wave exposure estimates that these models produce to variation in how key factors, such as the density of points from which fetch distances are measured (point spacing), the angle increment at which the fetch lines are drawn around each point (fetch angle spacing), and the adjustment of fetch line lengths based on bathymetry, are set in the model. This paper presents a preliminary analysis that shows the extent to which estimated relative wave exposure changed when the above model settings were varied for four case study areas within the Great Barrier Reef selected for their characteristic spatial arrangement (number and density) of obstacles. This was done using a new GIS-based generic model for estimating relative wave exposure (GREMO) which integrates many existing techniques into a single modeling platform