# GenEst v1.4.5

0

0th

Percentile

## Generalized Mortality Estimator

Command-line and 'shiny' GUI implementation of the GenEst models for estimating bird and bat mortality at wind and solar power facilities, following Dalthorp, et al. (2018) <doi:10.3133/tm7A2>.

# GenEst

## GenEst: Generalized Fatality Estimator

GenEst is a tool for estimating mortalities from efficiency, persistence, and carcass data.

## DISCLAIMER

This software is preliminary or provisional and is subject to revision. It is being provided to meet the need for timely best science. The software has not received final approval by the U.S. Geological Survey (USGS). No warranty, expressed or implied, is made by the USGS or the U.S. Government as to the functionality of the software and related material nor shall the fact of release constitute any such warranty. The software is provided on the condition that neither the USGS nor the U.S. Government shall be held liable for any damages resulting from the authorized or unauthorized use of the software.

## Installation

Setup and installation require several steps. Do not skip any steps.

### Updated version of R (>= 3.5.0, released on 23 April 2018):

NOTE TO EXPERIENCED R USERS: When you install a new version of R, packages that you previously installed under an older version may not be immediately available to the new R. If not, the easiest way to make them available is to copy the package folders in your old "library" folder into the "library" folder in your new R installation. Then, enter update.packages() in R. If asked about a CRAN mirror, choose the nearest location. If you are working in Windows OS and are asked whether you want to install packages "from source", choose "No".

### Third-party packages:

Several third-party pacakges are required; all are free and open source and available from CRAN. The easiest way to install them is to run the following commands in R (with guidance concerning potential dialog boxes given below the commands):


package_req <- c("corpus", "DT", "gsl", "gtools", "htmltools", "htmlwidgets", "lubridate", "MASS", "matrixStats", "mvtnorm", "Rcpp", "shiny", "shinyjs",  "survival")
package_new <- package_req[!(package_req %in% installed.packages()[,"Package"])]
if(length(package_new) > 0) install.packages(package_new)
if (packageVersion("htmlwidgets") < "1.5") install.packages("htmltools")
if (packageVersion("shiny") < "1.4.0") install.packages("shiny")


-- If asked whether you want to use a "personal library", choose "Yes"

-- If you are on Windows and are asked whether you want to install packages and their dependencies "from source", choose "No" (unless you are ready to go to lunch, in which case, you can select "Yes" and the installation may well be done by the time you get back).

### GenEst:

Click on "Tags" under the "Repository" tab on the left sidebar at https://code.usgs.gov/ecosystems/GenEst and then click the link for the specific release you want.

-- For Windows, download the compressed folder GenEst_1.x.x.zip (do not unzip) and note where it is stored. You will install from the local .zip folder.

-- For Mac OS or Unix-like OS, download the compressed file GenEst_1.x.x.tar.gz and note where it is stored. You will install from the local .tar.gz file.

If you are working directly in R (not R Studio), run the following command:

install.packages(file.choose()) # and navigate to the package archive file you just downloaded: GenEst_1.x.x.xxx


If you are working in R Studio:

Click "Install" in the Packages pane.

Select "Package Archive File (.zip; .tar.gz)" as "Install from:" in the dialog box.

Browse to where you saved the zip file, and open it so it appears in the "Package archive" space.

Click the Install button on the dialog box.

## Getting Started

### Graphical user interface (GUI): easy-to-use buttons and menus

To start the GUI, open R and enter the command:

library(GenEst)
runGenEst()


Download the User Guide from a link near the bottom of the "Help" page in the app or from https://pubs.usgs.gov/tm/7c19/tm7c19.pdf

### R command line: more functionality and flexibility

library(GenEst)
browseVignettes("GenEst")
?GenEst


Also, help files for GenEst functions are accessible in the standard R way, for example:

?pkm


GenEst User Guide: https://doi.org/10.3133/tm7C19

GenEst Statistical Models: https://doi.org/10.3133/tm7A2

## Functions in GenEst

 Name Description aicc.cpmSet Create the AICc tables for a set of carcass persistence models averageSS Tabulate an average search schedule from a multi-unit SS data table dateCols Select the date columns from a data table aicc.pkmSet Create the AICc tables for a set of searcher efficiency models app_widgets Create and manage widgets for data input, function execution, data output aicc.cpm Extract AIC and AICc for a carcass persistence model cpmSetFail Check if cpm models fail dateToDay Calculate day of study from calendar date cpmFail Check if a CP model is well-fit aicc.cpmSetSize Create the AICc tables for a list of sets of searcher efficiency models SEsi_left Calculate conditional probability of observation at a search app_ui_utilities HTML parameters app_utilities app utilities SEsi0 Calculate decayed searcher efficiency for a single pk aicc.pkmSetSize Create the AICc tables for a list of sets of searcher efficiency models app_download_functions GenEst app download funtions app_msg_functions GenEst App Messages checkSpecificModelCP Error check a specific model selection for a CP plot calcRate Estimate the number of fatalities in each search interval throughout the monitoring period. aicc.pkm extract AICc value from pkm object app_server The GenEst server definition function calcSplits Estimate the number of fatalities by up to two splitting covariates app_content GenEst Information cpLogLik Calculate the negative log-likelihood of a carcass persistence model model_utility_functions model utility functions (not exported) aicc.pkmSize Create the AICc tables for a list of sets of searcher efficiency models estM Estimate mortality estg Estimate all carcass-level detection rates and arrival intervals countCarcs Count the minimum number of carcasses in the cells checkComponents Check for model components app_ui Create the GenEst User Interface HTML checkSpecificModelSE Error check a specific model selection for an SE plot combinePreds Combine predictors SEsi_right Calculate conditional probability of observation after a series of searches app_output_utilities app utilities for formatting text, tables, figs, etc. for display calcTsplit Estimate the number of fatalities by time interval aicc Generic S3 function for summarizing AICc calcg Calculate detection probability for given SE and CP parameters and search schedule. app_panels app panel utility functions SEboxes Produce boxplots p and/or k for all cells for reference model and specific model cpmSetFailRemove Remove failed cpm models from a cpmSet object cpmSetSizeFail Check if all of the cpm models fail expandModelSetCP Expand a CP model set for plotting combinePredsAcrossModels Combine predictors across models pkmSetFail Check if pkm models fail estgGeneric Estimate generic g estgGenericSize Estimate generic detection probability for multiple carcass classes cpm Fit cp carcass persistence models defineUnitCol Auto-parsing to find the name of the unit column (unitCol) If a unit column is not explicitly defined by user in the arg list to estM or estg, then defineUnitCol parses the CO, DWP, and SS files to extract the unit column if possible. Criteria that a column must meet to be a unit column are that it is found in both data_CO and data_DWP, all units in data_CO must also be included among units in data_DWP, all units in both data_CO and data_DWP must be included among the column names in data_SS. If data_DWP = NULL, then the unit column must be included in data_CO and all its units must be included among the column names of data_SS. pkmSetFailRemove Remove failed pkm models from a pkmSet object cpmCPCellPlot Plot cell-specific decay curve for carcass persistence desc Descriptive statistics for a fitted CP model checkDate Checks whether a vector of data can be interpreted as dates obsCols_SE Select the columns from a data table that could be SE observations cpmSetSizeFailRemove Remove failed cpm models from a cpmSetSize object cpmSetSpecCPCellPlot Plot cell-specific decay curve for carcass persistence dlModTabSE Create the download version of the Searcher Efficiency model table obsCols_ltp Select the columns from a data table that could be CP Last Time Present observations obsCols_fta Select the columns from a data table that could be CP First Time Absent observations plot.gGeneric Plot results of a single generic ghat estimation logit Compute the logit or anti-logit pkmFail Check if a pk model is well-fit ltranspose Transpose a list of arrays plot.cpmSet Plot results of a set of CP models pkmParamPlot Plot parameter box plots for each cell for either p or k pkm Fit pk searcher efficiency models. dwpm Fit density-weighted proportion (DWP) models. pkLogLik Calculate the negative log-likelihood of a searcher efficiency model plot.estM Plot total mortality estimation pllogis The CDF of the loglogistic distribution mock A mock example data set pkmSECellPlot Plot cell-specific decay curve for searcher efficiency plotSEBoxTemplate template box plot pkmSetAllFail Check if all of the pkm models fail within a given set plot.gGenericSize Plot results of a set of size-based generic ghat estimations plot.cpm Plot results of a single CP model plotCPHeader The CP plot header pkmSetSizeFailRemove Remove failed pkm models from a pkmSetSize object plotSEBoxPlots p and k box plots for an SE model set pkmSetSizeFail Check if all of the pkm models fail plotSECells Plot the cellwise results of a single model in a set of SE models plot.pkmSet Plot results of a set of SE models refMod Return the model with the greatest log-likelihood prettySplitTab Create the pretty version of the split summary table removeCols Remove selected columns from column names print.corpus_frame Generic S3 function for printing corpus_frame runGenEst Launch the GenEst Application plot.splitFull Plot summary statistics for splits of mortality estimates pkmSetSpecParamPlot p or k box plots for an SE model set pkmSetSpecSECellPlot Plot cell-specific decay curve for searcher efficiency for a specific model with comparison to the cellwise model plotCPFigure Plot results of a single CP model in a set plot.splitSummary Plot summary statistics for splits of mortality estimates summary.gGeneric Summarize the gGeneric list to a simple table plot.pkm Plot results of a single pk model plotCPCells Plot the cellwise results of a single model in a set of CP models plotSEFigure Plot results of a single SE model in a set rpk Simulate parameters from a fitted pk model plotSEHeader The SE plot header prepPredictors Prepare predictors based on inputs summary.gGenericSize Summarize the gGenericSize list to a list of simple tables prepSS Create search schedule data into an prepSS object for convenient splits analyses ppersist Calculate the probability of persistence to detection qpk Quantiles of marginal distributions of $$\hat{p}$$ and $$\hat{k}$$ simpleMplot Plot a total mortality estimation for a simple situation prettyModTabCP Create the pretty version of the Carcass Persistence model table sizeCols Select the potential carcass class columns from a data table prettyModTabSE Create the pretty versions of model and summary tables print.cpm Print a cpm model object predsCols Select the predictor-ok columns from a data table rdwp Simulate parameters from a fitted dwp model readCSV Read in csv files in either format print.pkm Print a pkm model object rcp Simulate parameters from a fitted cp model summary.splitFull Summarize results of mortality estimate splits tidyModelSetSE Tidy an SE model set tidyModelSetCP Tidy a CP model set wind_cleared Wind cleared plot (60m) Search Example trimSetSize Trim a Model-Set-Size Complex to a Single Model Per Size update_input Update the inputs when an event occurs transposeSplits Transpose a splitFull array (preserving attributes) solar_PV Photovoltaic Example Dataset solar_trough Trough-based solar thermal power simulated example summary.estM Summarize total mortality estimation solar_powerTower Power Tower Example Dataset update_output Update the outputs when an event occurs update_rv Update the reactive value list when an event occurs wind_RP Wind Road and Pad (120m) Example wind_RPbat Wind Bat-Only Road and Pad (120m) Example CO_DWP Associate CO carcasses with appropriate DWP values (by unit and carcass class) SEpanel Produce a single panel in an SE summary/diagnostic plot GenEst Generalized estimation of mortality SEcols Produce a named vectory with standard SE plot colors CPcols Produce a named vector of standard CP plot colors CPdistOptions Produce the options for the distributions in the CP model SEfig Plot results of a single SE model in a set SEsi Calculate decayed searcher efficiency DWPCols Select the DWP-ok columns from a data table No Results!