The Groundwater Commons Game synthesises and extends existing work on human cooperation and collective action, to elucidate possible determinants and pathways to regulatory compliance in groundwater systems globally.

Decision-makers often have to act before critical times to avoid the collapse of ecosystems using knowledge \textcolor{red}{that can be incomplete or biased}. Adaptive management may help managers tackle such issues. However, because the knowledge infrastructure required for adaptive management may be mobilized in several ways, we study the quality and the quantity of knowledge provided by this knowledge infrastructure. In order to analyze the influence of mobilized knowledge, we study how the following typology of knowledge and its use may impact the safe operating space of exploited ecosystems: 1) knowledge of the past based on a time series distorted by measurement errors; 2) knowledge of the current systems’ dynamics based on the representativeness of the decision-makers’ mental models of the exploited ecosystem; 3) knowledge of future events based on decision-makers’ likelihood estimates of extreme events based on modeling infrastructure (models and experts to interpret them) they have at their disposal. We consider different adaptive management strategies of a general regulated exploited ecosystem model and we characterize the robustness of these strategies to biased knowledge. Our results show that even with significant mobilized knowledge and optimal strategies, imperfect knowledge may still shrink the safe operating space of the system leading to the collapse of the system. However, and perhaps more interestingly, we also show that in some cases imperfect knowledge may unexpectedly increase the safe operating space by suggesting cautious strategies.
The code enables to calculate the safe operating spaces of different managers in the case of biased and unbiased knowledge.

Swidden Farming is designed to explore the dynamics of agricultural land management strategies.

An algorithm implemented in NetLogo that can be used for searching resources.

This is an Agent Based Model of a generic food chain network consisting of stylized individuals representing producers, traders, and consumers. It is developed to: 1/ to describe the dynamically changing disaggregated flows of crop items between these agents, and 2/ to be able to explicitly consider agent behavior. The agents have implicit personal objectives for trading. Resilience and efficiency are quantified using the ascendency concept by linking these to the fraction of fulfillment of the overall explicit objective to have all consumers meet their food requirement. Different types of network structures in combination with different agent interaction types under different types of stylized shocks can be simulated.

The Retail Competition Agent-based Model (RC-ABM) is designed to simulate the retail competition system in the Region of Waterloo, Ontario, Canada, which which explicitly represents store competition behaviour. Through the RC-ABM, we aim to answer 4 research questions: 1) What is the level of correspondence between market share and revenue acquisition for an agent-based approach compared to a traditional location-allocation-based approach? 2) To what degree can the observed store spatial pattern be reproduced by competition? 3) To what degree are their path dependent patterns of retail success? 4) What is the relationship between retail survival and the endogenous geographic characteristics of stores and consumer expenditures?

Under the Kyoto Protocol, governments agreed on and accepted CO2 reduction targets in order to counter climate change. In Europe one of the main policy instruments to meet the agreed reduction targets is CO2 emission-trading (CET), which was implemented as of January 2005. In this system, companies active in specific sectors must be in the possession of CO2 emission rights to an amount equal to their CO2 emission. In Europe, electricity generation accounts for one-third of CO2 emissions. Since the power generation sector, has been liberalized, reregulated and privatized in the last decade, around Europe autonomous companies determine the sectors’ CO2 emission. Short-term they adjust their operation, long-term they decide on (dis)investment in power generation facilities and technology selection. An agent-based model is presented to elucidate the effect of CET on the decisions of power companies in an oligopolistic market. Simulations over an extensive scenario-space show that there CET does have an impact. A long-term portfolio shift towards less-CO2 intensive power generation is observed. However, the effect of CET is relatively small and materializes late. The absolute emissions from power generation rise under most scenarios. This corresponds to the dominant character of current capacity expansion planned in the Netherlands (50%) and in Germany (68%), where companies have announced many new coal based power plants. Coal is the most CO2 intensive option available and it seems surprising that even after the introduction of CET these capacity expansion plans indicate a preference for coal. Apparently in power generation the economic effect of CO2 emission-trading is not sufficient to outweigh the economic incentives to choose for coal.

In order to test how prosocial strategies (compassionate altruism vs. reciprocity) grow over time, we developed an evolutionary simulation model where artificial agents are equipped with different emotionally-based drivers that vary in strength. Evolutionary algorithms mimic the evolutionary selection process by letting the chances of agents conceiving offspring depend on their fitness. Equipping the agents with heritable prosocial strategies allows for a selection of those strategies that result in the highest fitness. Since some prosocial attributes may be more successful than others, an initially heterogeneous population can specialize towards altruism or reciprocity. The success of particular prosocial strategies is also expected to depend on the cultural norms and environmental conditions the agents live in.

This is an agent-based model, simulating wolf (Canis Lupus) reappearance in the Netherlands. The model’s purpose is to allow researchers to investigate the reappearance of wolves in the Netherlands and the possible effect of human interference. Wolf behaviour is modelled according to the literature. The suitability of the Dutch landscape for wolf settlement has been determined by Lelieveld (2012) [1] and is transformed into a colour-coded map of the Netherlands. The colour-coding is the main determinant of wolf settlement. Human involvement is modelled through the public opinion, which varies according to the size, composition and behaviour of the wolf population.

[1] Lelieveld, G.: Room for wolf comeback in the Netherlands, (2012).

This model was developed to test the usability of evolutionary computing and reinforcement learning by extending a well known agent-based model. Sugarscape (Epstein & Axtell, 1996) has been used to demonstrate migration, trade, wealth inequality, disease processes, sex, culture, and conflict. It is on conflict that this model is focused to demonstrate how machine learning methodologies could be applied.

The code is based on the Sugarscape 2 Constant Growback model, availble in the NetLogo models library. New code was added into the existing model while removing code that was not needed and modifying existing code to support the changes. Support for the original movement rule was retained while evolutionary computing, Q-Learning, and SARSA Learning were added.