Current Practices in Reporting Uncertainty in Ecosystem Ecology

Abstract

Ecosystem budgets of water and elements can be difficult to estimate and are often unreplicated, making it challenging to provide confidence in estimates of ecosystem pools and fluxes. We conducted a survey to learn about current practices in reporting uncertainties in precipitation, streamflow, soils, and vegetation. Uncertainty derives from natural variation, which is commonly characterized by replicate samples, and from imperfect knowledge, which includes measurement error and model error (model fit and model selection). We asked questions about whether researchers report uncertainties in these sources, whether they know how to do so, and how important they believe the sources to be. We also asked questions about identifying missing or unusable values, filling gaps in data, and dealing with analytical concentrations below detection limits. We obtained responses from 140 researchers representing 90 research sites around the world. Natural variation was the most important source of uncertainty in calculations of biomass and soil pools, according to respondents in these fields, and sampling error was the source they most often reported. In contrast, uncertainty in the chemical analysis of precipitation and stream water was the source most commonly reported by hydrologists, although they rated this one of the least important sources of uncertainty to calculations of hydrologic flux. Awareness of types of uncertainty can help identify sources of uncertainty that may have been overlooked, and quantifying them will help determine which sources are most important to report.