|
Quantitative Modeling
The sample scenario presented thus far is based on a qualitative analysis of feedback loops. Such an analysis can tell you a great deal about the tendencies and possible behaviors of a system, but if you really need accurate and precise predictions about how the system will behave under various conditions, you need to enter the world of quantitative systems modeling.
In quantitative modeling, precise numbers are entered to specify the size or amount of each component in the model as well as the strength of each of the cause-and-effect relationships between components. Then the model is “run” (usually as a computer simulation) to perform a detailed analysis of how each and every quantity in the model changes under the influence of all the other quantities.
A detailed introduction to quantitative modeling is beyond the scope of this web site. Suffice it to say that good quantitative systems modeling is not trivial. It is as much an art as a science and these days usually involves sophisticated computer programming.
Most importantly, all models are necessarily simplified representations of reality and are therefore, doomed to be wrong at some level. The predictions of quantitative systems models are only as good as the simplifying assumptions that have gone into those models. If everything important has been included accurately, the model predictions will be good enough to be useful. If something important was left out or misrepresented, the model’s results will be unreliable. The danger, of course, is that it’s not always obvious whether the assumptions made were good ones or not. There are methods for dealing with this type of issue, but they too are advanced.
In most cases, it is unrealistic to expect a high school student or undergraduate college student to develop an accurate model of a real environmental issue without considerable expert guidance. Keep in mind, for example, that present-day weather forecasting models represent the best efforts of hundreds of Ph.D.s working with millions of dollars in funding over many decades, and these models are run on some of the world’s most powerful computers, yet their predictions are reliable a few days into the future at best.
In spite of these limitations, introducing students to quantitative systems models through user-friendly, yet powerful dynamic modeling software packages, such as Stella (High Performance Systems, Inc.) is of great educational value. It gives students a taste of what a computer model is and what it can (and cannot) tell them. It helps them to be good critical thinkers about “facts” based on models.
The “Resources” section of this web site provides a number of excellent references for those interested in exploring the fascinating world of quantitative systems modeling.
If Kendra decides to pursue graduate research in coral conservation, she may well use quantitative systems models to predict how ocean currents will transport coral larvae in or out of marine protected areas, to examine how coral physiology enables them to build their skeletons, or to estimate how various proposed regulations or economic incentive programs might impact coral reef health.
|
