February 2, 2023
By the Translation Working Group

Translational ecology aims to increase the usability of knowledge and products beyond scientific communities by integrating socio-environmental knowledge and stakeholders or end users as partners in the process and projects. For the past few months, the Translation and Actionable Science Working Group has been working to gauge the translational needs of the EFI community and to identify where modelers and physical scientists see gaps in connecting to stakeholders and end users to conduct translational research.  The goal of this post is to share what the working group has identified as priorities for translational needs and share where there are connections to what other working groups and organizations are doing.

Michael Gerst (University of Maryland) led interviews in October-November 2022 with nine individuals across EFI that represent a range of early to late career stages, institutional type (academic, NGO, U.S. government agency), and gender to learn about interviewees’ experiences with forecasting projects that required stakeholder interaction, what went well, what didn’t, and what would have been useful to improve stakeholder engagement. The EFI Translation Working Group is using the results from these interviews to prioritize and develop activities that can help to fill those gaps.

The following seven topics were identified from the interviews and Working Group discussions that could be developed into self-contained seminars, manuals, or guidance documents (for short-term activities) or hosting larger workshops or proposal writing opportunities (for longer-term activities).  Cases where topics overlap with other EFI working group discussions and activities are also highlighted. 

1. How can EFI serve as an advocate for ecological forecasting to stakeholder groups, especially the public sector?
   In some cases, the individuals interviewed found that the stakeholders they work with were not interested as much in the forecasts provided as they were in the real-time data that was provided during the data collection and forecast process.  These real-time data allowed the stakeholders to use their expert knowledge to make informed decisions that may or may not have been related to the forecasts that the teams provided. Within academia, there has been quite a bit of work to raise the profile of ecological forecasting, but there is now the opportunity to bring this awareness outside academia to private and government sectors to both promote the benefits of ecological forecasts in settings outside academia and listen to the needs of stakeholders. 

2. What’s the appropriate level of stakeholder engagement? Is co-production always the right answer?
   Co-production is increasingly seen as a method for improving the fit between science and stakeholder needs. However, it can be time-consuming and expensive, necessitating assessment of whether its potential benefits are a good match for a project. While understanding of the contexts in which co-production can be fruitful has improved, there is still a gap in distilling guidelines for scientists on when it is appropriate, and if not, what other options exist.

3. How to help colleagues find collaborators across disciplines (i.e., matchmaking) as well as providing guidance on what makes a fruitful multi-disciplinary collaboration?
   This topic, as well the previous point about the appropriate level of stakeholder engagement connects with conversations and a blog post the EFI Diversity, Equity, and Inclusion (DEI) working group is having and developing that highlights the need to bring in collaborators at the beginning of a project to learn what is of most interest to them. The DEI group has focused on underrepresented individuals in terms of race/gender, but this can also be applicable to underrepresented disciplines in ecological forecasting like social science.
   This topic also connects with previous discussions in the former Partners and Knowledge Translation working group (which merged with the former Social Science group to become the current Translation working group). Previous discussions have revolved around how to keep a database of individuals and groups that support matchmaking connections. The group has also discussed the need for forums or meetings to allow groups to pitch ideas and socialize early ideas that can then be followed up to develop proposals to fund those ideas.  Clearly, this is something that resonates both within the working groups and across the broader EFI community.

4. How to initiate, foster, and maintain stakeholder relationships?
   In 2021 and early 2022 the Partners and Knowledge Transfer and Social Science working groups hosted a number of seminars on science communication and co-production.  In particular, the May 4, 2021 seminar on co-production provides some initial resources that could be built out further for this topic.  The discussions and the upcoming blog post the EFI DEI working group is developing that highlight ways to connect to underrepresented groups will also provide useful resources related to fostering relationships.

5. How to understand stakeholder decision-making processes?
   In the interviews, a few anecdotes were shared that ultimately can be summarized as: we thought we knew how stakeholders make decisions (with forecasts!) but experience eventually proved otherwise. In addition to learning the process of stakeholder engagement, interviewees thought there would be utility in helping modelers understand the universe of ways stakeholders might make decisions and where forecasts fit in (or don’t!).

6. How to set up extended project planning to ensure continued operationalization?
   It is important to have a plan in place for how ecological forecasts will be operationalized after the initial set of funding expires. Stakeholders are frustrated if they start to use a forecast and then it is no longer available when the funding is over.  NASA provides one example of how to avoid this issue.  In NASA’s Ecological Conservation Applications, solicitations (e.g., A.40 Earth Science Applications: Ecological Conservation) often require proposal teams to include partners/end users who will also be responsible for maintaining the ecological forecasting products (e.g, web apps) beyond the NASA funding period.

7. How to make data, models, and systems that are documented and reusable (FAIR data, models)?This is a topic that is of interest across multiple working groups in the EFI community. The Forecasting Standards working group has recently submitted a manuscript for publication titled “A Community Convention for Ecological Forecasting: Output Files and Metadata.” The preprint is available here: https://ecoevorxiv.org/9dgtq/. The manuscript focuses on suggestions for documenting ecological forecasts.  The Cyberinfrastructure and Methods working group has also been thinking about the issue where groups creating ecological forecasts continue to develop one-off or boutique workflows.  That working group is writing a workshop proposal to bring the together people from government agencies, industry, NGOs, and academia together to develop a way to share forecasts and workflows so people don’t need to reinvent a forecast workflow. Instead, new forecasts can be created that borrow strength from resources already developed for similar models or workflows and instead be able to focus on the details and nuances of applying a forecast in their own study system or domain. This also resonates with what the EFI NEON Ecological Forecasting Challenge is working on with the cyberinfrastructure that is set up for accessing target and meteorological data and accepting, scoring, and visualizing forecasts across multiple NEON Data streams.