Biology subjects

Chades, I.

Publications and source records attributed to Chades, I..

3 recordsLinked to original sources

bioRxiv2026

When should we use non-stationary adaptive management? A value of information analysis

O_LIMaking informed conservation decisions under climate change is a challenging task for practitioners, since decisions depend on changing environmental conditions and uncertain ecosystem responses to climate change. Given such uncertainties, the best practice to manage natural systems is adaptive management, where decisions dynamically adapt to the response of the ecosystem to previous conservation actions. Although adaptive approaches are optimal, they are also difficult to implement, have high computational costs, and recommend strategies that can be complex to interpret. These factors can hinder their on-ground application. On the other hand, simpler but suboptimal decision models can result in more interpretable recommendations, and might still yield good outcomes for ecosystems. Exploring trade-offs between complex optimal solutions and simpler sub-optimal solutions is essential for maximising conservation impact. C_LIO_LIIn this manuscript, we use value of information theory to help managers simplify their decision-models, while balancing optimality of strategies. Our approach provides modelling recommendations by determining the benefits of modelling non-stationary ecosystem dynamics and the uncertain ecosystem response to climate change. We illustrate our approach on four scenarios inspired from the management of the Great Barrier Reef, Australia, under different climate change trajectories. C_LIO_LIWe find that the two main drivers of the recommended reduction in model complexity are the strength of non-stationarity (e.g. climate change trajectory) and the degree of uncertainty in ecosystem responses to climate change (e.g. uncertainty in the thermal resistance of a coral reef). When non-stationarity is weak, the decision problem can be reduced from a non-stationary to a stationary formulation. Similarly, when uncertainty in the response to climate change is low, this uncertainty can be safely ignored in the decision-making process. Conversely, when non-stationarity is strong and/or uncertainty is high, our approach justifies the need to account for these complexities when making decisions, as simpler approaches would yield poor outcomes. C_LIO_LIThis manuscript guides managers in simplifying a modelling approach to manage ecosystems in the face of climate change. Our protocol can help simplify complex decision problems, allowing to reduce computational costs and enhance interpretability. By finding the balance between simplicity and optimality of models, this work contributes to bridging the gap between complex modelling and on-ground applications. C_LI

ecology↗

bioRxiv2024

Developing new technologies to protect ecosystems: planning with adaptive management

Technology development is an essential investment for policymakers to address contemporary global crises, including climate change, biodiversity loss, the energy transition, and emergent infectious diseases. However, investing limited resources in the development of new technologies is risky. The research and development process is unpredictable, with unknown timelines and outcomes. In addition, even after successful development, the effects of deploying a new technology remain uncertain. When confronted with these uncertainties, policymakers must determine how long they should allocate resources to developing new technologies. Informed decisions require anticipating possible successes and failures of both technology development and deployment, which is a challenging optimisation task when managing dynamic systems, such as threatened ecological systems. Using an adaptive management approach from Artificial Intelligence, we discover a time limit new technologies should be developed for, which balances costs, benefits, and uncertainties during development and deployment. We extract clear and transparent general rules for investing in new technologies, building on an analytical approximation. Using Australias Great Barrier Reef as a case study, we demonstrate how characteristics of the managed system influence the optimal investment strategy. Our approach can inform the development of new technologies in multiple domains including biodiversity conservation, public health, energy production, and the technology industry more broadly. SignificanceTechnology development is essential to address the crises our world faces, such as ecosystem collapse. With limited resources, policymakers must decide whether to invest in developing new technologies and, if ever, when to stop. Informed decisions require anticipating possible failures of both technology development and deployment, a challenging task when dealing with changing systems. Using an Artificial Intelligence approach, we find a time limit for technology development that depends on characteristics of the managed ecosystem. This work can guide technology investments in many domains such as biodiversity conservation, epidemiology, energy production and the technology industry more broadly.

ecology↗

bioRxiv2024

Interpretable Solutions for Stochastic Dynamic Programming

O_LIIn conservation of biodiversity, natural resource management and behavioural ecology, stochastic dynamic programming, and its mathematical framework, Markov decision processes (MDPs), are used to inform sequential decision-making under uncertainty. Models and solutions of Markov decision problems should be interpretable to derive useful guidance for managers and applied ecologists. However, MDP solutions that have thousands of states are often difficult to understand. Difficult to interpret solutions are unlikely to be applied, and thus we are missing an opportunity to improve decision-making. One way of increasing interpretability is to decrease the number of states. C_LIO_LIBuilding on recent artificial intelligence advances, we introduce a novel approach to compute more compact representations of MDP models and solutions as an attempt at improving interpretability. This approach reduces the size of the number of states to a maximum number K while minimising the loss of performance compared to the original larger number of states. The reduced MDP is called a K-MDP. We present an algorithm to compute K-MDPs and assess its performance on three case studies of increasing complexity from the literature. We provide the code as a MATLAB package along with a set of illustrative problems. C_LIO_LIWe found that K-MDPs can achieve a substantial reduction of the number of states with a small loss of performance for all case studies. For example, for a conservation problem involving Northern Abalone and Sea Otters, we reduce the number of states from 819 to 5 states while incurring a loss of performance of only 1%. For a dynamic reserve selection problem with seven dimensions, while an impressive reduction in the number of states was achieved, interpreting the optimal solutions remained challenging. C_LIO_LIModelling problems as Markov decision processes requires experience. While several models may represent the same problem, reducing the number of states is likely to make solutions and models more interpretable and facilitate the extraction of meaningful recommendations. We hope that this approach will contribute to the uptake of stochastic dynamic programming applications and stimulate further research to increase interpretability of stochastic dynamic programming solutions. C_LI

ecology↗