|
|
Maufroy, Alexandra; Chassot, Emmanuel; Joo, Rocio; Kaplan, David. |
Since the 1990s, massive use of drifting Fish Aggregating Devices (dFADs) to aggregate tropical tunas has strongly modified global purse-seine fisheries. For the first time, a large data set of GPS positions from buoys deployed by French purse-seiners to monitor dFADs is analysed to provide information on spatio-temporal patterns of dFAD use in the Atlantic and Indian Oceans during 2007-2011. First, we select among four classification methods the model that best separates "at sea" from "on board" buoy positions. A random forest model had the best performance, both in terms of the rate of false "at sea" predictions and the amount of over-segmentation of "at sea" trajectories (i.e., artificial division of trajectories into multiple, shorter pieces due to... |
Tipo: Text |
|
Ano: 2015 |
URL: https://archimer.ifremer.fr/doc/00271/38256/73660.pdf |
| |