American Geophysical Union (AGU) Fall Meeting December 12-16, 2016, San Francisco, CA
Quantifying uncertainy: An obligation or tool of discovery? Mark E. Harmon
International Long Term-Ecological Research Open Science Meeting October 9-13, 2016, Kruger National Park, South Africa
The International LTER Network held its first global Open Science Meeting. QUEST collaborators organized an "Estimating Uncertainty in Measurements, Experiments, and Models Workshop."
Long Term-Ecological Research All Scientist's Meeting August 30-September 2, 2015, Estes Park, CO
This year's conference was organized around the theme "From Long-term data to understanding: toward a predictive ecology", and showcased 300 posters and more than 75 formal and ad-hoc working group meetings. QUEST collaborators meet informally, but also used their sites data in these presentations:
UFRO World Congress: Sustaining Forests, Sustaining People, The Role of Research October 5-11, 2014 Salt Lake City, Utah, USA.
Presenters from four different countries described sources of uncertainty in estimates of forest carbon and nutrient pools and fluxes, including natural spatial and temporal variation, measurement error, model uncertainty, and model selection, and addressed how these uncertainties can guide monitoring designs and affect management and policy decisions.
Technical Sessions #58A & B: Quantifying Uncertainty in Forest Measurements and Models: Approaches and Applications.
Presentations shared approaches to analyzing uncertainty in forest measurements and giving examples of applications of uncertainty in above- and belowground estimates of forest biomass, carbon, and nutrient pools and fluxes, as well as other ecosystem attributes.
Tools for Estimating Uncertainty in Ecology @ ESA August 10, 2014
Although methods are well established for statistical analysis of most experimental designs, there are fields in Ecology where it is more difficult to establish confidence in results (e.g., in catchment studies, treatments are rarely replicated). For environmental networks, using standardized approaches ensures that results are comparable, but sometimes the same statistical technique is not applicable to comparable data sets (e.g. when there are significant differences in the sample size of the same population at two geographically distinct locations). Many of these concerns can be addressed through the appropriate use of tools for uncertainty analysis. This workshop will highlight current developments in uncertainty estimation across many fields of ecology. Overview presentations will focus on practical examples of how uncertainty calculations can inform data over small-to-large scales. Data packages and software tools will be shared with the attendees. Participants are encouraged to bring their own data sets and laptop computers; we will provide data for the exercises if you donít bring your own. We welcome participation by researchers in all career stages and from a broad array of ecological disciplines.
Uncertainty Analysis: A Critical Step in Ecological Synthesis August 5, 2013
Ruth Yanai, Jeffrey Taylor and Mark Harmon organized, and John Battles moderated an "Organized Oral Session" on Uncertainty Analysis: A Critical Step in Ecological Synthesis at the 2013 Ecological Society of America Annual Meeting in Minneapolis, MN.
Ecology is entering an exciting era in which the number and availability of long-term data sets are increasing exponentially. There is an unprecedented need to synthesize these data to address current scientific and societal problems. Great progress has been made on linking data and theory, including spatial integration and interdisciplinary combination. The question is no longer how to synthesize, but how well we link information from disparate sources and how to identify the most important areas for improvement. These synthetic approaches will demand increased proficiency and rigor in uncertainty analysis, to provide a metric of progress in synthesis science. This OOS highlights current developments in uncertainty estimation across many fields of ecology and provided guidance for large-scale synthesis research. Speakers were encouraged to provide recommendations for standardized approaches to uncertainty estimation and a vision for meeting future needs. Further development, understanding, and dissemination of the latest statistical techniques for deriving these estimates both inform ecological sampling design and equip up-and-coming ecologists with critical skills. Speakers examined sources of uncertainty and its general role in synthesis science. Case studies included a range of topics and approaches ranging from population ecology and small watershed nutrient cycling budgets to landscape carbon budgets. Methodologies presented include parametric statistical approaches, bootstrap analysis, Monte Carlo sampling, and Bayesian hierarchical analysis. Uncertainty introduced by spatial and temporal interpolation are common themes across scales from plots to the continental ecological observatory network.
Excerpts from the February 8 2011 Webinar
This preceded the first Quantifying Uncertainty in Ecosystem Studies meeting in Boston, March 14-15 2011.
The webinar and meeting were the first phases of our Long Term Ecological Research Network Office Synthesis Working Group (LTER SWG)'s evaluation of uncertainty in hydrologic inputs, outputs, and net hydrologic flux of major elements across small watersheds with diverse characteristics.