readData: Read in data object from files.

In MIAmaxent: A Modular, Integrated Approach to Maximum Entropy Distribution Modeling

Description

readData reads in occurrence data in CSV file format and environmental data in ASCII raster file format and produces a data object which can be used as the starting point for the functions in this package. This function is intended to make reading in data easy for users familiar with the maxent.jar program. It is emphasized that important considerations for data preparation (e.g. cleaning, sampling bias removal, etc.) are not treated in this package and must be dealt with separately!

Usage

readData(
  occurrence,
  contEV = NULL,
  catEV = NULL,
  maxbkg = 10000,
  PA = FALSE,
  XY = FALSE,
  duplicates = FALSE
)

Arguments

occurrence	Full pathway of the '.csv' file of occurrence data. The first column of the CSV should code occurrence (see Details), while the second and third columns should contain X and Y coordinates corresponding to the ASCII raster coordinate system. The first row of the csv is read as a header row.
contEV	Pathway to a directory containing continuous environmental variables in '.asc' file format.
catEV	Pathway to a directory containing categorical environmental variables in '.asc' file format.
maxbkg	Integer. Maximum number of grid cells randomly selected as uninformed background locations for the response variable. Default is 10,000. Irrelevant for presence/absence data (PA = TRUE) and ignored for presence-only data (PA = FALSE) if occurrence contains 'NA' values. See Details.
PA	Logical. Does occurrence represent presence/absence data? This argument affects how the values in occurrence are interpreted, and controls what type of data object is produced. See Details.
XY	Logical. Include XY coordinates in the output. May be useful for spatial plotting. Note that coordinates included in the training data used to build the model will be treated as explanatory variables.
duplicates	Logical. Include each coordinate in occurrence as a separate row in the output, even if multiple coordinates fall in the same raster cell. If TRUE, a presence data point in a given cell does not preclude absence data points in the same cell.

Details

When occurrence represents presence-only data (PA = FALSE), all rows with values other than 'NA' in column 1 of the CSV file are treated as presence locations. If column 1 contains any values of 'NA', these rows are treated as the uninformed background locations. Thus, 'NA' can be used to specify a specific set of uninformed background locations if desired. Otherwise uninformed background locations are randomly selected from the full extent of the raster cells which are not already included as presence locations. Only cells which contain data for all environmental variables are retained as presence locations or selected as uninformed background locations.

When occurrence represents presence/absence data (PA = TRUE), rows with value '0' in column 1 of the CSV are treated as absence locations, rows with value 'NA' are excluded, and all other rows are treated as presences. If duplicates = FALSE, raster cells containing both presence and absence locations result in a single presence row.

The names of the ASCII raster files are used as the names of the explanatory variables, so these files should be uniquely named. readData replaces underscores '_', spaces ' ' and other special characters not allowed in names with periods '.'. In MIAmaxent, underscores and colons are reserved to denote derived variables and interaction terms, respectively.

Value

Data frame with the Response Variable (RV) in the first column, and Explanatory Variables (EVs) in subsequent columns. When PA = FALSE, RV values are 1/NA, and when PA = TRUE, RV values are 1/0.

Examples

toydata_sp1po <- readData(system.file("extdata/sommerfeltia", "Sp1.csv", package = "MIAmaxent"),
   contEV = system.file("extdata/sommerfeltia", "EV_continuous", package = "MIAmaxent"))
toydata_sp1po

## Not run: 
# From vignette:
grasslandPO <- readData(
 occurrence=system.file("extdata", "occurrence_PO.csv", package="MIAmaxent"),
  contEV=system.file("extdata", "EV_continuous", package="MIAmaxent"),
  catEV=system.file("extdata", "EV_categorical", package="MIAmaxent"),
  maxbkg=20000)
str(grasslandPO)

# From vignette:
grasslandPA <- readData(
  occurrence = system.file("extdata", "occurrence_PA.csv", package="MIAmaxent"),
  contEV = system.file("extdata", "EV_continuous", package="MIAmaxent"),
  catEV = system.file("extdata", "EV_categorical", package="MIAmaxent"),
  PA = TRUE, XY = TRUE)
head(grasslandPA)
tail(grasslandPA)

## End(Not run)

MIAmaxent documentation built on Dec. 1, 2020, 5:08 p.m.