{"id":403,"date":"2023-08-09T15:56:59","date_gmt":"2023-08-09T19:56:59","guid":{"rendered":"https:\/\/pressbooks.bccampus.ca\/ewemodel\/chapter\/defining-the-ecosystem\/"},"modified":"2025-10-30T08:56:15","modified_gmt":"2025-10-30T12:56:15","slug":"defining-the-ecosystem","status":"publish","type":"chapter","link":"https:\/\/pressbooks.bccampus.ca\/ewemodel\/chapter\/defining-the-ecosystem\/","title":{"raw":"Defining the ecosystem","rendered":"Defining the ecosystem"},"content":{"raw":"

The ecosystems to be modelled using EwE can be of nearly any kind: the modeller sets the limits. The general rule for descriptive network analysis is to define the system so that the interactions within the system add up to a larger flow than the interactions between it and the adjacent system(s). In practice, this means that the import to and export from a system should not exceed the sum of the transfer between the groups of the system. If necessary, one or more groups originally left outside the system should be included in order to achieve this.<\/p>\r\nFor predictive models, it's a bit simpler or at least different. You formulate your research question, identify the key species and include the groups of importance for the key species. That would typically be the predators and prey of the key species along with fisheries that impacts these groups. Given that this indeed is much simpler, we'll focus the rest of this chapter on descriptive models.\r\n\r\nWhat should \u00a0you include in your model? \u00a0There is no clear answer to that, it really depends on what your research or policy question<\/a> is. \u00a0For this, there are basically two major categories, descriptive vs. predictive.\r\n\r\nFor descriptive models, you want to be complete, to include as much of the ecosystem components as practically possible. We\u2019ve often initially erred on the side of being inclusive, adding groups even if they probably had low biomasses, flows and potential impact. Then as the model took shape and it became clear that certain groups were unimportant, we\u2019d aggregate those groups to make things more manageable. \u00a0The strength of descriptive models is that they can be used to characterize the form and functioning of ecosystem, including notably for network analysis, e.g. about the ecosystem stage of development[footnote]Christensen, V., 1995. Ecosystem maturity \u2014 towards quantification. Ecological Modelling 77, 3\u201332. https:\/\/doi.org\/10.1016\/0304-3800(93)E0073-C<\/a>[\/footnote]. But their ability to address \u2018what-if\u2019 questions is severely limited or non-existent \u2013 one can use the mixed trophic impact analysis to give indications for what would happen as a result of a change in abundance or fisheries impact, but it will only be indications, not actual predictions.\r\n\r\nTo make predictions, we need predictive models, which means dynamic models that can be used to address more targeted research or policy questions than those posed to descriptive models. While descriptive models can be used as the foundation for predictive models, they come with a ballast of long development time, longer run time, and a flurry of detailed output that can make it difficult to actually get to addressing the policy \/ research questions that should drive the effort.\r\n\r\nInstead the best advice is to develop predictive models with what is needed to address the questions at hand. Focus on the target species, add important predators and fisheries, and include key prey groups and lower trophic level groups. Start off with a simple model, get to address the questions, then explore what happens if you add more details. So, don\u2019t try to make a model that perhaps someday may be useful for addressing a variety of questions, you\u2019ll spend so much time and energy doing that that you may never get to actually addressing the questions that were to drive the research.\r\n

\r\n\r\nDescriptive or predictive?<\/strong>\r\n\r\nLet's illustrate the difference between descriptive and predictive studies with an example. First, clearly define your research or policy question<\/a>. Say your task is to understand the importance of the fisheries of the Azores Islands. For that you would build a model of the EEZ and include the commercially important species along with their predators, prey and production system, and you would include the various Azorean fisheries in the model. Given your task, you would include the important skipjack tuna in your model. But skipjack is a highly migratory species with an ocean-wide distribution, not confined to the Azorean EEZ, so while it's important to include skipjack in your model, the skipjack population area is much wider than the EEZ. How do you handle that?\r\n\r\nThe simple way is to recognize that your model is restricted to the EEZ. You include skipjack tuna with both immigration and emigration. If in the Ecosim model runs, you increase the tuna fisheries, more skipjack will be caught, fewer skipjack will leave the EEZ, and the following year the same number of skipjack will immigrate to the EEZ. Your model will have skipjack included, it can fish them more or less, but it will not impact the overall skipjack population.\r\n\r\nIs that OK?\r\n\r\nTo answer that you have to consider your research question. If you are indeed describing the fisheries of the Azores Islands, all seems good with this approach. But if what you really want is to evaluate skipjack population dynamics, you need to define a model area that encompasses the skipjack distribution area and all-important fisheries. That would not be a model of the Azores Islands, but a dedicated MICE<\/a> type model, i.e. a simpler predictive model focused on the key species of relevance for your research question.\r\n\r\nDefine your model question(s), that's your focus!\r\n\r\n<\/div>\r\n

For descriptive models<\/h2>\r\n
\r\n

The groups of a system may be (ecologically or taxonomically) related species, single species, or size\/age groups, i.e., they must correspond to what is called \"functional groups.\" Using single species as the basic units has clear advantages, especially as one then can use estimated or published consumption and mortality rates without having to average between species. On the other hand, averaging is straightforward and should lead to unbiased estimates if there is information about all the components of the group. The input parameters of the combined groups should simply be the means of the component parameters, weighted by the relative contribution of the species in the group. Often one does not, however, have all the data needed for weighting the means. In such cases, try to aggregate species that have similar sizes, growth and mortality rates, and which have similar diet compositions.<\/p>\r\n

There is a facility in FishBase (www.fishbase.org<\/a>) that assembles, for any country and many ecosystems, a list of the freshwater and marine fish occurring in different habitat types, and other information useful for Ecopath models (maximum size, growth parameters, diet compositions, etc.)<\/p>\r\n\r\n<\/div>\r\n

\r\n

For tropical applications, grouping of species is always needed: there are simply too many species for a single-species approach to be appropriate for more than a few important populations. It is difficult to provide specific guidelines on how to make the groupings, as this may differ among ecosystems. Generally however, one should consider the whole ecosystem, e.g., for an aquatic model, one or two types of detritus (e.g., one to include mainly marine snow, the other discarded bycatch, if any), phytoplankton, benthic producers, herbivorous and carnivorous zooplankton, micro- and macrobenthos, herbivorous fish, planktivorous fish, predatory fish, etc., and that at least 12 groups are included, including the fishery (any number of fleets\/gears), if any. But most important is the personal judgment of what is appropriate for your system.<\/p>\r\n\r\n

The recommendation of including at least 12 groups is based on Christensen (1994),[footnote]Christensen, V. 1994. Emergy-based ascendency. Ecological Modelling 72:129-144. https:\/\/doi.org\/10.1016\/0304-3800(94)90148-1<\/a>[\/footnote] and is based on impact of number of groups on network indicators. If your model is predictive and you're not comparing network indicators with other models, you can safely ignore this recommendation.<\/div>\r\n<\/div>\r\n
\r\n

Special consideration needs to be given to the bacteria associated with the detritus. One option, applicable in cases where no special emphasis needs to be given to bacterial biomass, production and respiration, is to disregard the flows associated with these processes, which are, in any case, hard to estimate reliably, and which tend to completely overshadow the other flows in a system. (In such cases, one assumes that the bacteria belong to a different, adjacent ecosystem linked to yours only through detritus export). Alternatively, bacteria can be attached to one or all of the detritus boxes included in a system. To do this, create a \"box\" for the bacteria, and have them feed on one or several of the detritus boxes. (This is required because detritus, in the Ecopath model is assumed not to respire). Consider, finally, that there is no point including bacteria in your model if nothing feeds on them.<\/p>\r\n\r\n<\/div>\r\n

\r\n

For an overview of the ecosystem concept in ecology, we suggest you consult the book by Golley[footnote]Golley, F.B. 1993. A History of the Ecosystem Concept in Ecology: More Than the Sum of the Parts. Yale University Press, New Haven, CT.[\/footnote].<\/p>\r\n\r\n<\/div>\r\n

Open system problems<\/a><\/h2>\r\nFor almost every defined ecosystem study area, there will be some species that have life cycles that take them outside the defined area for at least part of each year.\u00a0 Movements (exchange) of biomass across the area boundary can be of two types: dispersal, involving unidirectional movement of organisms to and from sink and source populations outside the study area; and migration, involving regular, repeated movements into and out of the area by the same individuals.\u00a0 These are fundamentally different processes, with very different policy consequences.\u00a0 Dispersal acts as an extra mortality-agent and recruitment-source independent of fisheries and other impacts in the study area, while migration exposes organisms from the study area to particular risks and opportunities for part of the time, without acting as a \"permanent\" drain or source of those organisms.\r\n\r\nDispersal can (as described in the Azores Islands example above) be represented in both Ecopath and Ecosim by setting immigration and emigration rates in the Ecopath > input > <\/em>Other production<\/em> form. These rates are used in the Ecopath mass balance and are treated in Ecosim as unidirectional (non-migratory) dispersal rates. True migration is more complex to deal with, but Ecospace can handle it \u2013 see the Migration<\/a> chapter.\r\n\r\nThere are two broad options for dealing with directed migration to and from the Ecopath study area so as to avoid misleading predictions in Ecosim and Ecospace:\r\n