Search USGS⌕ Search

Geology topics

Daniel Dalthorp

Publications and source records attributed to Daniel Dalthorp.

17 recordsLinked to original sources
USGS2024

Accounting for the fraction of carcasses outside the searched area in the estimation of bird and bat fatalities at wind energy facilities

Accurate estimation of bird and bat mortality at wind energy facilities requires accounting for carcasses that lie outside the search plots because they lie beyond the search radius or in areas within the search radius that remain unsearched due to sub-optimal search conditions such as thick vegetation, rough or dangerous ground, water, or restricted access to the land. However, carcass density is not constant around a turbine and the fraction of carcasses within the unsearched area can vary greatly depending on where the area lies relative to the turbine. The density-weighted proportion approach takes into account the changing density of carcasses around turbines to estimate the fraction of carcasses lying in unsearched areas (dwp). It involves tallying the carcasses found in concentric rings centered at the turbine, fitting a curve to the carcass densities in the rings, and dividing the integral of the curve over the area searched by the integral over the total area. Accounting for unsearched area presents special difficulties such as extrapolation beyond the search radius, spatial prediction, and model selection, which are frequently ignored or under-appreciated, potentially resulting in substantial estimation errors. A powerful new R software package (dwp) is available to perform the calculations, given the distances at which carcasses were found from turbines and a map of the searched area used to discern the fraction of the ground searched at each distance. If all ground within a given search radius has been searched, the map is simply the search radius. For more complicated search plots, other kinds of maps may be used: R polygons for plots that can be readily delineated into searched and not-searched areas (for example, searches restricted to access roads and turbine pads), GIS shape files for complicated search patterns (for example, non-uniform vegetation or ground texture resulting in spatially varying search conditions), or raster files for complicated search patterns coupled with carcass spatial distribution that depends on both distance and direction from turbines. This study discusses estimation and interpretation of dwp in the context of several realistic examples; provides guidance for use of the dwp software for doing the analyses; and addresses questions of extrapolation, spatial prediction, and model selection.

Techniques and Methods↗
USGS2023

Reanalysis indicates little evidence of reduction in eagle mortality rate by automated curtailment of wind turbines

Unintended consequences of renewable energy development include collision-caused deaths of birds and bats. Energy companies may risk prosecution if protected species are among the casualties. Shutting down turbines during high collision-risk conditions could reduce mortality rates, and several companies are developing systems to identify such conditions. A recent peer-reviewed article published in the Journal of Applied Ecology reported a remarkable ‘82% (75%–89%) reduction in the fatality rate’ of eagles at a wind energy facility due to a device marketed as Identiflight®—remarkable because of the impressive effect size and the extremely high level of precision. We show that reported results stem from four major errors, which, when corrected, give an unremarkable estimate of 50% (−159%, 89%) reduction (or possible increase) in the fatality rate. The errors include the following: (i) Ignoring annual variation . They compare the average number of eagle fatalities over 4 years before activation of Identiflight® to the number in a single year after, ignoring annual variation in fatalities. (ii) Unfounded causal inference . Lack of replication (one treatment year at one site) is ignored, leading to unwarranted causal inference. (iii) Inflated effect size . Effect size is inflated by assuming (without providing evidence) that the difference in fatality relative to the mean at a neighbouring site would be exactly repeated at the treatment site. Furthermore, the observed difference in fatalities at the control site depends strongly on the arbitrarily chosen date distinguishing the ‘Before’ and ‘After’ periods, yielding unreliable results. (iv) Inconsistency of data . It is unclear why 7 of 42 reported eagle fatalities were not included in the data analyzed, potentially further inflating the estimated effect size. Synthesis and applications . The recent claim, published in the Journal of Applied Ecology, that ‘Eagle fatalities are reduced by automated curtailment of wind turbines’ is not supported by the data but stems from errors that led to strongly overstated effect size and precision, and unfounded inference. In theory, automated curtailment has obvious potential for reducing eagle fatalities, but several more years of data at several locations and appropriate statistical analyses will be required to evaluate its effectiveness and to inform management prescriptions involving this technology.

Journal of Applied Ecology↗
USGS2023

Modeling the spatial distribution of carcasses of eagles killed by wind turbines

Currently, the US Fish and Wildlife Service makes eagle permitting and management decisions nationwide based on a limited understanding of the impacts of wind power generation on eagles, and the factors that influence risk at a given facility. Accurate estimates of eagle mortality at wind power facilities form the basis for comparing the magnitudes of mortality rates in different areas and for measuring the benefits of proposed methods of minimizing the collision-caused impacts to eagle populations. Simple counts of observed eagle carcasses at wind facilities are almost certainly underestimates of the true mortality because fatalities can be removed by scavengers, be missed by searchers, or fall outside searched areas. For the latter, models of relative carcass density as a function of distance from the turbine can be fit to observed carcass locations and used to estimate the proportion of carcasses expected to land within an area of any configuration beneath a turbine. In the USA, however, it has been difficult to estimate these models for large birds such as Bald Eagles ( Haliaeetus leucocephalus ) and Golden Eagles ( Aquila chrysaetos ) due to inadequate numbers of dead eagles found at any single facility. In this case, analysis of a surrogate species might be useful to inform carcass distributions. We chose to model the carcass distribution of White-tailed Eagles ( Haliaeetus albicilla ) in Norway as an informative surrogate for Bald Eagles and Golden Eagles in the USA. Our three best-fitting parametric models were very consistent in estimating that 50% (95% CI: 40–60%) of White-tailed Eagle carcasses land within approximately 42 m of the turbines that had 70-m hubs and approximately 40-m blades. Although our models were fit to data from White-tailed Eagles and not Bald or Golden Eagles, applying these models when calculating mortality impacts of wind developments on both eagle species will likely improve the accuracy of post-construction mortality estimates, particularly at sites where substantial areas may be unsearchable. Accurate post-construction mortality estimates can inform pre-construction fatality prediction models. Resource managers can determine whether their conditions are sufficiently similar to those we modeled to warrant the use of these models for Bald and Golden Eagle carcass distributions.

Journal of Raptor Research↗
USGS2021

Relative energy production determines effect of repowering on wildlife mortality at wind energy facilities

Reduction in wildlife mortality is often cited as a potential advantage to repowering wind facilities, that is, replacing smaller, lower capacity, closely spaced turbines, with larger, higher capacity ones, more widely spaced. Wildlife mortality rates, however, are affected by more than just size and spacing of turbines, varying with turbine operation, seasonal and daily weather and habitat, all of which can confound our ability to accurately measure the effect of repowering on wildlife mortality rates. We investigated the effect of repowering on wildlife mortality rates in a study conducted near Palm Springs, CA. We controlled for confounding effects of weather and habitat by measuring turbine‐caused wildlife mortality rates over a range of turbine sizes and spacing, all within the same time period, habitat and local weather conditions. We controlled for differences in turbine operation by standardizing mortality rate per unit energy produced. We found that avian and bat mortality rate was constant per unit of energy produced, across all sizes and spacings of turbines. Synthesis and applications . In the context of repowering a wind facility, our results suggest that the relative amount of energy produced, rather than simply the size, spacing or nameplate capacity of the replacement turbines, determines the relative rate of mortality prior to and after repowering. Consequently, in a given location, newer turbines would be expected to be less harmful to wildlife only if they produced less energy than the older models they replace. The implications are far‐reaching as 18% of US and 8% of world‐wide wind power capacity will likely be considered for repowering within ~5 years.

California↗
USGS2021

Performance of the GenEst Mortality Estimator Compared to The Huso and Shoenfeld Estimators

The impacts of wind power development on bat and bird populations are commonly assessed by estimating the number of fatalities at wind power facilities through post-construction monitoring (PCM) studies. Standard methodology involves periodic carcass searches on plots beneath turbines (Strickland et al. 2011, US Fish and Wildlife Service 2012). The resulting counts are adjusted to compensate for bias due to imperfect carcass detection by searchers, removal of carcasses by scavengers or other processes (Korner-Nievergelt et al. 2011), and carcasses that may have fallen outside of searched areas. To account for the bias in counts due to imperfect detection and carcass removal, investigators typically conduct bias trial experiments to inform models of carcass detection probability. Many different estimators have been proposed that combine information about the bias trial experiments to estimate a detection probability for carcasses (g) and ultimately obtain an estimate of total mortality (M). The two estimators that have seen the most widespread use in North America recently are the Huso (Huso 2011, Huso et al. 2012) and Shoenfeld (Shoenfeld 2004; also called the Erickson estimator) estimators. GenEst (Dalthorp et al. 2018a, 2018b, 2018c) is the newest statistical estimator to become available and was designed to improve upon the Huso and Shoenfeld estimators by generalizing the key assumptions in both, and to improve comparability among new PCM studies. In addition to relaxing some of the assumptions inherent to the Huso and Shoenfeld estimators, GenEst uses a parametric bootstrap applied to a novel approach to variance estimation (Madsen et al. 2019). The current study was undertaken to document the performance of GenEst relative to the Huso and Shoenfeld estimators. We took a simulation approach to the study because simulation data provides the basis to compare mortality estimators under conditions where the “truth” is known. The estimators were compared on three metrics: 1) bias—the tendency of an estimator to over- or under-estimate actual mortality, 2) precision—the ability of an estimator to constrain an estimate to a narrow range (measured here as the width of a 90% confidence interval [CI] around the point estimate divided by the true, known mortality), and 3) CI coverage—the probability a CI with a specified level of confidence actually includes the true level of mortality. Although our simulations were conceived and designed—and are discussed—with respect to wind power facilities, it is important to note that the estimators and results discussed here are relevant to any post-construction fatality monitoring study that may occur (such as at solar facilities) where detection is imperfect. Although our study treats the problem of mortality estimation when detection is imperfect, it is also important to note that all of the estimators considered here are Horvitz-Thompson (Horvitz and Thompson 1952) style estimators, that is, none are designed to estimate the mortality of rare species as might be necessary under an Incidental Take Permit. The Evidence of Absence estimator (Dalthorp et al. 2017) is still the most appropriate statistical tool for rare event estimation. The simulations cover a broad range of conditions that may occur in field studies and complete results are presented without commentary in the appendix. The main body of this report does not provide a comprehensive treatment of our results; rather, we try to identify some of the more important differences among the estimators and some conditions under which reliable mortality estimates are especially challenging.

AWWI Technical Report↗
USGS2020

Comparing methods to estimate the proportion of turbine-induced bird and bat mortality in the search area under a road and pad search protocol

Estimating bird and bat mortality at wind facilities typically involves searching for carcasses on the ground near turbines. Some fraction of carcasses inevitably lie outside the search plots, and accurate mortality estimation requires accounting for those carcasses using models to extrapolate from searched to unsearched areas. Such models should account for variation in carcass density with distance, and ideally also for variation with direction (anisotropy). We compare five methods of accounting for carcasses that land outside the searched area (ratio, weighted distribution, non-parametric, and two generalized linear models ( glm )) by simulating spatial arrival patterns and the detection process to mimic observations which result from surveying only, or primarily, roads and pads (R&P) and applying the five methods. Simulations vary R&P configurations, spatial carcass distributions (isotropic and anisotropic), and per turbine fatality rates. Our results suggest that the ratio method is less accurate with higher variation relative to the other four methods which all perform similarly under isotropy. All methods were biased under anisotropy; however, including direction covariates in the glm method substantially reduced bias. In addition to comparing methods of accounting for unsearched areas, we suggest a semiparametric bootstrap to produce confidence-based bounds for the proportion of carcasses that land in the searched area.

Environmental and Ecological Statistics↗
USGS2020

Some approaches to accounting for incidental carcass discoveries in non-monitored years using the Evidence of Absence model

Executive Summary We evaluate three approaches to accounting for incidental carcasses when estimating an upper bound on total mortality (𝑀) as 𝑀 ∗ using the Evidence of Absence model (EoA; Dalthorp and others, 2017) to assess compliance with an Incidental Take Permit (ITP) (Dalthorp & Huso, 2015) under a monitoring protocol that includes formal, dedicated carcass surveys that achieve an overall detection probability of 𝑔 𝑠 =0.15 in the first year, followed by 4 years with no formal monitoring but with carcasses potentially discovered incidentally by operations and maintenance crews in their normal course of activity or otherwise discovered outside the formal searches. We refer to carcasses discovered incidentally as “incidentals” and define 𝑥 𝑖 as the count of incidentals. Similarly, we define 𝑥 𝑠 as the number of carcasses found during the formal searches conducted the first year.

Open-File Report↗
USGS2019

Estimating population size with imperfect detection using a parametric bootstrap

We develop a novel method of estimating population size from imperfectly detected counts of individuals and a separate estimate of detection probability. Observed counts are separated into classes within which detection probability is assumed constant. Within a detection class, counts are modeled as a single binomial observation X with success probability p where the goal is to estimate index N . We use a Horvitz–Thompson‐like estimator for N and account for uncertainty in both sample data and estimated success probability via a parametric bootstrap. Unlike capture–recapture methods, our model does not require repeated sampling of the population. Our method is able to achieve good results, even with small X . We show in a factorial simulation study that the median of the bootstrapped sample has small bias relative to N and that coverage probabilities of confidence intervals for N are near nominal under a wide array of scenarios. Our methodology begins to break down when P ( X =0)>0.1 but is still capable of obtaining reasonable confidence coverage. We illustrate the proposed technique by estimating (1) the size of a moose population in Alaska and (2) the number of bat fatalities at a wind power facility, both from samples with imperfect detection probabilities, estimated independently.

Environmetrics↗
USGS2018

GenEst statistical models—A generalized estimator of mortality

Introduction GenEst (a generalized estimator of mortality) is a suite of statistical models and software tools for generalized mortality estimation. It was specifically designed for estimating the number of bird and bat fatalities at solar and wind power facilities, but both the software (Dalthorp and others, 2018) and the underlying statistical models are general enough to be useful in various situations to estimate the size of open populations when detection probabilities and search coverages are less than 1. In this report, we outline the statistical models and data structures underlying the estimator. The models are numerous, complex, and intricately interwoven. Discussion begins with broad, high-level overviews of the general models. The lower-level technical details are then gradually added. Broader and less technical discussions on the general context and applications of the models and the use of the software are available in the software user guide (Simonis and others, 2018), vignettes bundled with the software, and the help files within the software itself.

Techniques and Methods↗
USGS2018

GenEst user guide—Software for a generalized estimator of mortality

GenEst (Generalized Estimator) is a software tool for estimating the total number of individuals arriving in an area during a specific time period when their detection probability is unknown but estimable. Its development was motivated by the need to accurately estimate the total number of bird and bat fatalities occurring at wind and solar energy facilities, but it is applicable in a variety of other contexts as well. Simple counts of carcasses are not an accurate measure of the true number of fatalities because some carcasses are inevitably missed in carcass searches. Furthermore, simple carcass counts do not allow comparison among locations or years because carcasses may be detected at different rates. This software uses data collected during carcass searches and estimates of detection rates to accurately estimate the number of fatalities and to provide a measure of precision associated with the estimate. These estimates are fundamental to understanding acute and cumulative effects of renewable energy development on wildlife populations. The software package is available with a user-friendly graphic interface as well as a flexible and powerful command-line implementation. GenEst includes tools for estimating searcher efficiency, carcass persistence, and other detection probability parameters from experimental field trials. Included in the software are example datasets for analyses, standard R package help files, this user guide, and vignettes detailing use at the command-line.

Techniques and Methods↗
USGS2017

Evidence of absence (v2.0) software user guide

Evidence of Absence software (EoA) is a user-friendly software application for estimating bird and bat fatalities at wind farms and for designing search protocols. The software is particularly useful in addressing whether the number of fatalities is below a given threshold and what search parameters are needed to give assurance that thresholds were not exceeded. The software also includes tools (1) for estimating carcass persistence distributions and searcher efficiency parameters (𝑝 and 𝑘 ) from field trials, (2) for projecting future mortality based on past monitoring data, and (3) for exploring the potential consequences of various choices in the design of long-term incidental take permits for protected species. The software was designed specifically for cases where tolerance for mortality is low and carcass counts are small or even 0, but the tools also may be used for mortality estimates when carcass counts are large.

Data Series↗
USGS2016

Wind energy development: Methods to assess bird and bat fatality rates post-construction

Monitoring fatalities at wind energy facilities after they have been constructed can provide valuable information regarding impacts of wind power development on wildlife. The objective of this monitoring is to estimate abundance of a super-population of carcasses that entered the area within a designated period of time. By definition, the population is not closed and carcasses can enter as they are killed through collision with turbines, and leave as they are removed by scavengers or decompose to a point where they are not recognizable. In addition, the population is not inherently mobile, but can only change location through some external force. A focus on number of animal carcasses comprising the super-population, combined with peculiar traits that resist classic assumptions, distinguish fatality estimation at wind-power facilities from more classic abundance estimates that can be addressed through mark-recapture techniques or other well-known abundance estimators. We review the available methods to estimate the super-population of carcasses at wind power facilities. We discuss the role of these estimates in determining appropriate levels of minimization and mitigation of impacts to individual species of concern. We discuss the potential to extrapolate these measurements to reflect the cumulative effect of the industry on individual species. Finally, we suggest avenues of research needed to strengthen our understanding of the effect wind power development has, and might have in the future, on wildlife on this continent and worldwide.

Human-Wildlife Interactions↗
USGS2015

A framework for decision points to trigger adaptive management actions in long-term incidental take permits

Introduction The U.S. Fish and Wildlife Service (USFWS) has begun to issue incidental take permits (ITPs) to wind power companies to allow limited take of bird and bat species that are protected under the Endangered Species Act, the Bald and Golden Eagle Protection Act, or the Migratory Bird Treaty Act (Huso and others, 2015). Expected take rates are determined using scientifically based collision-risk models and knowledge about the ecology of the population of interest. ITPs often include mitigation requirements to compensate for estimated take and further describe (1) adaptive management actions (AMAs) that may be required to reduce take rates if permitted rate is exceeded, or (2) additional compensatory mitigation to offset take that exceeds permitted levels. Confirming the accuracy of predicted take and providing evidence that permitted take levels have not been exceeded can be challenging because carcasses may be detected with probability much less than 1, and often no carcasses are observed. When detection probability is high, finding 0 carcasses can be interpreted as evidence that none (or few) were actually killed. As the probability of observing an individual decreases, the likelihood of missing carcasses increases, making it unclear how to interpret having observed 0 (or few) carcasses. In a practical sense, the consequences of incorrect inference can be significant: overestimating take could result in costly and unjustified mitigation, whereas underestimating could result in unanticipated declines in species populations already at risk.

Open-File Report↗
USGS2015

Estimating wind-turbine-caused bird and bat fatality when zero carcasses are observed

Many wind-power facilities in the United States have established effective monitoring programs to determine turbine-caused fatality rates of birds and bats, but estimating the number of fatalities of rare species poses special difficulties. The loss of even small numbers of individuals may adversely affect fragile populations, but typically, few (if any) carcasses are observed during monitoring. If monitoring design results in only a small proportion of carcasses detected, then finding zero carcasses may give little assurance that the number of actual fatalities is small. Fatality monitoring at wind-power facilities commonly involves conducting experiments to estimate the probability ( g ) an individual will be observed, accounting for the possibilities that it falls in an unsearched area, is scavenged prior to detection, or remains undetected even when present. When g < 1, the total carcass count ( X ) underestimates the total number of fatalities ( M ). Total counts can be 0 when M is small or when M is large and g ≪1. Distinguishing these two cases is critical when estimating fatality of a rare species. Observing no individuals during searches may erroneously be interpreted as evidence of absence. We present an approach that uses Bayes' theorem to construct a posterior distribution for M , i.e., P ( M | X , ĝ ), reflecting the observed carcass count and previously estimated g . From this distribution, we calculate two values important to conservation: the probability that M is below a predetermined limit and the upper bound ( M * ) of the 100(1 &minus; &alpha;)% credible interval for M . We investigate the dependence of M * on &alpha;, g , and the prior distribution of M , asking what value of g is required to attain a desired M * for a given &alpha;. We found that when g < ~0.15, M * was clearly influenced by the mean and variance of ĝ and the choice of prior distribution for M , but the influence of these factors is minimal when g > ~0.45. Further, we develop extensions for temporal replication that can inform prior distributions of M and methods for combining information across several areas or time periods. We apply the method to data collected at a wind-power facility where scheduled searches yielded X = 0 raptor carcasses

Ecological Applications↗
USGS2015

Mortality estimation from carcass searches using the R-package carcass: a tutorial

This article is a tutorial for the R-package carcass . It starts with a short overview of common methods used to estimate mortality based on carcass searches. Then, it guides step by step through a simple example. First, the proportion of animals that fall into the search area is estimated. Second, carcass persistence time is estimated based on experimental data. Third, searcher efficiency is estimated. Fourth, these three estimated parameters are combined to obtain the probability that an animal killed is found by an observer. Finally, this probability is used together with the observed number of carcasses found to obtain an estimate for the total number of killed animals together with a credible interval.

Wildlife Biology↗
USGS2014

Evidence of Absence software

Evidence of Absence software (EoA) is a user-friendly application used for estimating bird and bat fatalities at wind farms and designing search protocols. The software is particularly useful in addressing whether the number of fatalities has exceeded a given threshold and what search parameters are needed to give assurance that thresholds were not exceeded. The software is applicable even when zero carcasses have been found in searches. Depending on the effectiveness of the searches, such an absence of evidence of mortality may or may not be strong evidence that few fatalities occurred. Under a search protocol in which carcasses are detected with nearly 100 percent certainty, finding zero carcasses would be convincing evidence that overall mortality rate was near zero. By contrast, with a less effective search protocol with low probability of detecting a carcass, finding zero carcasses does not rule out the possibility that large numbers of animals were killed but not detected in the searches. EoA uses information about the search process and scavenging rates to estimate detection probabilities to determine a maximum credible number of fatalities, even when zero or few carcasses are observed.

Data Series↗