In fbahoken/cartogRaflow: Filtering Matrix for Flow Mapping

This vignette of Cartograflow is dedicated to filtering an Origin-Destination matrix with a continuous distance matrix. The aim is to plot flow less than a maximum distance value, above a minimum distance travelled criterion or on a range of distances.

Main function

flowdist

-- flowdist() computes a continous distance matrix from spatial features (area or points). The result is a matrix of the distances travelled between ODs, with flows filtered or not.

Useful additional functions for a complete analysis

flowjointure

-- flowjointure() is to performs a spatial join between a flow dataset and a spatial features layer (as a map background) or another matrix

flowreduct

-- flowreduct() is to reduce the flow dataset regarding another matrix, e.g. distances travelled. \cr \cr

metric is the metric of the distance matrix : continuous (e.g. for meters) or ordinal (e.g. for adjacency). \cr

If the metric is continuous (e.g for filtering flows by kilometric distances travelled), use:\cr

d.criteria is for selecting dmin or dmax distance criteria for "continuous" metric ; Argument dmin is for keeping only flows up to a dmin criterion in km ;\cr Argument dmax for selecting values less than a dmax criterion in km.\cr
d is the value of the selected dmin or dmax criteria.

Notice that these arguments can be used as a filter criterion in flowmap().

flowmap

-- flowmap() is to plot flows as segments or arrows, by acting on the following arguments:\cr

filter is to filter or not flow's information or features \cr
threshold is used to set the filtering level of the flows when filter="True" \cr
taille is the value of the width of the flow feature \cr
a.head is the arrow head parameter (in, out, in and out) \cr
a.length is the length of the edges of the arrow head (in inches) \cr
a.angle is the angle from the shaft of the arrow to the edge of the arrow head \cr
a.col is the arrow's color \cr
plota is to add spatial features as map background to the flows's plot \cr
add is to allow to overlay flow features on external spatial features background \cr

rm(list=ls())

library(sf)
library(dplyr)
library(cartograflow)
library(cartography)

1. Load datasets

Flow dataset

# Load Statistical information
data<-read.csv2("./data/MOBPRO_ETP.csv",
                header=TRUE,
                sep=";",
                stringsAsFactors=FALSE,
                encoding="UTF-8",
                dec=".",
                check.names=FALSE)
str(data)

# Variable typing 
data$i<-as.character(data$i)
data$j<-as.character(data$j)
data$Fij<-as.numeric(data$Fij)
data$count<-as.numeric(data$count)

# Selecting useful variables
tabflow<-data%>%select(i,j,Fij)

Geographical dataset

# Load a list of geo codes
ID_CODE<-read.csv2("./data/COD_GEO_EPT.csv",
                   header=TRUE,
                   sep=";",
                   stringsAsFactors=FALSE,
                   encoding="UTF-8",
                   dec=".",
                   check.names=FALSE)
#head(ID_CODE)

CODE<-ID_CODE%>% dplyr::select(COD_GEO_EPT)

colnames(CODE)<-c("CODGEO")
#head(CODE)

# Map Background
map<-st_read("./data/MGP_TER.shp")

2. Compute distance matrix

Function aims first to compute a distance matrix then to reduce the matrix and finally to plot the (filtered) flows

Example is for euclidian distance.

head(tabflow)

tab<-flowjointure(geom="area",DF.flow=tabflow,origin = "i",destination = "j",
                   bkg=map,id="EPT_NUM",x="X",y="Y")

tab.distance<-flowdist(tab,
                       dist.method = "euclidian",
                       result = "dist")

tab.distance<-tab.distance %>% select(i,j,distance)
tab<-tab %>% select(i,j,ydata)
head(tab.distance)

3. Flowmap reduction according to distance matrix

3.1. Flow reduction according to a maximum distance matrix

Using for mapping flow less than the maximum distance travelled criterion

#reduce the flow dataset from a selected distance travelled (eg. 8.5 km)
library(rlang)

tab.flow<-flowreduct(tab,
                     tab.distance,
                     metric = "continous",
                     d.criteria = "dmax", #max distance parameter 
                     d = 8567)        #max distance value - Q1 : 8567 km

#select for all i,j flow values up to 0
flow.d<-tab.flow %>%
        select(i,j,flowfilter) %>%
        filter(flowfilter !=0)
head(flow.d)

3.2. Flowmap (filtered) according to a maximum distance travelled parameter

Using the "dmax" distance parameter to plot flows less than the maximum distance criterion (here : 8,5 km)

# Graphic parameters
knitr::opts_chunk$set(fig.width=6, fig.height=6)
par(mar=c(0,0,1,0))
extent <- c(2800000, 1340000, 6400000, 4800000)
resolution<-150

# Overlay a spatial background 
par(bg = "NA")

# Add the corresponding background 
plot(st_geometry(map), col = NA, border=NA, bg="#dfe6e1")
plot(st_geometry(map), col = "light grey", add=TRUE)


#Flowmap : flow travelled less than 8.5 km  (as the first quartile Q1)

flowmap(tab=flow.d,
        fij="flowfilter",origin.f = "i",destination.f = "j",
        bkg = map,code="EPT_NUM",nodes.X="X",nodes.Y = "Y",
        filter=TRUE,
        taille=8,           
        a.head = 1,
        a.length = 0.11,
        a.col="#f7714f",
        add=TRUE)

#Map legend
legendPropLines(pos="topleft",
                title.txt="Number of commuters\n(distance travelled less than 8,5 km)",
                title.cex=0.8,    
                cex=0.5,
                values.cex= 0.7,  
                var=c(min(flow.d$flowfilter),8567), 
                col="#f7714f",
                lwd=8,
                frame = FALSE,
                values.rnd = 0
                )


#Map cosmetic
layoutLayer(title = "Professional mobility in Greater Paris : short distance travelled",
            author = "Cartograflow, 2020",
            sources = "Data : INSEE, 2017 ; Basemap : APUR, RIATE, 2018.",
            scale = 2,
            tabtitle = FALSE,
            frame = TRUE,
            col = "grey",
            coltitle ="black"
            )

# north arrow
north("topright")

3.3. Flowmap (filtered) according to a minimum distance travelled parameter

Using the "dmin" distance parameter to plot flows above the minimum distance criterion (here : 20 km)

#head(tabflow)

tab<-flowjointure(geom="area",DF.flow=tabflow,origin = "i",destination = "j",
                   bkg=map,id="EPT_NUM",x="X",y="Y")

tab.distance<-flowdist(tab,
                       dist.method = "euclidian",
                       result = "dist")

tab.distance<-tab.distance %>% select(i,j,distance)
tab<-tab %>% select(i,j,ydata)
head(tab.distance)

tab.flow<-flowreduct(tab,
                     tab.distance,
                     metric = "continous",
                     d.criteria = "dmin",  
                     d = 19234)        #Q2 : 14518 km - Q3:19234 km

#select for all i,j flow values above to 0
flow.d<-tab.flow%>%
       select(i,j,flowfilter)%>%
        filter(flowfilter !=0)

# Graphic parameters
knitr::opts_chunk$set(fig.width=6, fig.height=6)
par(mar=c(0,0,1,0))
extent <- c(2800000, 1340000, 6400000, 4800000)
resolution<-150

# Overlay a spatial background 
par(bg = "NA")

# Add the corresponding background 
plot(st_geometry(map), col = NA, border=NA, bg="#dfe6e1")
plot(st_geometry(map), col = "light grey", add=TRUE)


#Flowmap : flow travelled up to 20 km (as the third quartile Q3)


par(mar=c(0,0,1,0))

extent <- c(2800000, 1340000, 6400000, 4800000)
resolution<-150

flowmap(tab=flow.d,
        fij="flowfilter",origin.f = "i",destination.f = "j",
        bkg = map,code="EPT_NUM",nodes.X="X",nodes.Y = "Y",
        filter=TRUE,
        taille=8,           
        a.head = 1,
        a.length = 0.11,
        a.col="#f7714f",
        add=TRUE)

# Map legend
legendPropLines(pos="topleft",
                title.txt="Number of commuters\n(distance travelled more than 19 km)",
                title.cex=0.8,    
                cex=0.5,
                values.cex= 0.7,  
                var=c(19234, max(flow.d$flowfilter)), 
                col="#f7714f",
                lwd=8, 
                frame = FALSE,
                values.rnd = 0
                )
# Map cosmetics
layoutLayer(title = "Professional mobility in Greater Paris : long distance travelled",
            author = "Cartograflow, 2020",
            sources = "Data : INSEE, 2017 ; Basemap : APUR, RIATE, 2018.",
            scale = 2,
            tabtitle = FALSE,
            frame = TRUE,
            col = "grey",
            coltitle ="black")

# north arrow
north("topright")

4. Flowmap reduction and filtering according to distance matrix

Use the "dmax" distance parameter in association with the "threshold" flowmap parameter to map flows between two values (here between the first and third quartiles [Q1 - Q3])

#head(tabflow)

tab<-flowjointure(geom="area",DF.flow=tabflow,origin = "i",destination = "j",
                   bkg=map,id="EPT_NUM",x="X",y="Y")

tab.distance<-flowdist(tab,
                       dist.method = "euclidian",
                       result = "dist")

tab.distance<-tab.distance %>% select(i,j,distance)
tab<-tab %>% select(i,j,ydata)
head(tab.distance)

tab.flow<-flowreduct(tab,
                     tab.distance,
                     metric = "continous",
                     d.criteria = "dmax",  
                     d = 19234)       #Q3:19234 km

#select for all i,j flow values up to 8567  
flow.d<-tab.flow%>%
       select(i,j,flowfilter)%>%
        filter(flowfilter >8567)      #Q1=8567km


# Graphic parameters
knitr::opts_chunk$set(fig.width=6, fig.height=6)
par(mar=c(0,0,1,0))
extent <- c(2800000, 1340000, 6400000, 4800000)
resolution<-150

# Overlay a spatial background 
par(bg = "NA")

# Add the corresponding background 
plot(st_geometry(map), col = NA, border=NA, bg="#dfe6e1")
plot(st_geometry(map), col = "light grey", add=TRUE)


#Flowmap : flow travelled up to (Q3)

flowmap(tab=flow.d,
        fij="flowfilter",origin.f = "i",destination.f = "j",
        bkg = map,code="EPT_NUM",nodes.X="X",nodes.Y = "Y",
        filter=TRUE,
        taille=8,           
        a.head = 1,
        a.length = 0.11,
        a.col="#f7714f",
        add=TRUE)

# Map legend
legendPropLines(pos="topleft",
                title.txt="Number of commuters\n(distance travelled between 8,5 and 19 km)",
                title.cex=0.8,    
                cex=0.5,
                values.cex= 0.7,  
                var=c(8567, max(flow.d$flowfilter)), 
                col="#f7714f",
                lwd=8, 
                frame = FALSE,
                values.rnd = 0
                )

# Map cosmetic
layoutLayer(title = "Professional mobility in Greater Paris : median distance travelled",
            author = "Cartograflow, 2020",
            sources = "Data : INSEE, 2017 ; Basemap : APUR, RIATE, 2018.",
            scale = 5,
            tabtitle = TRUE,
            frame = TRUE,
            col = "grey",
            coltitle ="black")

Reference

-- Bahoken Francoise (2016), Programmes pour R/Rtudio annexés, in : Contribution à la cartographie d'une matrix de flux, Thèse de doctorat, Université Paris 7, pp. 325-346. URL : https://halshs.archives-ouvertes.fr/tel-01273776, pp. 480-520.