Open Access Repository

Biogeochemical marine ecosystem models II: the effect of physiological detail on model performance


Downloads per month over past year

Fulton, EA, Parslow, JS, Smith, ADM and Johnson, CR 2004 , 'Biogeochemical marine ecosystem models II: the effect of physiological detail on model performance' , Ecological Modelling, vol. 173 , pp. 371-406 , doi: 10.1016/j.ecolmodel.2003.09.024.

[img] PDF
2004_Fulton,_Pa...pdf | Request a copy
Full text restricted
Available under University of Tasmania Standard License.


The level of detail required to efficiently capture system dynamics in ecosystem models has not been well defined. To this
end an ecosystem model of a generalised temperate bay, Bay Model 2 (BM2), was constructed. It is a trophically diverse
biogeochemical model built using the functional groups from another ecosystem model, the Integrated Generic Bay Ecosystem
Model (IGBEM) and the general framework from a model of Port Phillip Bay (PPB), Australia. BM2 captures the essential
features of real marine systems, it is also capable of reproducing realistic levels of biomass and conforms with known ecological
relationships. The model's performance is not as good for some of the poorly known groups (like infauna) or when environmental
conditions undergo extreme change. Despite this, the overall performance of BM2 indicated, it is as capable of representing
systems as accurately as more physiologically detailed ecosystem models, such as IGBEM. This shows that physiological detail
is not always required and that simpler formulations, such as those employed in BM2, are generally adequate for learning and
general predictive purposes. This is important because, in comparison with IGBEM, BM2 uses substantially fewer parameters
and has lower development, computation and maintenance costs.

Item Type: Article
Authors/Creators:Fulton, EA and Parslow, JS and Smith, ADM and Johnson, CR
Keywords: Biogeochemical; Model; Ecosystem; ERSEM; Port Phillip Bay; IGBEM; BM2
Journal or Publication Title: Ecological Modelling
ISSN: 0304-3800
DOI / ID Number: 10.1016/j.ecolmodel.2003.09.024
Item Statistics: View statistics for this item

Actions (login required)

Item Control Page Item Control Page