Spatial predictive distribution modelling of the kelp species Laminaria hyperborea

Abstract

Bekkby, T., Rinde, E., Erikstad, L., and Bakkestuen, V. 2009. Spatial predictive distribution modelling of the kelp species Laminaria hyperborea. – ICES Journal of Marine Science, 66: 2106–2115. The kelp species Laminaria hyperborea constitutes highly productive kelp forest systems hosting a broad diversity of species and providing the basis for commercial kelp harvesting and, through its productivity, the fishing industry. Spatial planning and management of this important habitat and resource needs to be based on distribution maps and detailed knowledge of the main factors influencing the distribution. However, in countries with a long and complex coastline, such as Norway, detailed mapping is practically and economically difficult. Consequently, alternative methods are required. Based on modelled and field-measured geophysical variables and presence/absence data of L. hyperborea, a spatial predictive probability model for kelp distribution is developed. The influence of depth, slope, terrain curvature, light exposure, wave exposure, and current speed on the distribution of L. hyperborea are modelled using a generalized additive model. Using the Akaike Information Criterion, we found that the most important geophysical factors explaining the distribution of kelp were depth, terrain curvature, and wave and light exposure. The resulting predictive model was very reliable, showing good ability to predict the presence and absence of kelp.

Importance of wave and current exposure to fauna communities in Laminaria hyperborea kelp forests

“…Data on wind speed and direction were delivered by the Norwegian Meteorological Institute and averaged over a 5 yr period just prior to the study period. The model has been applied to a number of research studies on kelp distribution (Bekkby et al 2009, Bekkby & Moy 2011, diversity ) and secondary production within the study area. The dominating winds in the area come from the south-southwest (195−225°).…”

“…The kelp Laminaria hyperborea dominates rocky sea beds along the Norwegian coastal areas with moderate to high levels of wave exposure (Kain 1967, Bekkby et al 2009. A rich and abundant mobile macrofauna, dominated by crustaceans and gastropods, is associated with kelp, and the largest abundances are found in association with epiphytic algae on kelp stipes (Norderhaug et al 2002).…”

Quantifying wave exposure in shallow temperate reef systems: applicability of fetch models for predicting algal biodiversity

“…However, fetch modelling can also be conducted within a readily available GIS program and is quite flexible. Fetch-based indices can be generated for specific sites of interest (Ruuskanen et al 1999, Bekkby et al 2009), for entire coastlines, or on a customised grid (Kelly et al 2001, Burrows et al 2008.…”

“…As fetch models are designed to estimate local conditions, they do not specifically model waves due to swell that may originate from offshore systems. They also do not account for nearshore processes such as the refraction, diffraction and breaking of waves.Despite these limitations, fetch models have been successfully applied within an ecological modelling 84 Hill et al: Quantifying wave exposure on shallow reefs context to describe and predict relationships between exposure and rocky shore communities (Burrows et al 2008), seagrass beds (Fonseca & Bell 1998), algal occurrence (Ruuskanen et al 1999) and algal morphology (Bekkby et al 2009). As may be expected, most fetch-based studies have been conducted in semienclosed waters (where locally-generated wind waves dominate) on local to regional scales (but see Burrows et al 2008).…”

“…However, fetch modelling can also be conducted within a readily available GIS program and is quite flexible. Fetch-based indices can be generated for specific sites of interest (Ruuskanen et al 1999, Bekkby et al 2009), for entire coastlines, or on a customised grid (Kelly et al 2001, Burrows et al 2008.In terms of predicting algal biodiversity, the performance of the different exposure indices can be summarised as follows. Indices derived from an offshore wave hindcast model were the best predictor of algal patterns in only a few instances.…”