FN Archimer Export Format
PT J
TI Influence of oceanic conditions in the energy transfer efficiency estimation of a micronekton model
BT 
AF Delpech, Audrey
   Conchon, Anna
   Titaud, Olivier
   Lehodey, Patrick
AS 1:1,2;2:2,3;3:2;4:2;
FF 1:;2:;3:;4:;
C1 Laboratoire d’Etudes Géophysiques et d’Océanographie Spatiale, LEGOS - UMR 5566 CNRS/CNES/IRD/UPS, Toulouse, France
   Collecte Localisation Satellite, CLS, Toulouse, France
   Mercator Ocean, Toulouse, France
C2 LEGOS, FRANCE
   CLS, FRANCE
   MERCATOR OCEAN, FRANCE
TC 0
UR https://archimer.ifremer.fr/doc/00603/71463/69917.pdf
   https://archimer.ifremer.fr/doc/00603/71463/70965.pdf
LA English
DT Article
CR PIRATA
AB Micronekton – small marine pelagic organisms mostly in the size range 1–10 cm – is a key component of the ocean ecosystem, as it constitutes the main source of forage for all larger predators. Moreover, the mesopelagic component of micronekton that undergoes Diel Vertical Migration (DVM) likely plays a key role in the transfer and storage of CO2 in the deep ocean: the so-called ‘biological pump’ mechanism. SEAPODYM-MTL is a spatially explicit dynamical model of micronekton. It simulates six functional groups of migrant and non-migrant micronekton, in the epipelagic and mesopelagic layers. Coefficients of energy transfer efficiency between primary production and each group are unknown. But they are essential as they control the predicted biomass. Since these coefficients are not directly measurable, a data assimilation method is used to estimate them. In this study, Observing System Simulation Experiments (OSSE) in the framework of twin experiments are used to test various observation networks at a global scale regarding energy transfer coefficients estimation. Observational networks show a variety of performances. It appears that environmental conditions are crucial to determine network efficiency. According to our study, ideal sampling areas are warm, non-dynamic and productive waters like the eastern side of tropical Oceans. These regions are found to reduce the error of estimated coefficients by 20 % compared to cold and dynamic sampling regions. The results are discussed in term of interactions between physical and biological processes. 
PY 2020
SO Biogeosciences
SN 1726-4189
PU Copernicus GmbH
VL 17
IS 4
BP 833
EP 850
DI 10.5194/bg-2019-353
ID 71463
ER

EF