Trajectories of a binary varible"

In binaryTimeSeries: Analyzes a Binary Variable During a Time Series

binaryTimeSeries

This tutorial illustrates the concepts of the trajectories of a binary variable using an example data.

knitr::opts_chunk$set(
  collapse = TRUE,
  fig.width=8.5, 
  fig.height=6,
  comment = "#>"
)
# Make sure you load all the libraries below before you start using the package.
library(binaryTimeSeries)
library(ggnewscale)
library(reshape2)
library(raster)
library(terra)
library(dplyr)
library(maptools) 

Data

The example data is a time series of five binary maps at five time points: 2000, 2001, 2002, 2003, and 2005. The tutorial analyzes the trajectory of variable 1, which in this instance indicates presence, whiles 0 represents absence.

The time series maps are available in GeoTIFF and CSV formats and included in the package. The first step is to load the raster files.

# Read the raster files from "externa"l folder as terra raster format.
rasstackY <- terra::rast(system.file("external/Example_raster_Y.tif",package="binaryTimeSeries"))

#NB: The package comes with "Example_Data_Y.tif", "Example_Data_Y.cvs", "Example_raster_X.tif", and "Example_Data_X.cvs."

# Set the spatial resolution of the data. 
datares <- c( 1000, 1000)

# Set the crs of the data.
datprj <-"+proj=utm +zone=32 +datum=WGS84 +ellps=GRS80  +units=m +no_defs"

Calling the changeData function

Next, use the changeData function to create a list containing the raster data sets concerning the number of times the category of interest is present and the number of times the category of interest changes during the time series. Output from the changeData function serves as input data for the presencePlot function. Type "?changeData" in your R console to see the help file.

# Create input data for the presencePlot function and assign the results to a variable, e.g., "a."
a <- changeData (rasstackY,
                 nodata = 2,
                 category = 1,
                 spres = datares,
                 datacrs = datprj)
a

Calling the presencePlot function

Use the presencePlot function to create the number of presences and change maps. Type "?presencePlot" in your R console to see the help file.

presencePlot (input = a,
              pltunit = "m",
              dataEpsg = 32632,
              scalePos = "bottomleft",
              narrowPos = "topright",
              narrowSize = 1,
              categoryName = "marsh",
              xAxis = "Longitude (m)",
              yAxis = "Latitude (m)",
              axisText = 1.2,
              axisLabel = 1.2,
              plotTitle = 1)

Calling the trajData function

The trajData function creates the data that serves as input for the trajPlot function. Type "?trajData" in your R console to see the help file.

# Create input data for the trajPlot function and assign the results to a variable, e.g., "b."
b <- trajData(x = rasstackY,
             nodata = 2,
             category = 1,
             spres = datares,
             datacrs = datprj,
             unified = "yes")# if "no", the analysis considers the entire study region.

b

Calling the trajPlot function

The trajPlot function creates a map and pie chart of the trajectories of the category of interest. Type "?trajPlot" in your R console to see the help file.

trajPlot(input = b,
         pltunit = "m",
         dataEpsg = 32632,
         categoryName = "marsh's",
         scalePos = "bottomleft",
         narrowPos = "topright",
         narrowSize = 1,
         xAxis = "Longitude (m)",
         yAxis = "Latitude (m)",
         axisText = 1.2,
         axisLabel = 1.4,
         plotTitle = 1.5,
         legendTex = 0.9)

Calling the dataStack function

The dataStack function create the data which serves as input for the "stackbarPlots" function. Type "?dataStack" in your R console to see the help file.

# Create a vector variable containing the time points
timepoints <- c(2000,2001,2002,2003,2005)

# Use the dataStack function to create input for the stackbarPlots function.
c <- dataStack(x = rasstackY,
               category = 1,
               nodata = 2,
               unified = "yes",
               timePoints = timepoints,
               categoryName = "marsh",
               regionName = "extent")
c

Calling the stackbarPlots function

The stackbarPlots function creates stacked bar plots showing the sizes of a category's trajectory during each time interval and the three Change Components. Type "?stackbarPlots" in your R console to see the help file.

# Create the stacked bars 
stackbarPlots(input = c,
              axisSize = 10,
              lbAxSize = 12,
              lgSize = 12,
              titleSize = 12,
              datbreaks = "no",
              upperlym = 60,
              lowerlym = - 50,
              lymby = 5,
              upperlym2 = 35,
              lymby2 = 5,
              xAngle = 0)

References

Pontius Jr, R. G. (2022). Metrics That Make a Difference: How to Analyze Change and Error. Springer Nature Switzerland AG.

Bilintoh, T.M., (2022). Intensity Analysis to Study the Dynamics of reforestation in the Rio Doce Water Basin, Brazil. Frontiers in Remote Sensing, 3 (873341), 13.

Pontius Jr, R. G. (2019). Component Intensities to relate difference by category with difference overall.International Journal of Applied Earth Observation and Geoinformation, 77: 94-99.

Pontius Jr, R. G., Krithivasan, R., Sauls, L., Yan, Y., & Zhang, Y. (2017). Methods to summarize change among land categories across time intervals. Journal of Land Use Science, 12(4), 218–230.

Pontius Jr, R. G., & Santacruz, A. (2014). Quantity, exchange, and shift components of difference in a square contingency table. International Journal of Remote Sensing, 35(21), 7543–7554.

binaryTimeSeries documentation built on July 18, 2022, 5:07 p.m.