Natural Selection of Model Ecosystems for Industrial Ecology

by Jessica Bass

Whether gathering observations virtually, outdoors, or in a laboratory, the process of establishing the conditions for and collecting a strong body of data is often a lengthy exchange of trials and errors. Constant advancements in information and technology, and a dependence upon simulated modeling in some areas, can leave scientists, and potentially entire fields of study, continually struggling to catch up. Bollinger et al. (2015) suggest, however, that efforts to create such scientific models and to gather data, can be approached strategically, offering a potential for more quickly-developed and longer-lasting systems that may benefit fields highly dependent upon simulations for data collection and research. Through analysis of various modeling techniques, the study produces a list and detailed rationale of suggested guidelines for timely and effective development, integration, and use of model ecosystems. Their findings demonstrate the importance of simplicity, openness, and resource and information sharing to support fields such as Industrial Ecology.

The theory of Universal Darwinism recognizes evolution as a generic process of natural selection through which all forms and ideas are tested and refined with the passing of time. The additions, eliminations, and combinations that occur over the course of many trials and triumphs bring rise to the most resilient forms and concrete ideas and theories, especially those with strength in the ability to adapt to a changing environment. Continued tests, through research as well as through the changing external environment, will gradually shape resilient scientific models that will respond to a variety of factors and give rise to data that may lead the way into the future of fields such as industrial ecology, Bollinger et al. predict. Yet scientists wondered if it might be possible to cultivate these superior, evolved models without the large amounts of time typically required to shape and refine them. Bollinger et al. sought to analyze the various approaches and influential actors in the process of designing scientific models in order to identify best practices and establish a set of guidelines that may allow for the timely creation of a body of models that will nurture further simulations.

Scientific modeling is, ideally, composed of simplistic structures that allow for clear and direct analysis while helping avoid error.   However, the growing need for a single, sustainable system capable of embracing and responding to a wide variety of influencing factors typically requires use of highly complex, integrated systems. Observing the resilience of various models, Bollinger et al. noted the tendency of those composed by and of diverse groups of people and multiple different systems that supplement one another, to outlast more simplistic models as conditions changed and the models were required to evolve. The sociotechnical factors of modeling, and the defining aspects of model variation, selection, and heredity, have allowed researchers to identify and observe the factors that influence and support the creation of strong model ecosystems.

The keys to generating such fertile ground for modeling and data collection for the advancement of any field are based, first and foremost, upon the need for diversity, teamwork, and openness in creation and interaction. In order to cultivate mutual support for the advancement of a field such as industrial ecology, Bollinger et al. suggest a strong focus upon sharing and receiving information and resources. The first step in this direction is to ensure equal access through the use of freely accessible software and formatting and, to the greatest extent possible, simplicity in model components and explanations.   Second, researchers of a common field must foster a community that can work together to build an abundant model ecosystem that encourages collaboration, even promoting sharing or borrowing designs. Research teams may pass on information in research journals, or often even by issuing digital object identifiers for data or code. Beyond those deeply and directly involved in a particular field of study, colleagues, institutions, and even the general public, hold strong potential to assist, offering a new perspective as they engage in research and review efforts.

Cultivating multimodel ecology will demand patience and flexibility, at times requiring a rearrangement of designs or responsibilities in order to serve the community as a whole.   While time limitations with the need for additional focus upon outreach initiatives, difficulties in model simplification, and challenges in direct sharing and borrowing, may serve as noteworthy barriers to realizing the these guidelines in full, the potential for model growth, interaction, and coevolution as part of a supportive ecosystem, holds significant promise to facilitate the advancement of fields such as Industrial Ecology.

Bollinger, L. A., Nikolić, I., Davis, C. B., Dijkema, G. P.J., 2015. Multimodel Ecologies: Cultivating Model Ecosystems in Industrial Ecology. Journal of Industrial Ecology, 19, 252–263.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s