Statistical Tools for Wind and Solar Energy Development and Operations
Science Center Objects
Solar and wind power development is increasing exponentially in the United States. However, these energy sources may affect wildlife, either directly from collisions with the turbine blades or photovoltaic arrays or indirectly from loss of habitat and migration routes. An important component to understanding the effects of these renewable energy projects on wildlife is accurate and precise estimates of fatality.
Current protocols for estimating bird and bat fatality at wind-power facilities is to search designated areas below turbines to find carcasses. Carcasses may be scavenged or difficult to detect, so statistical tools and search protocols are available to estimate actual fatality. However, current estimators only adjust the observed number of carcasses found on search plots for scavenger removal, search efficiency and time between searches, not for variation in plot size among studies nor portion of the plot actually searched. Simple counts of carcasses found at wind farms also do not reflect actual fatalities because some are removed by scavengers, or are overlooked by or fall within areas inaccessible to searchers. Because the density of carcasses generally declines with distance from the turbine, the location and configuration of inaccessible areas can greatly affect the proportion of carcasses that might be missed.
Solar energy development can present several different sources of mortality to wildlife, especially birds. Estimates of the total birds killed at an entire facility must consider each of the various sources of fatality in order to be accurate. Monitoring and estimation tools similar to those available for wind energy facilities are in short supply.
Biological Statistician Manuela Huso, along with her colleage Dan Dalthorp, lead the USGS in developing more accurate fatality detection tools and monitoring protocols to assist both the solar and wind energy industry and the agencies that regulate their development and operations.
Explore the library of USGS Tools for Solar and Wind Energy Development and Operations.
GenEst - A Generalized Estimator of Mortality
Over the past few years several new estimators of wildlife mortality at renewable energy facilities - solar as well as wind, power lines as well as oil spills - have been developed, each contributing some improvement over the previous ones. However, each new estimator also leads those carrying out post-construction mortality monitoring to potential confusion regarding which estimator to use when, and how and why the estimates differ. USGS statisticians brought together the academic and private consultant developers of these different estimators to collaborate on a single approach that can provide accurate estimates of mortality when mortality takes place over an extended period of time and probability of discovery of all fatalities is less than one. GenEst is being developed as an R-package that will be available for Windows, Mac, and Linux, with both GUI and command line implementations. GenEst will be open source so anyone can see the data and all underlying statistical models will be described in detail. GenEst is supported by the Bureau of Land Management, Department of Energy, USGS, Bat Conservation International, American Wind Wildlife Association, Western EcoSystems Technology, Inc., Avangrid Renewables, and Pattern Energy.
GenEst Training Sessions are planned at several locations around the U.S., starting in late Fall 2018. Click here to send an email for more information.
A software tool used to estimate the possible number of bird and bat fatalities at wind farms when few or even zero carcasses are found. An accompanying user guide describes the underlying statistical models and contains step-by-step instructions that walk users through the calculations and interpretations of results. Version 1.0 also offers guidance in designing monitoring studies and contains step-by-step instructions that walk managers through calculations and help with interpretations of results. A training video is also available.
USGS scientists developed statistical tools to help wildlife managers and wind energy operators infer whether incidental wildlife take levels are consistent with, higher than, or lower than permitted levels, over both short-term and long-term time scales. The tools examine the accuracy of the triggers that signal wind energy companies to undertake actions to remain within allowed take rates.
Accurate estimates of wildlife fatality at wind-power facilities are critical for predicting potential fatality prior to construction, developing techniques to reduce fatalities, and assessing cumulative effects on wildlife populations. New software and a user’s guide were designed to provide estimates that are accurate and unbiased.
Federal researchers reviewed currently used methods to estimate bird and bat fatality rates at wind power facilities, and discussed the role of these estimates in determining minimization and mitigation impacts to species of concern. Using appropriate monitoring protocols and an unbiased estimator on statistically selected sites may help evaluate the cumulative effect of wind power development on wildlife at national or regional scales.
USGS scientists developed a method to statistically determine how many animals may have been killed from collisions with wind turbines based on surveys for wildlife carcasses beneath the turbines. Wildlife managers can use the method to design monitoring protocols and to determine whether a zero count really means “zero killed” or “none found.”
Accurate counts of wildlife mortalities from collisions with man-made structures can be difficult because carcasses may not be detected or may be scavenged. An international team of researchers devised software to estimate wildlife mortality from the number of carcasses found and the probability that an animal that has died from a collision is found.
USGS statisticians evaluated models relating carcass density to distance from wind turbines to estimate the proportion of carcasses expected to fall within searched areas of varying configuration. Accurately modeling change in carcass density with distance resulted in less bias in fatality estimates than expanding the search area.
A USGS report includes suggested methods for mortality monitoring at solar facilities. Authors present case studies to illustrate how distance-sampling techniques may improve overall detectability without substantially increasing monitoring costs. Following these protocols will inform efforts to avoid, minimize, and mitigate wildlife impacts from solar energy.