Fishing the waves: comparing GAMs and random forest to evaluate the effect of changing marine conditions on the energy performance of vessels
(EN)
BONANOMI, SARA
NOTTI, EMILIO
COLOMBELLI, ALESSANDRO
SALA, ANTONELLO
MORO, FABRIZIO
PULCINELLA, JACOPO
The optimization of consumption and the reduction of gas emissions in fisheries rely on a thorough examination of all factors affecting the energy balance of fishing vessels. Engines, propellers, or the hydrodynamic characteristics of vessels and gears are unquestionably the primary factors affecting this balance, and an improvement in energy efficiency based on these factors is typically attained through technical modifications to existing technologies. Behavioral modifications, such as a reduction in operational speeds or the selection of closer fishing grounds, are another option. There may still be room for improvement in behavioral responses, for instance by adapting fishing strategies in response to changing weather and sea conditions. As far as the authors are aware, the influence of varying sea state and wind conditions on the energy expenditure of fishing vessels has not yet been investigated and is the focus of this research. In this study, wind and wave actions were associated with the observed activity of three fishing vessels operating in the northern Adriatic Sea: an OTB, a PTM, and a TBB trawler. The analysis made use of a comparison between two different approaches, generalized additive models (GAMs) and random forest, in order to quantify the significance of each variable on the response and generate consumption forecasts. In our analysis, the observed influence of predictors was significant albeit occasionally ambiguous. Wave height had the most obvious impact on energy expenditure, with the towing and gear handling phases being the most affected by wave action. Conversely, navigation seemed to be mostly unaffected by significant wave heights up to 1.5 meters, with unclear effects on consumption estimated above this threshold. The relationship between winds and fuel consumption was found to be nonlinear and ambiguous; hence, its significance should be investigated further.
(EN)
