Using Data to Improve Ecological Forecasts

×

Provenance: Tadhg Moore, Virginia Polytechnic Institute and State Univ
Reuse: This item is offered under a Creative Commons Attribution-NonCommercial-ShareAlike license http://creativecommons.org/licenses/by-nc-sa/3.0/ You may reuse this item for non-commercial purposes as long as you provide attribution and offer any derivative works under a similar license.

• Define data assimilation
• Generate an ecological forecast for primary productivity
• Describe how to assess ecological forecast accuracy
• Describe how data assimilation affects forecast accuracy and uncertainty
• Explain how updating models with data collected at different time scales (e.g., daily, weekly) and with different levels of associated uncertainty affects ecological forecasts

This entire Macrosystems EDDIE module can be completed in one 2 to 3-hour lab period or two 60-minute lecture periods for upper-level undergraduate students in Ecology, Environmental Science, Ecological Modelling, and Quantitative Ecology classes. This module is designed to be taught independently of the other Macrosystems EDDIE modules and includes materials to introduce students to general ecological forecasting concepts. If instructors are looking for a more in-depth curriculum on ecological foreasting, then we suggest partnering this module with Macrosystems EDDIE Module 5 "Introduction to Ecological Forecasting" which provides students with a more general introduction to ecological forecasting. If you teach both modules, we recommend teaching Module 5 first, then Module 7. We found that teaching this module in one longer lab section with short breaks was more conducive for introductory students than multiple 1-hour lecture periods. This module is part of a four-module series on ecological forecasting as part of the Macrosystems EDDIE project.

Quick overview of the activities in this module

See the instructor manual, provided below, for a step-by-step guide for carrying out this module. A student handout describing Activities A, B, and C, and instructor PowerPoint are also provided.

• Activity A: Students build a model to simulate primary productivity for their chosen NEON site and generate a forecast.
• Activity B: Students generate multiple forecasts while assimilating data at different temporal frequencies and with different amounts of observation uncertainty.
• Activity C: Students then explore the effect of assimilating sensor data with different levels of observation uncertainty on forecast accuracy and make management decisions using an ecological forecast.

Why macrosystems ecology and ecological forecasting?

Macrosystems ecologyis the study of ecological dynamics at multiple interacting spatial and temporal scales (e.g., Heffernan et al. 2014). For example,globalclimate change can interact with local land-use activities to control how an ecosystem changes over the next decades. Macrosystems ecology recently emerged as a new sub-discipline of ecology to study ecosystems and ecological communities around the globe that are changing at an unprecedented rate because of human activities (IPCC 2013). The responses of ecosystems and communities are complex, non-linear, and driven by feedbacks across local, regional, and global scales (Heffernan et al. 2014). These characteristics necessitate novel approaches for making predictions about how systems may change to improve both our understanding of ecological phenomena as well as inform resource management.

Forecasting is a tool that can be used for understanding and predicting macrosystems dynamics. To anticipate and prepare for increased variability in populations, communities, and ecosystems, there is a pressing need to know the future state of ecological systems across space and time (Dietze et al.2018). Ecological forecasting is an emerging approach which provides an estimate of the future state of an ecological system with uncertainty, allowing society to prepare for changes in important ecosystem services. Ecological forecasts are a powerful test of the scientific method because ecologists make a hypothesis of how an ecological system works; embed their hypothesis in a model; use the model to make a forecast of future conditions; and then when observations become available, assess the accuracy of their forecast, which indicates if their hypothesis is supported or needs to be updated. This process of iteratively updating the forecast with ecological data is called data assimilation, which can greatly improve forecast accuracy over time. Macrosystems ecologists are increasingly using ecological forecasts to predict how ecosystems are changing over space and time (Dietze and Lynch 2019).

In this module, students will generate an ecological forecast for a NEON site and explore how to use ecological data to improve forecast accuracy. This module will introduce students to the concept of data assimilation within an ecological forecast; how data assimilation can be used to improve forecast accuracy; how the level of uncertainty and temporal frequency of observations affects forecast output; and how data assimilation can affect decision-making using ecological forecasts.

Workflow for this module:

1. Give students their handout ahead of time to read over prior to class or distribute handouts when they arrive to class. The R Shiny app is set up for students to complete discussion questions as they navigate through the module activities. Thus, students could answer questions 1-2 prior to the start of instruction and can save their progress, which will allow them to return at a different time. The questions can be saved and downloaded as a Microsoft Word file at the end of the module, which could be submitted to their instructor for potential grading.
2. Instructor gives a brief PowerPoint presentation that introduces ecological forecasting, the process of data assimilation, and the ecological model students will be using (~20mins).
3. After the presentation, the students divide into pairs. Each pair selects their own NEON site and visualizes their site's data (Site selection). The two students within a pair each build their own model for predicting primary productivity (as represented by chlorophyll-a), explore different initial condition distributions and generate a forecast. For virtual instruction, we recommend putting two pairs together (n=4 students) into separate breakout rooms during this activity so the two pairs can compare results.
4. The instructor can instruct students to wait until all students are finished Activity A and then they will all begin Activity B together. For virtual instruction, this would entail having the students come back to the main room for a short check-in.
5. The students work in their pairs to generate different forecasts with and without data assimilation, as well as forecasts that assimilate data with different levels of observation uncertainty and at different temporal frequencies (e.g., daily, weekly). Students assess forecast accuracy using different data assimilation techniques by comparing their forecasts with their partners and working together to hypothesize how different methods of data assimilation will affect forecast output. 
6. Students then individually apply what they have learned to complete the management scenario individually and discuss with their partner how their decision affected forecast accuracy.

Teaching Materials:

20SEP22: Note that the materials below are in draft form. We are currently piloting this module in classrooms and are regularly updating materials based on student and instructor feedback.

• R shiny app: https://macrosystemseddie.shinyapps.io/module7/
• Student handout.docx (Microsoft Word 2007 (.docx) 1.8MB Sep27 22) - Handout for students to complete prior to the module
• Instructor Manual.docx (Microsoft Word 2007 (.docx) 595kB Oct6 22) - Instructor manual and troubleshooting for the module 
• PowerPoint presentations to introduce core concepts & module activities 
  • Instructor's Powerpoint stand-alone.pptx (PowerPoint 2007 (.pptx) 3.9MB Oct11 22) - "Stand-alone" version: use this version if this is the first or only Macrosystems EDDIE ecological forecasting module your students will complete
  • Instructor's Powerpoint multiple modules.pptx (PowerPoint 2007 (.pptx) 3.2MB Oct6 22) - "Multiple module" version: use this version if your students have already completed one or more Macrosystems EDDIE ecological forecasting modules (Module 5, 6, or 8)
• Getting Started with Shiny.pptx (PowerPoint 2007 (.pptx) 1.9MB Sep27 22) - Additional PowerPoint slides that provide a basic orientation to using an R Shiny app

Important Note to Instructors:

The R Shiny app used in this module is continually being updated, so these module instructions will periodically change to account for changes in the code. If you have any questions or have other feedback about this module, please contact the module developers (see "We'd love your feedback" below).

We highly recommend that instructors familiarize themselves with the Shiny app prior to the lesson. This will enable you to be more prepared to answer questions related to certain areas of the app's functionalities.

• Activity A:Students explore ecological model structure, prepare inputs to the ecological model and generate an ecological forecast.
• Activity B:Students assimilate data into their forecast, explore how observation uncertainty affects data assimilation, and explore how data assimilation frequency affects forecast output.
• Activity C:Students use their understanding of data assimilation to design an environmental monitoring system to inform ecological forecasts for a reservoir management scenario.

Optional pre-class readings and videos:

Articles:

• Silver, N. (2012) Chapter 6: How to drown in three feet of water.Pages 176-203 inThe Signal and the Noise: Why so many Predictions Fail – but some Don't. Penguin Books.
• Dietze, M., & Lynch, H. (2019). Forecasting a bright future for ecology.Frontiers in Ecology and the Environment,17(1), 3. https://doi.org/10.1002/fee.1994
• Dietze, M. C., Fox, A., Beck-Johnson, L. M., Betancourt, J. L., Hooten, M. B., Jarnevich, C. S., Keitt, T. H., Kenney, M. A., Laney, C. M., Larsen, L. G., Loescher, H. W., Lunch, C. K., Pijanowski, B. C., Randerson, J. T., Read, E. K., Tredennick, A. T., Vargas, R., Weathers, K. C., & White, E. P. (2018). Iterative near-term ecological forecasting: Needs, opportunities, and challenges. Proceedings of the National Academy of Sciences, 115(7), 1424–1432. https://doi.org/10.1073/pnas.1710231115

Videos:

• NEON's Ecological Forecast: The Science of Predicting Ecosystems
• Fundamentals of Ecological Forecasting Series
  • Why Forecast?
  • Forecast Analysis Cycle
  • Ensemble Kalman Filter

We'd love your feedback!

We frequently update this module to reflect improvements to the code, new teaching materials and relevant readings, and student activities. Your feedback is incredibly valuable to us and will guide future module development within the Macrosystems EDDIE project. Please let us know any suggestions for improvement or other comments about the module by sending an email to MacrosystemsEDDIE@gmail.com or filling out the form at the following link: https://serc.carleton.edu/eddie/macrosystems/faculty_feedback.