# A R function for converting to long data to wide data Converting regular survey data (long data, similar to the treedata and herbdata in the package) to species matrix, and attributes matrix (wide data). ```r longtowide(longdata, left, up, inner = NULL, fun = rep("sum", length(inner)), freq = FALSE) ``` ## Arguments - `longdata`: Regular survey data (long data, similar to the treedata and herbdata in the package) - `left`: Row names of matrix data after converted - `up`: Column names of matrix data after converted - `inner`: Attributes data need to converted - `fun`: Functions need to applied in the converting process - `freq`: Frequency of attributes ## Returns Wide data,such species matrix, and attributes matrix, etc. ## Author(s) Zongzheng Chai ## Examples ```r #####Example###### ##Following two example data with different subject, pineoakTree ##is for the base data of individual trees, and the herb data is for ##the base data of cluster. data(treedata) data(herbdata) ####Merge a species matrix from base data ##Note: following calculation only for the base data of individual not cluster pineoak.spematrix<-longtowide(longdata=treedata,left="plot.id",up="spe.id") pineoak.spematrix ##Equivalent method to merge species matrix from base data treedata$abundance<-1 head(treedata) tree.spematrix<-longtowide(longdata=treedata,left="plot.id",up="spe.id",inner="abundance") tree.spematrix herb.spematrix<-longtowide(longdata=herbdata,left="plot.id",up="spe.id",inner="abundance") herb.spematrix ####Compute the frequency of species in the multiple plots ##Merge a frequency matrix from base data tree.freq<-longtowide(longdata=treedata,left="plot.id",up="spe.id",freq=TRUE) tree.freq herb.freq<-longtowide(longdata=herbdata,left="plot.id",up="spe.id",freq=TRUE) herb.freq ####Compute the attribute matrix from base data ##Single attribute matrix tree.dbh<-longtowide(longdata=treedata,left="plot.id",up="spe.id",inner="dbh.cm") tree.dbh herb.cover<-longtowide(longdata=herbdata,left="plot.id",up="spe.id",inner="coverage") herb.cover ##Multiple attribute matrix tree.multattr<-longtowide(longdata=treedata,left="plot.id",up="spe.id", inner=c("dbh.cm","ht.m")) tree.multattr herb.multattr<-longtowide(longdata=herbdata,left="plot.id",up="spe.id", inner=c("coverage","abundance")) herb.multattr #######Calculation of Importance value (IV) for species###### ####Calculation of Importance value (IV) for individual tree data ##Covert long data to wide data tree.freq<-longtowide(longdata=treedata,left="plot.id",up="spe.id",freq=TRUE)$frequency tree.density<-longtowide(longdata=treedata,left="plot.id",up="spe.id")/(20*400) treedata$base<-pi*(treedata$dbh.cm)^2/10000 tree.dominance<-longtowide(longdata=treedata,left="plot.id",up="spe.id",inner="base")/(20*400) ##Compute the relative value tree.refreq<-tree.freq/rowSums(tree.freq)*100 tree.redensity<-tree.density/rowSums(tree.density)*100 tree.redominance<-tree.dominance/rowSums(tree.dominance)*100 ##Compute the importance value matrix treeIVmatirx<-(tree.refreq+tree.redensity+tree.redominance)/3 ##Compute the total importance value for species treeTIV<-colSums(treeIVmatirx)/20 ##Order the total importance value for species sortedtreeTIV<-sort(treeTIV,decreasing=TRUE) sortedtreeTIV ####Calculation of Importance value (IV) for clust herb data ##Covert long data to wide data herb.freq<-longtowide(longdata=herbdata,left="plot.id",up="spe.id",freq=TRUE)$frequency herb.dencover<-longtowide(longdata=herbdata,left="plot.id",up="spe.id", inner=c("abundance","coverage")) herb.density<-herb.dencover$abundance/100 herb.coverage<-herb.dencover$coverage/100 ##Compute the relative value herb.refreq<-herb.freq/rowSums(herb.freq)*100 herb.redensity<-herb.density/rowSums(herb.density)*100 herb.recoverage<-herb.coverage/rowSums(herb.coverage)*100 ##Compute the importance value matrix herbIVmatirx<-(herb.refreq+herb.redensity+herb.recoverage)/3 ##Compute the total importance value for species herbTIV<-colSums(herbIVmatirx)/20 ##Order the total importance value for species sortedherbTIV<-sort(herbTIV,decreasing=TRUE) sortedtreeTIV ```

longtowide function