Not my field of expertise, but if ecosystem modeling is anything like the modeling I am more familiar with, the answer to your question will likely depend on the particular context you are interested in (particular ecosystem and application of interest). And it looks like their is a entire journal dedicated to this:
http://www.journals.elsevier.com/ecological-modelling/
Happy reading! ;-)
Dola,
The effectiveness in predictions of a computer model are going to be largely dependent on the quality of input data and knowledge of the specific environmental features that affect the target factor. As you model further out in time the uncertainty of your output increases.
Think of it like this, predicting the weather tomorrow is easier than predicting the weather one year from now.
I agree with being the modeler tha main source of reliability. The second source is the software, as certain software are somewhat semplified and not allow complex ecological modeling. If quantitative modeling, high skills in algebra, statistics, and knowledge of the topics in terms of exact variables to be input and the exact relationships among each other are absolutely required. If the spatial component is required in the ecological model, a high GIS knowledge and, if necessary of remote sensing, are also a must from the modeler. Then, the only suggestion I can give you is that if you trust the scientist modeler, then keep him/her seriously into consideration...and consider what software was used.
Please Contact Prof. P.S. Roy. (https://www.researchgate.net/profile/P_Roy/). He can help u alot !!
Hi, Mi thots about it is to rememberthat all the models (by definition) are wrong!, because they are a simplification of the reality, but some of them are useful!!, and this is where I found their reliability. You should understand your own ssystem (as said before), know the sources of uncertainity and finally get an abstaction of the reliality that gives to you information about the proccess....of course the mathematical/software part of the problem is crucial (they are your main tools!), but your knowledge of the system, is crucial.
A good reference for one example (species distribution modelling) is Elith and Leathwick (I attached it)
"How reliable the approaches of ecological modelling?
Predicting ecosystem functioning."
it is case-dependent
why are you asking?
Accuracy of ecological modelling is strongly dependent on the choice of variables being factored into the model.
Considering the dynamics of the environment it is not feasible to include all the variables so making the right choice of variables that have the predictive sensitivity for the intended model is a key factor.
Not my field of expertise, but if ecosystem modeling is anything like the modeling I am more familiar with, the answer to your question will likely depend on the particular context you are interested in (particular ecosystem and application of interest). And it looks like their is a entire journal dedicated to this:
http://www.journals.elsevier.com/ecological-modelling/
Happy reading! ;-)
Of course we all must try to test the results of our models using either current or historical data. When making future projections, we should take care that some of it can be tested in the near future, if possible.
Hello Dola,
Robin Dunbar (1989 -- "Ecological Modelling in an Evolutionary Context". Folia Primatologica 53: 235-246) argued quite convincingly that in evolutionary biology modelling can be done in two very different ways. One way ecological modelling can be done (that Dunbar terms "top-down modelling") is to compile comparative data on extant species and use those data as a descriptive tool, such that emergent patterns from those comparative data are used as the basis for making predictions about either other unstudied/under-studied extant species, or about extinct species, or about the resilience of species under changed conitions in the future. A classic example of this approach is: Brown, J.H., and B.A. Mauer. 1989. Macroecology: The division of food and space among species and communities. Science 243:1145–50.
The other main way to undertake ecological modelling (that Dunbar terms "bottom-up modelling") is to generate multivariate models that integrate key aspects of the ecology and morphology of a species into what Dunbar call "a single functional model" (e.g., to consider the combined effects of key ecological variables like ambient temperature, altitude, and habitat productivity on fundamental biological variables of a species like body size, time budgets, and reproductive strategies). Dunbar himself uses this "bottom-up modelling" approach in another paper he published in 1998: "Impact of global warming on the distribution and survival of the gelada baboon: a modelling approach". Global Change Biology 4: 293-304.
Several of the answers to this question have noted that modelling will be context or species-specific. Another way of looking at this limitation is to say that the model will only be as "good", or accurate, as the quality of the data on which the model is based. A couple of great examples where this caution about data quality is stressed are:
Lozier, J.D., Aninello, P., and Hickerson, M.J., 2009. "Predicting the distribution of Sasquatch in western North America: anything goes with ecological niche modelling. Journal of Biogeography 36: 1623-1627 (I think this short paper had to be the most interesting one I read in 2010!); and,
Long, P. 2011. Species distribution modelling -- an increasingly powerful tool. Biodiversity Science: Developments in biodiversity and conservation management 2 -- http://www.biodiversityscience.com/2011/04/27/species-distribution-modelling/
Dear Dola, In my personal opinion, dynamical systems theory is a poor metaphor for ecosystem dynamics. Ecosystems are not machines and don't yield very well to mechanical analogies . I would urge you to investigate instead aposteriori methods for analyzing ecological systems. Over the past three decades a number of ecologists have been developing such methods of ecosystem analysis . Best wishes, Bob Ulanowicz
- Badges
- Science method
- Similar topics
- Biological Science
- Ecology