What is modeling?

Well, if I had to come up with a one sentence definition I would say that modeling is the art of building an abstract representation of the world. Of course some would argue with my definition and with reason, but for our sack we can just go with it.
Why would we model to begin with you ask? Good question, the more we know the more it appears that the world is too complicated for us to understand by just looking at it. Hence, we build models of part of our surroundings to help us grasp what's happening. Evidently, the objective is for one day to be able to build a model as complex as the world, hit run and get ... the world! Until then, though, we have have to keep breaking apart what we investigate in little blocks that are manageable and that we can model easily. The idea is to be able to relatively accurately approximate a phenomenon (we can check the validity of our model by simply running it and comparing with the real world) and extract from that a better understanding of the moving pieces.

Why modeling and ecology?

We all know the concentric circles model of ecosystem (an example in the picture on the right), but is this a good model? It certainly is ... depending on your purpose! Indeed, if you want to introduce to someone the idea of ecosystem it's a great way to do it. But if you want to discuss the intricate water cycle on the planet maybe you need another model. The point is that models are great when scaled to the question at hand and in ecology we have plenty of scaling to do from microcosms to world wide systems!

The main visual that people have when talking about ecology is either the recycling/save/no waste movement or the naturalist/scientific persona in the field. Both images are powerful and legit in their own ways; however, a big part of ecology nowadays relies on laboratory experiments/analysis of data (mostly DNA collected in the field) and this is creating a ton of data ... A TON! It can be daunting, thankfully modeling is here to solve your issue ... or at least try by taking that huge amount of data and extracting the trends as well as discovering the emergent properties hidden in the hay stack of data out there.



First modeling endeavors

For my first modeling project (ever) I was wrestling with modeling one big class project at UK that ended up published in a peer-reviewed journal (Biological Invasions). We wanted to investigate what happened during biological invasions when the invader is bringing in a novel disease! Think about it as my own exploration of the horrible doing of European explorers in north america ... it was not cool!

Nonetheless, here we are looking at two species, take grey and red squirrels in Britain for example. The red squirrel is native from Britain and the grey squirrel invaded from America. The latter brought with them the squirrelpox, a benign virus for the american squirrels, but a lethal disease for the native red squirrel (historical irony if you ask me ... referring to the aforementioned disease transmission from Europeans to native Americans).

After the initial wiped-out response to the disease, one question that arose is what if the native start evolving a resistance to this disease? Would that stop the invasion? Would that even allow the native squirrels to fight back and conquer their territory back? So many questions that one can sit and wait to see the answers unravel beneath one's eyes ... or I can just sit behind a computer for a year and come up with the "probable" answer. I say probable because modeling involves assumptions and any results hinge on those assumptions ... if some of these become violated over time, the results will definitely change! No reason to panic though, if you check any modeling paper, scientists always explore what we call the "parameter space" (first picture on the left). Looking at different initial conditions to predict different results, it doesn't completely solve the assumption issue, but it sure helps.

So what's the story for our squirrels in Britain? Given our assumptions and enough time as well as space away from the invaders (doesn't sound too good for our island dwellers, right), we showed that native could evolve enough disease resistance to fight back and win the invasion war! We call this process the Collapse Spiral (see second image on the left).

Ongoing modeling endeavors

Building on my interest for crayfish, regeneration, and ecosystem engineering, I am now building up a mathematical model of crayfish demography using a life table approach. I'd like to quantify or simply qualify the effect of autotomy and regeneration on crayfish population dynamic at first. Then, if everything goes well I will implement new parts to this model including other freshwater stream dwellers and their relationship with crayfish via ecosystem engineering in their community. Stay tuned for upcoming information / publication.