occTest: Occurrence tests

In occTest: Characterizing and Filtering Species Occurrence Data

Occurrence tests

Description

Perform tests for data quality in species occurrence using several methods

Usage

occTest(
  sp.name,
  habitat = NULL,
  sp.table,
  r.env,
  tableSettings = NULL,
  analysisSettings = NULL,
  writeoutSettings = NULL,
  gradingSettings = NULL,
  return.spatial.data = FALSE,
  r.dem = NULL,
  ntv.ctry = NULL,
  inv.ctry = NULL,
  resolveAlienCtry = FALSE,
  resolveNativeCtry = FALSE,
  interactiveMode = FALSE,
  verbose = FALSE,
  doParallel = FALSE,
  mc.cores = 2
)

Arguments

sp.name	character. Name of the species.
habitat	NULL
sp.table	data.frame. Object with the coordinate data.
r.env	raster or rasterStack. Environmental data(e.g. typically climatic.
tableSettings	list. Elements corresponding to different settings of the input occurrence table.
analysisSettings	list. Elements corresponding to different settings of the analysis functions .
writeoutSettings	list. Elements corresponding to different settings of the analysis functions .
gradingSettings	list. Not implemented yet. Defaults to NULL.Elements corresponding to different settings of the analysis functions .
return.spatial.data	logical. Should the spatial dataset of analysisSettings attached to the metadata?, default is FALSE to save memory
r.dem	raster. Elevation data (in meters).
ntv.ctry	character. vector with ISO3 code of the countries where species is considered native
inv.ctry	character. vector with ISO3 code of the countries where species is considered invasive
resolveAlienCtry	logical. To automatically try to detect countries for which species is considered native
resolveNativeCtry	logical. To automatically try to detect countries for which species is considered alien
interactiveMode	logical. Should prompts be ouput for some decisions taken by the workflow? Deafult FALSE,
verbose	logical. Print workflow information? Default to FALSE
doParallel	logical. Should some operations be run in parellel when possible? Default to FALSE
mc.cores	numeric. If doParallel is TRUE, then how many cores to use? Default to 2

Value

data frame with the tests performed (field $_test), specific comment for the tests ($_comments), the exact value of the test ($_value), and scores summarizing results across tests with an objective ($_score)

Note

The output dataframe allows users to classify or scrub the occurrences the way they want to.
The names of the columns in the output object have the following naming convention:
$AnalysisType$_$SpecificTest$_value: numeric or logical. Shows the quantitative result of the test (sometimes the same as in the result of the _test)
$AnalysisType$_$SpecificTest$_test: logical Shows whether the occurrence passes or not the test, being TRUE a flag for a wrong record and NA indicating that the test was not performed on that record.
$AnalysisType$_$SpecificTest$_comment: character. Shows some comments related to the specific test.
Examples: HumanDetection_HumanInfluence_value gives you the score of current human influence in the record HumanDetection_HumanInfluence_test gives you whether we consider the former value an error/bias (TRUE) or not (FALSE) HumanDetection_HumanInfluence_comment gives you a commen that give further detail on the analysis. In this case that the threshold of 45 was used for the test. HumanDetection_score summarizes all the other HumanDetection tests and outputs a value from 0 to 1. A value of 0.5 would indicate that half of the tests used indicate that is an a Human signal in the record.

Examples

### THIS IS A CUT DOWN  EXAMPLE 
### visit vignetteXtra-occTest for more info
#load environmental raster
library (raster)
library (sf)
library (occTest)
#load occurrence data
occData <- read.csv (system.file('ext/exampleOccData.csv',package = 'occTest'))
#load environmental raster
renv <- raster (system.file('ext/AllEnv.tif',package = 'occTest'))
#load elevation raster
dem <- raster (system.file('ext/DEM.tif',package = 'occTest'))
#load settings
settings <- readRDS (system.file('ext/exSettings.rds',package = 'occTest'))
#run occTest
out = occTest(sp.name='MyFake species',
             sp.table = occData,ntv.ctry = 'ESP',inv.ctry = 'FRA',
             tableSettings = settings$tableSettings,
             writeoutSettings = settings$writeoutSettings,
             analysisSettings = settings$analysisSettings,
             r.env = renv,r.dem=dem)

occTest documentation built on Nov. 18, 2022, 5:07 p.m.