September 8, 2022
See highlights about the NEON Forecasting Challenge in a two part series from the National Ecological Observatory Network blog.
This second post provides examples of how the NEON Forecasting Challenge has been used in undergraduate and graduate classes.
https://www.neonscience.org/impact/observatory-blog/efi-neon-forecasting-challenge-classroom
You can find the first post about building a forecasting community and individual experiences participating in the Challenge here:
At the May 2022 EFI Virtual Conference we debuted the EFI Futures Outstanding Student Presentation Award. This award is given to promote, recognize, and reward an outstanding student presenter and provides valuable feedback to student presenters on their research and presentation skills. Students presenting both Posters and Oral Presentations were eligible for this award. Each poster or oral presentation were anonymously reviewed by at least two volunteer reviewers with no conflicts of interest with the presenters. In addition to being recognized for their outstanding work, award winners received $50 to spend at the EFI shop. We thank all the students who presented and the volunteers who reviewed the presentations!
Congratulations to this year’s Outstanding Presentation Award recipients!
See Yiluan and Whitney’s presentation, poster and abstracts below.
In her presentation, Yiluan introduced her pollen phenology forecast used for public health that leverages the rapidly accumulating data on leafing phenology with a nonlinear, data-driven model. Yiluan and her co-authors established a workflow for data assembly, prediction, and delivery and use the PhenoForecast platform to deliver forecasts of plant phenology and to communicate pollen allergy risks. The platform is not available online yet, but Yiluan and her collaborators are working toward that end.
In her poster, Whitney shared an overview of Module 8 of the Macrosystems EDDIE suite of educational modules that introduce students to concepts of macrosystems ecology, ecological forecasting, and quantitative literacy skills. This module in particular focuses on introducing decision-support and uncertainty communication skills, in addition to general ecological forecasting concepts and applications. Whitney’s poster also demonstrates how the effectiveness of the modules was assessed and how they increase students’ engagement and understanding of complex concepts.
Fullscreen ModeABSTRACTS
Ecological Forecasting of Leafing and Flowering Phenology During Climate Change to Inform Public Health
Yiluan Song1, Stephan B. Munch2,3,4, Kai Zhu1
1University of California, Santa Cruz, 2Southwest Fisheries Science Center, 3National Marine Fisheries Service, 4National Oceanic and Atmospheric Administration
Changes in plant phenology caused by climate change have important implications not only in ecology but also in public health. Global warming has led to an extended pollen season and increased pollen abundance. Exposure to pollen imposes significant costs on public health and is likely to exacerbate under future environmental change. The elevating risks from phenological shift calls for major improvements in the prediction of phenology and the understanding of driving mechanisms. The success in the prediction of phenology has been limited by the performance of phenology models, in particular, a mismatch between the common linear modeling approach and the nonlinear mechanisms. Traditional models with linear structures often fail to accurately predict phenology, as mechanisms of phenology can be highly nonlinear. Models that address nonlinear mechanisms with data-driven machine learning approaches are needed to overcome the limitations in phenology models. To model the nonlinear mechanisms of leafing and flowering phenology, we adopt a state-of-the-art machine learning method, Gaussian Process empirical dynamic modeling (GP-EDM). By forecasting leafing phenology observed with satellite remote sensing (MODIS) and near-surface camera imagery (PhenoCam) in the near term (as part of the EFI RCN NEON Ecological Forecast Challenge), we validate that our approach reveals complex causal relationships from time series and outperforms parametric alternatives in prediction. Applying our ecological forecasting method, we forecast the leafing and flowering phenology of four tree taxa that produce allergenic pollen at high spatial and temporal resolution (0.05 deg and daily) over the continental US. These forecasts, published daily through a Shiny App, serve to provide early allergy health advisories and warnings, which can limit pollen exposure and reduce healthcare costs. Further, the interactive query of plant phenology on the web app can raise awareness of the impacts of climate change on our health and the ecosystem. Overall, this project develops cutting-edge machine learning models for forecasting plant phenology and is a step towards mitigating the adverse impacts of unprecedented climate changes on public health.
Undergraduate Student Confidence and Understanding of Ecological Forecasting Concepts Significantly Increases after Completing a Macrosystems EDDIE Module
Whitney Woelmer1
1Virginia Tech
Ecological forecasting is an emerging approach to estimate future states of ecological systems with uncertainty, allowing society to prepare for and manage ecosystem services. Despite the increasing need to understand, create, and communicate forecasts, forecasting training has previously focused on graduate students, representing a gap in training undergraduate students as the next generation of ecologists. In response, we developed a hands-on teaching module within the Macrosystems EDDIE (Environmental Data-Driven Inquiry & Exploration; MacrosystemsEDDIE.org) educational program to introduce ecological forecasting and forecast communication to undergraduate students through an RShiny application. Tested with >250 undergraduate students, our assessment results suggest that the module significantly increased students’ ability to correctly define ecological forecasting terms, interpret forecast visualizations with uncertainty, and identify different ways to communicate forecast uncertainty for stakeholders. These results suggest that integrating ecological forecasting into undergraduate ecology curricula via software-based learning enhances students’ abilities to engage and understand complex ecological concepts.
June 15, 2022
See highlights about the NEON Forecasting Challenge in a two part series from the National Ecological Observatory Network blog.
Find the first post about building a forecasting community and individual experiences participating in the Challenge here:
You can find the second post about how the NEON Forecasting Challenge has been used in the undergraduate and graduate classes here:
https://www.neonscience.org/impact/observatory-blog/efi-neon-forecasting-challenge-classroom
December 7, 2021
December 2, 2021
Post by Jody Peters
The EFI Partner’s Working Group hosted Pablo Suarez on November 19 for an intensely participatory Zoom call session titled “Serious Fun: Reimagining Virtual Events on Ecological Forecasting”. This was the first of a 2-part series on science communication. The second session, “Sharing is Caring: Communicating Science Beyond Academic Publications” will be held on January 19. You can find details and register for the January session here.
The first session on November 19 was not recorded due to the intensely participatory nature of the call. However, below we share an overview of the call and resources shared by Pablo.
Pablo Suarez is innovation lead at the Red Cross Red Crescent Climate Centre, a visiting fellow at Boston University, and an artist in residence at the National University of Singapore. He directs the Climate Centre’s initiatives linking applied knowledge with humanitarian work, as well as new approaches to climate risk management. Examples of his work include participatory games for learning about climate change and dialogue, and forecast-based financing pilots in Togo and Uganda.
Pablo started with the premise that disasters are collaborating better than humans are – for example, things like the pandemic, climate issues, and conflicts are exponentially causing harm and we need to have better collaborations between those who know and those who do. And if we want to effectively work with other scientists, people impacted by what we study, or policymakers, we need to let people know how much we care about what they care about. Often we convey information and work to develop collaborations through meetings and presentations typically with a keynote or series of talks followed by a short Q&A session. Pablo advocated that there are better ways to communicate and develop collaborations during meetings and conferences and that is by using games and including humor and spontaneity as a way to broaden engagement and make connections.
As I worked with Pablo to coordinate the logistics for the call and compile the advertising material, I didn’t grasp what the “Serious Fun” part of his talk title meant. But after participating on the call, I am inspired to think about ways to incorporate his suggestions for reimagining virtual events which are low-carbon and low-budget compared to in-person events. We know Zoom fatigue is real, but by including spontaneity, humor, and games, virtual events can be more engaging and can lead to deepened networks and collective action. By having fun, we can be more effective at conveying serious ideas about science, ecological forecasting, and making decisions.
Below I describe the two major activities held during the call and end with a list of the resources Pablo shared.
One group activity was called Rant. Early on in the call, we had a short amount of time to write out all our frustrations with Zoom calls. It was fast-paced and Pablo really encouraged us to get into the Rant by expressing our frustrations through things like capitalization, exclamation marks, etc. Later after other activities in the call, we had the opportunity to go back to rants submitted by ourselves and by others on the call to think about changes that could be made to alleviate issues with Zoom calls. You can see a few example rants and changes below and you can find all the rants here. I particularly appreciated the second example below, since that was my rant.
The other major activity Pablo facilitated was a version of Decisions for Seasons, a game designed to support learning and dialogue about key aspects of planning for extremes and experiencing climate change impacts and managing risk. In this game, everyone had 10 forecasts about the climate conditions for the next year. We could choose if the conditions would be regular/normal (thumbs up) or if we should plan for drought insurance (buckets) or flood insurance (umbrellas). This game was also fast-paced with only about a minute to make these decisions. After everyone had made their predictions about the future conditions, Pablo rolled a die where a 1 meant there was a drought and a 6 meant there was a flood and we all hoped we had picked enough drought or flood insurance to handle those rolls! The fun part was then seeing whose predictions were correct and who ended up getting broken hearts for incorrect predictions (see the second image below). Although we didn’t have time during our call, an extension to this game is to incorporate climate change by using an 8-sided die where floods become represented by a 6, 7, or 8 increasing the chance of an extreme event. You can find more details about how this game can be played here.
Throughout the call we had fun, there were moments of heart-racing tension to get our predictions and answers in, there were opportunities for competition and collaboration, and plenty of joy. But there was also a seriousness moving forward to think about what we are doing with forecasts, how we can improve forecasts, how we can improve people’s access, understanding, and opportunity to act, and how we can enable people to make decisions with forecasts.
While this seminar focused on how to make virtual events more engaging, Pablo’s suggestions and resources can also be applied to in-person conferences and meetings, classes, and lab meetings. Let’s have more fun, let’s be more engaged!
For those interested in exploring how to incorporate games into your presentations and to find ways to make your audiences know you care about what they care about, the following are resources Pablo shared during the call.