Multi-approach mapping to help spatial planning and management of the kelp species L. digitata and L. hyperborea: Case study of the Molène Archipelago, Brittany

Abstract

a b s t r a c t a r t i c l e i n f o Keywords: Laminaria Lidar Acoustic imagery Zero-inflated model Habitat mapping Spatial managementThe Molène Archipelago in Brittany (France) hosts one of the largest kelp forests in Europe. Beyond their recognized ecological importance as an essential habitat and food for a variety of marine species, kelp also contributes towards regional economies by means of the alginate industry. Thousands of tons of kelp are collected each year for the needs of the chemical and food industries. Kelp harvesting in Brittany mainly concerns two species, Laminaria digitata (59,000 t) and Laminaria hyperborea (24,000 t), that, together, represent approximately 95% of the national landings. Estimating the available standing stock and its distribution is a clear need for providing appropriate and sustainable management measures. Prior to estimating the spatial distribution of biomasses, we produced a detailed seabed topography map with accurate hard substrate delineation thanks to surveys and appropriate processing of airborne optical and acoustic imaging. Habitat suitability models of presence-absence and biomass were then developed for each species by relating in situ observations from underwater video and sampling to the many biotic and abiotic factors that may govern kelp species distribution. Our statistical approach combining generalized additive models (GAM) in a delta approach also provided spatial uncertainty associated with each prediction to help management decisions. This study confirmed that the adopted strategy, based on an integrated approach, enhanced knowledge on kelp biomass distributions in the Molène Archipelago and provided a promising direct link between research and management. Indeed, the high resolution topography and hard substrate maps produced for the study greatly improved knowledge on the sea bottom of the area. This was also of major importance for an accurate mapping of kelp distribution. The quality of the habitat suitability models was verified with fishing effort data (RECOPESCA program) and confirmed by local managers and kelp harvesters. Based on the biomass maps produced and their associated confidence intervals, we proposed more precise management rules than those already in use for both L. digitata and L. hyperborea. Our mapping approach is a first step towards sustainable kelp species management in the area. Introducing higher resolution environmental variables and population dynamics would help interannual management.

Achieving Reliable Estimates of the Spatial Distribution of Kelp Biomass

“…These traditional methods have only limited predictive power because they perform accurately for the provided dataset but usually provide poor predictions when new datasets are fed into the models. The commonly used covariate for kelp distribution modelling, wave exposure at the sea surface (Bajjouk et al, 2015; i.e., wave exposure at the sea surface; Bekkby et al, 2009;Gorman et al, 2012;Gregr et al, 2018), did not perform as well as bottom wave exposure in predicting the spatial distribution of L. hyperborea in our study. Interestingly, the model containing significant wave height (i.e., wave exposure at the sea surface) appeared to perform quite well in the statistical modelling.…”

“…Two studies from France, however, developed statistical models of the biomass of Laminaria species and subsequently predicted the species' distribution of biomass in space (Gorman et al, 2012;Bajjouk et al, 2015). This also allowed them to directly estimate the standing stock.…”

On the Use of Maps and Models in Conservation and Resource Management (Warning: Results May Vary)

“…Boosted regression trees Supervised Costa et al, 2014;Hewitt et al, 2015 Classification rule with unbiased interaction selection and estimation Supervised Ierodiaconou et al, 2011 Discriminant function analysis Supervised Degraer et al, 2008 Ecological niche factor analysis Supervised Tong et al, 2012;Sánchez-Carnero et al, 2016 Fuzzy k-means Unsupervised Falace et al, 2015 Generalized additive model Supervised Schmiing et al, 2013;Touria et al, 2015 Generalized Quick, unbiased, efficient tree Supervised Ierodiaconou et al, 2011;Hasan et al, 2012 Random forest Both Hasan et al, 2012;Diesing et al, 2014;Piechaud et al, 2015 Support vector machine Supervised Hasan et al, 2012 Frontiers in Marine Science | www.frontiersin.orgFIGURE 2 | Example of how different methods can produce different outcomes. The input data were bathymetric data, backscatter data, and topographic data (i.e., slope, easterness, northerness, and relative deviation from mean value) (see Lecours et al, 2016b).…”

Nepenthes as Tourism Flagship Species: the Conservation Strategies in Dayak Seberuang Settlements Area

“…For example, N. bicalcarata in Dayak Seberuang settlement area only found in peat swamp forest, its mean that the peat swamp forest is a primary target to conserve and maintain the population of N. bicalcarata (Fig.1). Distribution map is a common tools in arrange the conservation strategy and policy to be more effective and accurate on target [27].…”