Biodiversity Knowledge Framework

The Biodiversity Knowledge Framework supports the Biodiversity 2037 MERF by providing a method for identify knowledge gaps, so that we can better invest in research, monitoring and data collection. The interactive Knowledge Portal makes this more accessible and dynamic.

The Biodiversity Knowledge Framework is part of the Biodiversity 2037 MERF and it provides the process for identifying knowledge gaps and prioritising them for research investment. This way, we’re using our time and resources as effectively as possible.

Targeted data collection is critical for effective evidence-based decision making. We know that there are knowledge gaps that will be critical to effective management; our challenge is to identify these knowledge gaps and prioritise them for research investment. New information is most valuable when it addresses large portions of uncertainty that, once resolved, could lead to highly beneficial actions.

Resources for biodiversity conservation are always limited and often contested. Ideally, the majority of program resources will go towards directly delivering on-ground outcomes, but this can easily slip to much less than this, particularly if efficient support processes are not used. For example:

Similarly, information is always limited and there are many opportunities for improvement that are unlikely to be realised. Biodiversity 2037 champions the use of decision-science approaches to help make sound choices about which opportunities to pursue. This starts by bringing together available evidence and expert opinion in structured processes to consider insights, gaps, and consequences. This allows us to be strategic in selecting new research projects.

The Knowledge Framework provides an approach to support investors by ranking the potential value of filling knowledge gaps. Rather than asking what researchers would like to do, then assessing whether this aligns with strategic biodiversity information needs, Expressions of Interest are sought for projects/approaches to directly address identified needs. This process creates more focused investment to generate more benefit for more species, rather than spreading the same budget over many smaller, less powerful projects.

The information and priorities identified using the Knowledge Framework processes (i.e. the Relative Benefit of Knowledge and causal maps) can also provide a guide for other investors (such as Parks Victoria) and researchers.

An example for the overall Biodiversity 2037 approach is managing the impact of deer on biodiversity. The approach is described below:

• Strategic Management Prospects (SMP) identifies that deer impact is a widespread issue;
• There are current control methods, but their level of effectiveness is not yet known
• A causal map (fuzzy cognitive map) of deer control and biodiversity outcomes is developed by a group of experts
• Knowledge gaps are identified. Currently, available knowledge is used to indicate how effective, affordable and feasible a new method would need to be in order to justify investment to fund further research into it. Knowledge gaps are ranked and then prioritised against others for investment.

Anyone interested in the methods behind DELWP’s knowledge acquisition process can read the Knowledge Framework, though it is particularly targeted at:

• Research scientists, universities and other organisations who are interested in addressing priority knowledge gaps
• Land managers to understand implications of their management strategies and the relationships between actions
• Investors to help them decide which projects to fund

Everyone will be able to contribute to the implementation of the Framework through the Biodiversity Knowledge Portal. There will be an annual review and update of models. Portal users will be able to interact with the models and external groups will be able to indicate where they are doing or planning to do research. Currently, queries and contributions can be directed to biodiversity.knowledge@delwp.vic.gov.au

It is currently in use. The Biodiversity Knowledge Framework has already been used as the basis of the 2019 Knowledge Grants.

The Biodiversity Knowledge Portal was released in May 2021. The portal provides access to all available models.

To identify knowledge gaps and prioritise research investment, there needs to be a consistent, quantifiable and systematic approach. Part of this process is the development causal maps. Causal maps show the relationship between important biodiversity values and intervention actions. This ensures that assessment of knowledge gaps uses a whole-of-ecosystem view.

This approach also allows knowledge gaps to be ranked in accordance with an index of Relative Benefit of Knowledge (the benefit to biodiversity gained from doing a particular piece of research compared to another). The technique we have chosen to develop the causal models is Fuzzy Cognitive Maps. Fuzzy Cognitive Maps are an established method in ecology and conservation to explore the impacts of different management scenarios and to combine knowledge from different sources. We have been collaborating with key experts in the decision science space at the University of Melbourne and Arthur Rylah Institute to design the process and develop the initial maps. Importantly, our approach has been designed to be drawn on benefit and uncertainty data from the expert elicitation conducted to support Strategic Management Prospects (SMP) and Specific Needs Assessments.

These tools are explicitly designed to capture uncertainty about the effectiveness of an action (in response to a threat) and the outcome for species, or groups of species. However, there is still uncertainty surrounding the impact of an action on a species (e.g. are were uncertain about fox control because of uncertainty surrounding the method or because of uncertainty around the impact of reduced fox density on the target species?). The intention is that the causal models will be easily accessible, reviewed, tested and updated.

There are a range of biodiversity knowledge improvement approaches, each with strengths and weaknesses The most appropriate approach should be used to address the question we are interested in, rather than setting up all the different approaches. For example:

• Adaptive management studies are desirable because they can be designed to provide feedback on specific treatments/outcomes, but due to the challenges of measuring change in complex systems and discriminating between correlation and causation, under-funded studies may provide limited insights
• Remote sensing provides repeatable overviews that can cost-effectively show trends in natural resource features, but these are typically general rather than species-specific
• Long-term monitoring sites can detect gradual changes, but these may be slow and expensive to identify, and depending on the number and layout of sites, typically do not provide any insights to management effectiveness or specific/urgent conservation issues (e.g. Forest Monitoring Plots Project has not been useful for solving Leadbeater’s Possum or Greater Glider issues)
• Broad scale citizen science projects (e.g. bird atlas) are good for engagement due to the ease of observing and identifying species, but generally confirm things we already know (e.g. movements due to seasons/climate variations; populations crashes due to droughts/fires; local extinctions due to loss of habitat) rather than improve our choices about what to do next