What´s the difference between Modeling and Simulating

[lalvarez]

Hello Everybody

"Modeling vs. Simulation"

Can anyone help me to clarify these two terms? I´m working with lobster larvae dispersion on the west coast of Baja Peninsula with ROMS, but for some of my advisors what i´m doing is a simulation and for the others it´s modeling.

So i need a clear definition of the terms in order to unify my advisors ideas of what i´m doing.

Thanks in advance

[ptimko]

Try this...

A simulation is a recreation of a past event to examine what has happened in the past...

Modelling is a study of the processes and forces involved to understand a physical or other other type of system...

If you are only recreating the past from a single event I would say you are doing a simulation...

If you are recreating several past events and looking for the common factors between the simulations I would say you were modelling...

This is of course just my opinion, other's might feel differently.

[rduran]

Perhaps they are being used interchangeably.

I think one could argue that if you are doing a simulation you are surely modeling (whatever you are simulating you are using a model to simulate it) and if you are modeling ... well you are certainly simulating something (if not reality then some idealized experiment).

My guess is that your advisers may not be distinguishing between the two, then again maybe they are. One thing is almost for sure: they do have different ideas of what you should be doing. The only real way to figure this out is talking with them.

My suggestion is that you think about what they told you, come up with a way to achieve it and discuss it with them before spending too much time working on it. That way you can make sure you iron out any inconsistencies and avoid spending valuable time doing something different from what you should be doing. The process of thinking how to achieve whatever they asked from you should help you understand what they want from you as well.

Animo!

edit: and if you can come up with some valuable suggestions of yourself in the process ... advisers like to see their advisee contributing scientifically.

[arango]

I agree with Timko and I think that you are doing both. If you are coding/adapting a partial differential equation and associated parameters to study the dispersion of lobster larvae, you are doing modeling. Now, you need to perform a lot of simulations with this physical/mathematical model to see if the solutions agree with the available data. Then, I will say that you are simulating because you are exploring the parameter space to verify your model.

A very interesting question! We sometimes use both modeling and simulating interchangeably. In my my mind there is a difference between both. We usually use both words in different contexts.

[kate]

I expect that some people use the word simulating for what we normally call modeling. Why? Because you can build models that don't require simulating - box models, idealized models, whatever. They want to call what we do something else to differentiate between the two. I agree that asking is the best policy.

[lalvarez]

Hello to All

Thanks a lot for your answers.

In deed, what is undestood as Model in physical oceanography has a very different meanings to other sciences. I agree with Kate, there is a clasification given by Haeffner (2005, Modelling Biological Systems -book)about models, from the verbal model (spoken knowledge passed from person to person) to the formal models (expresss by a mathematical differential equations); cannot remember right now but ROMS has a large clasification name, believe me.

The exercise we planned involves the California spiny lobster which has a larvae life history with a period of 1 year as a pelagic drifter, actually it has not been performed at least here in Mexico. We are going to work with a domain from Point Conception, California to Cabo San Lucas, Baja California because the biggest lobster population inhabits the middle of Baja Peninsula and the other not so big is in the Southern California Bight. We will explore and try to explain how the role of meso-scale circulation processes affect and define the dispersion and connectivity of this species along the southern region of the California Current. As Hernan Arango says a lot of simulating is waiting to be performed and some modelling inbetween with LTRANS, adjusting the model for the spiny lobster.

Thanks Ptimko and Rduran for your suggestions, they have been very useful to clarify this with the people i´m working with.

Have a nice day

[rduran]

Hi all,

I am not persuaded that you are not simulating when modeling, take a box model for example. A much simplified model but I still see it as simulating say a very basic physical principle or the likes. Nor am I persuaded that you are not modeling when simulating.

What I do agree is that giving modeling and simulating different definitions is very useful. I also agree this has been done (giving them specific definitions) and therefore one can use them non-interchangeably.

The reason mathematics is rigorous is because great care and tremendous effort is put into defining anything that is used. I don't know any way around carefully defining things if we want to be rigorous.

Looking into wikipedia (which may or may not be accurate)

A simulation is the implementation of a model ... A simulation brings a model to life and shows how a particular object or phenomenon will behave. Such a simulation can be useful for testing, analysis, or training in those cases where real-world systems or concepts can be represented by models.

http://en.wikipedia.org/wiki/Scientific_modelling

This agrees with what I said: you need to be using a model if you are simulating. It also seems that, as Hernan said, exploring parameter space would be simulating (testing/analysis of model).

Also from wikipedia:

Simulation is the imitation of the operation of a real-world process or system over time.[1] The act of simulating something first requires that a model be developed; this model represents the key characteristics or behaviors/functions of the selected physical or abstract system or process. The model represents the system itself, whereas the simulation represents the operation of the system over time.

Here simulation is restricted to imitating a real-world process and that makes a difference (now running a model may not necessarily be a simulation). It would seem to me from this latter definition that exploring parameter space would be part of developing the model (i.e. modeling) with which at the end a simulation can take place.

So in summary I insist that a specific definition must precede. In this case the relevant definition would be a standard definition for physical oceanography.

Nice project lalvarez! Sounds like nested grids might be a good idea for your model...

[nganju]

I think it's important for modelers to be able to defend their approaches, terminology, and philosophy to non-modelers. There are many disparaging comments made about models/modelers with little consideration for the value of model-derived information. There is substantial literature on the epistemology (study of knowledge) of models. The discussion of modeling vs. simulating is within this scope though we may be splitting hairs.

There are some great papers about numerical modeling by Prof. Eric Winsberg (USF, Philosophy Dept.) that are worth reading. Here's the link: http://usf.academia.edu/EricWinsberg/Papers

I also recommend "Models as Mediators" by Morgan and Morrison. Most of the book is about economic or population models, but the first few papers have general accounts and a few hydrodynamic examples.

[lalvarez]

Hello Nganju

I have visited the site of Eric Winsberg and i found some interesting readings

Thanks a lot for the information

[tony1230]

My opinion on "simulating" & "modelling" much agree with ptimko and Hernan but i have own words.

As i can understand within my scope of oceanography knowledge, the "Simulating" is an action with which we do the works/jobs repeatedly as

our predecessors have already been done and verified by many researchers to be ture. In the simulating what we want to know would no more

than what we have known, we just want a copy of results (as ptimko said is a recreation of a past event).

As to "Modelling", i treated it as a further step on simulating. We are modelling some processes to which we may do some exploration by

simulation, either on parameters or on equations, even just on results which were rarely known to others. Modelling is also you do

simulation to understand or corroborate some precesses/phenomena that perplexes you or to grobe for something that may exists.

I'm afraid i may not agree much with wikipedia rduran quoted (i have not referred to all in that page and am not sure if my thoughts are make

sense). Either we do a simulating or modelling, we are implementing a model. But the main difference between them is, "simulating" uses

a model that we built with parameters/equations unedited so you do not know whether the simulating results are correct. If not (by available

data), we may want to adjust/modify some of them as then can agree more to your specific project backgrounds. After some steps (adjust para,

calibrate and validate the model), we got results that can be explained by knowledge from textbooks or documentations/papers, i call this

"Modelling". To some extent, it seems to me that "simulating" likes a warm-up before we paly games and the game itself is "modelling" .


All is my own opinion. Any arguments were warmly welcomed. Thanks.

- Steve W. Shou

[rduran]

There is a recent article for the hydrological modeling literature that reads (section 3):

There is considerable confusion in the literature about the use of the terms simulation, prediction, projection, forecasting, and hindcasting; the usage in hydrological modeling in some cases conflicts with usage in other areas. It would be desirable for authors and referees to use of these terms with more rigor.

I feel this agrees with my suggestion that --for physical oceanography-- it is necessary to have a standard definition to be able to make sure that we are talking about the same thing.

I wonder if there is a similar paper defining these terms for physical oceanography -- I believe there should be.

Here is the link to the article:
http://onlinelibrary.wiley.com/doi/10.1 ... 3/abstract


A guide to good practice in modeling semantics for authors and referees
Keith Beven, Peter Young
Article first published online: 26 AUG 2013
DOI: 10.1002/wrcr.20393