merge.data.US.R

### MERGE us DATA
### Edited by FH
rm(list = ls()); source("./R/format.function.R"); library(reshape)

#########################
## READ DATA
####################
### read individuals tree data
data.us <- read.csv("./data/raw/DataUS/FIA51_trees_w_supp.csv",header=TRUE,stringsAsFactors =FALSE)

### read species names
species.clean <- read.csv("./data/species.list/REF_SPECIES.CSV",stringsAsFactors=FALSE)

######################################
## MASSAGE TRAIT DATA
############################
## HEIGHT DATA FOR TREE MISSING
## BRING US DATA FOR HEIGHT OVER WHEN WE ANALYZE THAT DATASET LATER ON

#####################################
## FORMAT INDIVIDUAL TREE DATA
#############

## change unit and names of variables to be the same in all data for the tree 
data.us$G <- 10*(data.us$FinalDbh-data.us$InitDbh)/data.us$Interval ## diameter growth in mm per year
data.us$G[which(data.us$InitDbh == 0 | data.us$FinalDbh == -999)] <- NA
data.us$year <- data.us$Interval ## number of year between measuremen
data.us$D <- data.us[["InitDbh"]]; data.us$D[data.us$D == 0] <- NA ;## diameter in cm
data.us$dead <- as.numeric(data.us$FinalDbh > 0) ## dummy variable for dead tree 0 alive 1 dead
data.us$sp <- as.character(data.us[["Species"]]) ## species code
data.us$plot <- as.character(data.us[["PlotID"]]) ## plot code
data.us$subplot <- paste(as.character(data.us[["PlotID"]]),as.character(data.us[["SubplotNumber"]]),sep=".") ## plot code
data.us$htot <- rep(NA,length(data.us[["Species"]])) ## height of tree in m - MISSING
data.us$tree.id <- as.character(data.us$TreeID); ## tree unique id
data.us$sp.name <- NA; 
## v <- species.clean$SPCD; for(i in 1:length(unique(data.us$sp))) { 
## 	data.us$sp.name[which(data.us$sp == unique(data.us$sp)[i])] <- species.clean$COMMON_NAME[which(v == unique(data.us$sp)[i])] }
	
######################
## ECOREGION
###################
## merge greco to have no ecoregion with low number of observation
greco <- read.csv(file = "./data/raw/DataUS/EcoregionCodes.csv", header = T); colnames(greco)[1] <- "Ecocode" 

table(data.us$Ecocode) 
data.us <- merge(data.us, greco[,-4], by = "Ecocode"); data.us$DIVISION <- factor(data.us$DIVISION)
## Some ecoregions still have small # of individuals, so create a variable which does division if # ind < 10000; else it reads Domain
#
data.us$eco_codemerged <- as.character(data.us$DIVISION);  tab.small.div <- table(data.us$eco_codemerged)
sel.small.div <- which(table(data.us$eco_codemerged) < 10000)
for(i in 1:length(sel.small.div)) { 
	find.ind <- which(data.us$eco_codemerged == names(tab.small.div)[sel.small.div[i]]); print(length(find.ind))
	data.us$eco_codemerged[find.ind] <- as.character(data.us$DOMAIN)[find.ind] 
	}
	
data.us <- data.us[,-c(2,24,25)] ## Remove other ecocode related stuff to save space
######################
## PERCENT DEAD
###################
## variable percent dead/cannot do with since dead variable is missing
## compute numer of dead per plot to remove plot with disturbance
perc.dead <- tapply(data.us[["dead"]],INDEX=data.us[["plot"]],FUN=function.perc.dead)
# ## VARIABLE TO SELECT PLOT WITH NOT BIG DISTURBANCE KEEP OFTHER VARIABLES IF AVAILABLE (disturbance record)
data.us <- merge(data.us,data.frame(plot=names(perc.dead),perc.dead=perc.dead), by = "plot", sort=FALSE)

###########################################################
### PLOT SELECTION FOR THE ANALYSIS
###################
## Remove data with dead == 1
table(data.us$dead)