ALBERT

Description: Increasingly complex nonlinear World-Earth system models are used for describing the dynamics of the biophysical Earth system and the socioeconomic and sociocultural World of human societies and their interactions. Identifying pathways toward a sustainable future in these models for informing policymakers and the wider public, e.g., pathways leading to robust mitigation of dangerous anthropogenic climate change, is a challenging and widely investigated task in the field of climate research and broader Earth system science. This problem is particularly difficult when constraints on avoiding transgressions of planetary boundaries and social foundations need to be taken into account. In this work, we propose to combine recently developed machine learning techniques, namely, deep reinforcement learning (DRL), with classical analysis of trajectories in the World-Earth system. Based on the concept of the agent-environment interface, we develop an agent that is generally able to act and learn in variable manageable environment models of the Earth system. We demonstrate the potential of our framework by applying DRL algorithms to two stylized World-Earth system models. Conceptually, we explore thereby the feasibility of finding novel global governance policies leading into a safe and just operating space constrained by certain planetary and socioeconomic boundaries. The artificially intelligent agent learns that the timing of a specific mix of taxing carbon emissions and subsidies on renewables is of crucial relevance for finding World-Earth system trajectories that are sustainable in the long term. We propose a framework for using deep reinforcement learning (DRL) as an approach to extend the field of Earth system analysis by a new method. We build our framework upon the agent-environment interface concept. The agent can apply management options to models of the Earth system as the environment of interest and learn by rewards provided by the environment. We train our agent with a deep Q-neural network extended by current state-of-the-art algorithms. We find that the agent is able to learn novel, previously undiscovered policies that navigate the system into sustainable regions in two exemplary conceptual models of the World-Earth system.

Description: Human societies depend on the resources ecosystems provide. Particularly since the last century, human activities have transformed the relationship between nature and society at a global scale. We study this coevolutionary relationship by utilizing a stylized model of private resource use and social learning on an adaptive network. The latter process is based on two social key dynamics beyond economic paradigms: boundedly rational imitation of resource use strategies and homophily in the formation of social network ties. The private and logistically growing resources are harvested with either a sustainable (small) or non-sustainable (large) effort. We show that these social processes can have a profound influence on the environmental state, such as determining whether the private renewable resources collapse from overuse or not. Additionally, we demonstrate that heterogeneously distributed regional resource capacities shift the critical social parameters where this resource extraction system collapses. We make these points to argue that, in more advanced coevolutionary models of the planetary social–ecological system, such socio-cultural phenomena as well as regional resource heterogeneities should receive attention in addition to the processes represented in established Earth system and integrated assessment models.

Description: Science and policy stand to benefit from reconnecting the many notions of social-ecological resilience to their roots in complexity sciences.We propose several ways of moving towards operationalization through the classification of modern concepts of resilience based on a multi-agent-environment perspective. Social-ecological resilience underlies popular sustainability concepts that have been influential in formulating the United Nations Sustainable Development Goals (SDGs), such as the Planetary Boundaries and Doughnut Economics. Scientific investigation of these concepts is supported by mathematical models of planetary biophysical and societal dynamics, both of which call for operational measures of resilience. However, current quantitative descriptions tend to be restricted to the foundational form of the concept: persistence resilience. We propose a classification of modern notions of social-ecological resilience from a multi-agent-environment perspective. This aims at operationalization in a complex systems framework, including the persistence, adaptation and transformation aspects of resilience, normativity related to desirable system function, first- vs. second-order and specific vs. general resilience. For example, we discuss the use of the Topology of Sustainable Management Framework. Developing the mathematics of resilience along these lines would not only make social-ecological resilience more applicable to data and models, but could also conceptually advance resilience thinking.