University Of Tasmania
2004_Fulton,_Smith_&_Johnson_Ecol_Mod_B.pdf (284.44 kB)

Effects of spatial resolution on the performance and interpretation of marine ecosystem models

Download (284.44 kB)
journal contribution
posted on 2023-05-16, 15:23 authored by Elizabeth FultonElizabeth Fulton, Smith, ADM, Craig JohnsonCraig Johnson
Simplifying models by using coarse spatial resolution can be desirable because it reduces structural, computational and data requirements and can make model interpretation easier. However, spatially simplified models may be incapable of reproducing important dynamics observed in nature. To consider this issue the effect of spatial structure on the output of two trophic ecosystem models (Bay Model 2 (BM2) and the Integrated Generic Bay Ecosystem Model (IGBEM)) was considered using a theoretical approach known as 'deep-shallow model' comparison. This involved comparing simulation runs of 1-, 3-, and 8-box versions of the ecosystem models (the 'shallow' models) with a 59-box version that was used to represent the real world (the 'deep' model). The results indicate that simpler spatial configurations (geometries) can result in less short-term variation, changes in predicted spatial patterns and trophic self-simplification (loss of functional groups), as the opportunity for spatial refuges is reduced. More importantly, models with very little spatial resolution (i.e. 1- and 3-box models) do not capture the effects of changes in nutrient loads or fishing pressure as well as more complex models. The results for the 8-box models used here indicate that some simplification is acceptable, as overall model performance is not overwhelmed by the impacts of trophic self-simplification and a loss of spatial heterogeneity. However, using models with very little spatial resolution (i.e. 1- and 3-box models) can be misleading, as the impacts on system dynamics of the reduced heterogeneity increase. © 2003 Elsevier B.V. All rights reserved.


Publication title

Ecological Modelling










School of Natural Sciences


Elsevier Science BV

Place of publication


Repository Status

  • Restricted

Socio-economic Objectives

Natural hazards not elsewhere classified

Usage metrics

    University Of Tasmania