Fuzzy Cognitive Mapping and Nonlinear Hebbian Learning for the Qualitative Simulation of the Climate System, from a Planetary Boundaries Perspective

In the present work, we constructed a collective fuzzy cognitive map for the qualitative simulation of the Earth climate system. The map was developed by considering the subsystems on which the climate equilibrium depends, and by aggregating different experts opinions over this framework. The resulting network was characterized by graph indexes and used for the simulation and analysis of hidden patterns and model sensitivity. Then, linguistic variables were used to fuzzify the edges and aggregated to produce an overall linguistic weight for each one. The resulting linguistic weights were defuzzified using the center of gravity technique, and the current state of the Earth climate system was simulated and discussed. Finally, a nonlinear Hebbian learning algorithm was used for updating the edges of the map until a desired state was reached, defined by target values for the concepts. The results are discussed to explore possible policy implementation, as well as environmental decision making.