Recognizing political influences in participatory social-ecological systems modeling

Abstract

Stakeholder participation in social-ecological systems (SES) modeling is increasingly considered a desirable way to elicit diverse sources of knowledge about SES behavior and to promote inclusive decision-making in SES. Understanding how participatory modeling processes function in the context of long-term adaptive management of SES may allow for better design of participatory processes to achieve the intended outcomes of inclusionary knowledge, representativeness, and social learning, while avoiding unintended outcomes. Long-term adaptive management contexts often include political influences -- attempts to shift or preserve power structures and authority, and efforts to represent the political and economic interests of stakeholders -- in the computer models that are used to shape policy making and implementation. In this research, we examine a period that included a major transition in the watershed model used for management of the Chesapeake Bay in the United States. The Chesapeake Bay watershed model has been in development since the 1980s, and is considered by many to be an exemplary case of participatory modeling. We use documentary analysis and interviews with participants involved in the model application and development transition to reveal a variety of ways in which participatory modeling may be subject to different kinds of political influences, some of which resulted in unintended outcomes, including: perceptions of difficulty updating the model in substantive ways, “gaming” of the model/participatory process by stakeholders, and increasing resistance against considering uncertainty in the system not captured by the model. This research suggests unintended or negative outcomes may be associated with both participatory decision-making and stakeholder learning even though they are so often touted as the benefits of participatory modeling. We end with a hypothesis that further development of a theory of computer model governance to bridge model impact and broader theories of environmental governance at the science-policy interface may result in improved SES modeling outcomes.

The Collaborative Policy Modeling Paradox: Perceptions of water quality modeling in the Chesapeake Bay Watershed

“…The role of CAST as a load estimation model, planning tool, and BMP accounting instrument has evolved over time as features have been added (and removed) and as it has become central to achieving the TMDL (Lim et al, 2023). The use and integration of models such as CAST into the planning and implementation processes underscores the evolving landscape of model governance.…”

“…Many of the noted technical challenges in how stakeholders use and interpret the model likely reflect difficult model implementation problems (e.g., interface design, formats of input/output data, visualization, training), especially in the context of diverse stakeholders. However, we believe the push towards simpler models also reflects the CPMP because simpler models are easier to manipulate for distributional gains (Lim et al, 2023).…”

“…However, like all policy processes, participatory modeling can be more or less effective depending on the context in which it is undertaken. Here, we introduce the concept of the CPMP as a barrier to effective participatory modeling and examine its impact on three key components of successful model governance: salience, credibility, and legitimacy (Lim et al, 2023;White et al, 2010).…”

“…CAST serves multiple key functions in helping jurisdictions meet their TMDL responsibilities and is deeply embedded in the water quality governance system (Chesapeake Bay Program, 2020). The model was developed through an ongoing collaborative -and sometimes contested -governance process marked by phases of new development, simplification, and challenges to its legitimacy (see Lim et al 2023). CAST was designed for use by member jurisdictions as a planning tool to create watershed implementation plans (WIPs), which contain sets of BMPs aimed at meeting pollution reduction goals.…”

Reflections on SES modeling: Stop me if you’ve heard this

“…This bias remains as relevant today as it did two decades ago, when Rouwette and colleagues (2002) lamented that the literature on collaborative modeling did not reflect efforts that failed to reach their goals or suffered from poor group dynamics. Specific to model development processes, while public participation is widely perceived as desirable or even necessary for SES relevant decision-making, the pitfalls or unintended consequences from broadly collaborative modeling efforts remain under-examined (but do see Cockerill et al, 2009;Lim et al, 2023).…”