Plotting

Plotting is done using ggplot2 which provides a powerful alternative paradigm for creating both simple and complex plots in R using the ideas the Grammar of Graphics¹ The idea of the grammar is to specify the individual building blocks of a plot and then to combine them to create the graphic desired².

The ggplot functions expects a data.frame for its first argument, data; then a geometric object geom that specifies the actual marks put on to a plot and an aesthetic that is “something you can see” have to be provided. Examples of geometic Objects (geom) include points (geom_point, for scatter plots, dot plots, etc), lines (geom_line, for time series, trend lines, etc) and boxplot (geom_boxplot, for, well, boxplots!). Aesthetic mappings are set with the aes() function and, examples include, position (i.e., on the x and y axes), color (“outside” color), fill (“inside” color), shape (of points), linetype and size.

The phase plot plots stock status against fishing mortality relative to target reference points as a two-dimensional phase plot.

Back to Top

Uncertainty

cv=rlnorm(100,log(stock(ple4)),0.3)
ggplot(cv)+
  geom_boxplot(aes(factor(year),data))

Age structure

sel  =apply(harvest(ple4),1,mean)
ggplot(sel)+
  geom_line(aes(age,data))

Trends

q    =FLQuant(cumprod(1+rep(.02,dim(fbar(ple4))[2])),dimnames=dimnames(fbar(ple4)))
plot(q)

Hyperstability

cpue   =stock(ple4)/mean(stock(ple4))
stable =cpue^0.1
deplete=cpue^2

plot(FLQuants("CPUE"          =cpue,
              `Hyper_Stable`  =stable,
              `Hyper_Depleted`=deplete))

trend=FLQuant(seq(1,2,length.out=dim(stock(ple4))[2]),dimnames=dimnames(stock(ple4)))
var  =trend*abs(cv)*sign(cv)

ggplot(FLQuants("Log Normal" =cv,
                "Trend in CV"=var))+
  geom_boxplot(aes(factor(year),data))+
  facet_grid(qname~.)

bias=FLPar(omega=1,ref=mean(stock(ple4)),q=0)

hyperstability<-function(object,omega=1,ref=apply(object,c(1,3:6),mean)) 
  ref%*%((object%/%ref)^omega)

bias<-function(object,bias=0.02) 
     FLQuant(cumprod(1+rep(bias,dim(object)[2])),dimnames=dimnames(object))

set.seed(1234)
u     =FLQuants("Unbiased"      =rlnorm(100,log(apply(oem(ple4),2:6,sum)),.3),
                "Hyperstability"=rlnorm(100,log(apply(oem(ple4),2:6,sum)%*%
                                                  hyperstability(stock(ple4),0.52)),.3),
                "Trend"         =rlnorm(100,log(apply(oem(ple4),2:6,sum)%*%bias(stock(ple4),0.02)),.3),
                "AR"            =apply(oem(ple4),2:6,sum)%*%
                                   exp(rnoise(100,apply(oem(ple4),2:6,sum)*0,.3,b=.7)),
                "Variable"      =var,
                "Juvenile"      =rlnorm(100,log(apply(oem(ple4,sel=mat(ple4)),2:6,sum)),.3),
                "Mature"        =rlnorm(100,log(apply(oem(ple4,sel=1-mat(ple4)),2:6,sum)),.3),
                "Numbers"       =rlnorm(100,log(apply(oem(ple4,mass=FALSE),2:6,sum)),.3))

u=FLQuants(llply(u,function(x) x/mean(x)))
u=ldply(u,as.data.frame)

u.=ddply(u,.(year,.id), with, quantile(data))
ggplot()+
  geom_line(aes(year,data,col=factor(iter)),
            data=subset(u,iter%in%c(2,11)))+
  geom_ribbon(aes(year,ymin=`25%`,ymax=`75%`),data=u.,col="grey",alpha=.5)+
  facet_wrap(~.id,ncol=2)+
  theme_bw()+theme(legend.position="none")

library(FLCore)
library(ggplotFL)
library(tmvtnorm)
library(plyr)
library(reshape) 

data(ple4)

n =survey(ple4)
n =round(1000*n%/%apply(n,2:6,sum))
l =stock.wt(ple4)^0.3

sigma=FLPar(0,dimnames=dimnames(n)[c(1,1:6)])
sigma=apply(sigma,3:7,function(x) {diag(x)=0.2;x})
sigma=FLPar(c(sigma),dimnames=dimnames(n)[c(1,1:6)])

sigma=FLPar(aperm(maply(dimnames(l)$year, function(year)           cor2cov(sigma@.Data[,,year,,,,,drop=T],l[,year])),3:1))
names(dimnames(sigma)[3])="year"

upper=n%=%20

x=rtmvnorm(n=max(n[,1]), mean=c(l[,1]), sigma=matrix(c(sigma[,,1,drop=T]),10,10), upper=c(upper[,1]))
x=melt(x)

names(x)[1:2]=c("iter","age")
n=as.data.frame(n,drop=TRUE)
x =subset(merge(n,x),iter<=data)

ggplot(x)+
  geom_histogram(aes(value,fill=factor(age)))+
  xlab("Year")+ylab("Count")

library(FLCore)
library(tmvtnorm)
library(reshape2)

lem2<-function(n,l,sd){

  se=sd/n^0.5
  se=(apply(se%*%se%*%n,c(2:6),sum)%/%apply(n,2:6,sum))^0.5
  
  ln=apply(l%*%n,2:6,sum)%/%apply(n,2:6,sum)
    
  FLQuants(len=ln,se=se)}

data(ple4)

n =survey(ple4)
n =round(1000*n%/%apply(n,2:6,sum))
l =stock.wt(ple4)^0.3
sd=l*0.2

lengths=lem2(n,l,sd)

ggplot()+
  geom_line(aes(year,data),data=as.data.frame(lengths[[1]]))+
  geom_errorbar(aes(year,ymin=min,ymax=max),
                data=model.frame(FLQuants(
                         min=lengths[[1]]-lengths[[2]],
                         max=lengths[[1]]+lengths[[2]])))+
  xlab("Year")+ylab("Mean Length")

source('~/Desktop/sea++/mydas/pkg/R/oemLn.R')

library(reshape)
lh=lhPar(FLPar(linf=89))

ak=alk(lh)  

source('~/Desktop/sea++/mydas/pkg/R/genTime.R')

library(FLBRP)
library(FLasher)

lh=FLPar(c(linf= 59.1,  k=0.28, t0=-0.4, a=0.01111,b=3.15,a50=4.0, l50=43.25),units="NA")

lh=lhPar(lh)  

eq=lhEql(lh)

gTime=c(round(gt(eq)))

fbar(eq)=refpts(eq)["msy","harvest"]%*%FLQuant(c(rep(.1,19),
                                                 seq(.1,2,length.out=30),
                                                 seq(2,1.0,length.out=gTime)[-1],
                                                 rep(1.0,61)))[,1:105]
om=as(eq,"FLStock")

om=fwd(om,f=fbar(om)[,-1], sr=eq)

ggplot(melt(lfd[,seq(1,45,10)]))+
  geom_histogram(aes(length,weight=value),binwidth=1)+
  facet_grid(year~iter,scale="free")+
  xlab("Length (cm)")+ylab("Frequency")+
  coord_cartesian(xlim=c(0,mean(lh["linf"])))

Figure 1 Observation error model for turbot.

Back to Top

More information

• You can submit bug reports, questions or suggestions on FLPKG at the FLPKG issue page,³ or on the FLR mailing list.
• Or send a pull request to https://github.com/flr/FLPKG/
• For more information on the FLR Project for Quantitative Fisheries Science in R, visit the FLR webpage.⁴
• The latest version of FLPKG can always be installed using the devtools package, by calling

    library(devtools)
    install_github('flr/FLPKG')

Software Versions

• R version 3.5.1 (2018-07-02)
• FLCore: 2.6.10.9004
• FLPKG:
• Compiled: Sat Dec 8 13:16:19 2018
• Git Hash: c308b3b

Author information

Laurence Kell. laurie@seaplusplus.es

Acknowledgements

This vignette and many of the methods documented in it were developed under the MyDas project funded by the Irish exchequer and EMFF 2014-2020. The overall aim of MyDas is to develop and test a range of assessment models and methods to establish Maximum Sustainable Yield (MSY) reference points (or proxy MSY reference points) across the spectrum of data-limited stocks.