Search for: All records

Abstract Knowledge co-production offers a promising approach to design effective and equitable pathways to reach development goals. Fisheries Strategies for Changing Oceans and Resilient Ecosystems by 2030 (FishSCORE), a United Nations Ocean Decade programme, will co-produce knowledge that advances solutions for climate resilient fisheries through networks and partnerships that include scientists, stakeholders, practitioners, managers, and policy experts. FishSCORE will establish (1) a global network that will develop broadly relevant information and tools to assess and operationalize climate resilience in marine fisheries and (2) local and regional partnerships that will apply those tools to identify and forward context-specific resilience strategies. FishSCORE's activities will be guided by a set of core principles that include commitments to inclusivity, equity, co-leadership, co-ownership, and reciprocity. FishSCORE will focus on identifying solutions for climate resilient fisheries, and it will also advance goals associated with capacity, power, and agency that will support iterative, pluralistic approaches to decision-making in fisheries experiencing ongoing climate-driven changes. This process of co-producing knowledge and strategies requires considerable investments of time from all partners, which is well aligned with the Ocean Decade. However, secure funding must be prioritized to support and implement co-production activities over this long time horizon. 

Abstract Both the ecological and social dimensions of fisheries are being affected by climate change. As a result, policymakers, managers, scientists and fishing communities are seeking guidance on how to holistically build resilience to climate change. Numerous studies have highlighted key attributes of resilience in fisheries, yet concrete examples that explicitly link these attributes to social‐ecological outcomes are lacking. To better understand climate resilience, we assembled 18 case studies spanning ecological, socio‐economic, governance and geographic contexts. Using a novel framework for evaluating 38 resilience attributes, the case studies were systematically assessed to understand how attributes enable or inhibit resilience to a given climate stressor. We found population abundance, learning capacity, and responsive governance were the most important attributes for conferring resilience, with ecosystem connectivity, place attachment, and accountable governance scoring the strongest across the climate‐resilient fisheries. We used these responses to develop an attribute typology that describes robust sources of resilience, actionable priority attributes and attributes that are case specific or require research. We identified five fishery archetypes to guide stakeholders as they set long‐term goals and prioritize actions to improve resilience. Lastly, we found evidence for two pathways to resilience: (1) building ecological assets and strengthening communities, which we observed in rural and small‐scale fisheries, and (2) building economic assets and improving effective governance, which was demonstrated in urban and wealthy fisheries. Our synthesis presents a novel framework that can be directly applied to identify approaches, pathways and actionable levers for improving climate resilience in fishery systems. 

A key problem in computational sustainability is to understand the distribution of species across landscapes over time. This question gives rise to challenging large-scale prediction problems since (i) hundreds of species have to be simultaneously modeled and (ii) the survey data are usually inflated with zeros due to the absence of species for a large number of sites. The problem of tackling both issues simultaneously, which we refer to as the zero-inflated multi-target regression problem, has not been addressed by previous methods in statistics and machine learning. In this paper, we propose a novel deep model for the zero-inflated multi-target regression problem. To this end, we first model the joint distribution of multiple response variables as a multivariate probit model and then couple the positive outcomes with a multivariate log-normal distribution. By penalizing the difference between the two distributions’ covariance matrices, a link between both distributions is established. The whole model is cast as an end-to-end learning framework and we provide an efficient learning algorithm for our model that can be fully implemented on GPUs. We show that our model outperforms the existing state-of-the-art baselines on two challenging real-world species distribution datasets concerning bird and fish populations.