You are here:
Assessing the transferability of ecosystem service production estimates and functions
Citation:
DeWitt, Ted, J. Moon, M. Errend, M. Gray, AND Randy Bruins. Assessing the transferability of ecosystem service production estimates and functions. A Community for Ecosystem Services, Washington, DC, December 08 - 12, 2014.
Impact/Purpose:
An EPA team of scientists are developing a methodology to assess the reliability of using estimates and models of ecosystem service stocks and production applied to locations or contexts different from where those were originally measured or developed. Communities that wish to sustain or increase the availability of ecosystem goods and services (EGS) for the benefit of their populace need ways to estimate EGS stocks and their sensitivity to human actions or stressors. Measurement of EGS stocks and production are costly and time consuming, and many studies rely on applying existing measurements or models (i.e., transferring ecological information) to their sites of interest. However, those information transfers may not be appropriate (and thus inaccurate) because of contextual differences between the development site and the application site. The “transferability assessment” (TA) methodology will help with the assessment of the accuracy and error (i.e., variability) of transferring estimates or models of EGS. This methodology will be useful for identifying which environmental models and data are most appropriate to be used at sites where data are limited.
Description:
Estimates of ecosystem service (ES) production, and their responses to stressors or policy actions, may be obtained by direct measurement, other empirical studies, or modeling. Direct measurement is costly and often impractical, and thus many studies transfer ES production estimates or models that were obtained or developed in other contexts. This is analogous to socioeconomic benefits transfer, but in the ecological realm. However, few studies rigorously assess the assumptions, accuracy, and errors associated with ecological information transfers, nor the implications for using these estimates at new sites. We have developed methodology to help assess the risk of transferring ES estimates and models. For ES estimates, the transferability assessment (TA) focuses on identifying the expected level of variance in a transferred estimate by describing the contextual differences between the estimate’s origin and the transfer site. For ES models, the TA focuses on evaluating the transferred model’s potential performance level given both contextual differences between sites and possible changes made to the model. A key part of this work is identifying contextual information that might predict the success of a transfer and creating metrics of contextual difference. The TA was designed in collaboration with the development of EPA’s Ecosystem Service Production Function Library (EPFL). We demonstrate an example of a TA, including how transfer reliability can be visualized across dimensions of context.