minor changes to Fushan and NZ in accordance with milestones

merge.data.FUSHAN.R

@@ -26,7 +26,7 @@ data.fushan$D <- data.fushan[["dbh1"]] ## diameter in cm
 data.fushan$dead <- as.numeric(data.fushan$status2 == "dead")  ## dummy variable for dead tree 0 alive 1 dead
 data.fushan$plot <- rep(NA,nrow(data.fushan))  ## plot code - MISSING
 data.fushan$htot <- rep(NA,nrow(data.fushan)) ## height of tree in m - MISSING
-data.fushan$tree.id <- data.fushan$tag  ## tree unique id
+data.fushan$tree.id <- data.fushan$tag  ## tree unique id - use tag here as it is unique for each row
 data.fushan$sp.name <- rep(NA,nrow(data.fushan)) 
 v <- species.clean$sp
 for (i in 1:length(unique(data.fushan$sp))) {

merge.data.NZ.R

@@ -3,42 +3,48 @@ rm(list = ls())
 source("./R/format.function.R")
 library(reshape)
 
-######################### READ DATA read individuals tree data
+######################### READ DATA read individuals tree and environmental data
 data.nz <- read.csv("./data/raw/DataNVS/nz_treedata_growth_130801.csv", header = TRUE, 
     stringsAsFactors = FALSE, skip = 9)
 data.nz <- data.nz[, -1]
-data.trait <- read.csv("./data/raw/DataNVS/nz_traits_130801.csv", , header = TRUE, 
-    stringsAsFactors = FALSE)
-data.trait <- data.trait[, -1]
+data.nz$plid <- gsub("__", "_", data.nz$plid)
+data.nz$plid <- gsub("_", ".", data.nz$plid)  ## Replace all underscores with a single dot
+
 data.plot <- read.csv("./data/raw/DataNVS/nz_plotinfo_130801.csv", , header = TRUE, 
     stringsAsFactors = FALSE)
 data.plot <- data.plot[, -1]
-colnames(data.trait)[1] <- "sp"
 data.plot$plid <- gsub("__", "_", data.plot$plid)
 data.plot$plid <- gsub("_", ".", data.plot$plid)  ## Replace all underscores with a single dot
-data.nz$plid <- gsub("__", "_", data.nz$plid)
-data.nz$plid <- gsub("_", ".", data.nz$plid)  ## Replace all underscores with a single dot
 
+data.nz <- merge(data.nz, data.frame(plid = data.plot$plid, Easting = data.plot$Easting, 
+	Northing = data.plot$Northing, Locality = data.plot$Locality, MAP = data.plot$map, MAT = data.plot$mat, Broadclass = data.plot$BroadClass, 
+	Specificclass = data.plot$SpecificClass, stringsAsFactors = F), sort = T, by = "plid")
+rm(data.plot)
+	
 ###################################### MASSAGE TRAIT DATA Maximum height per species is already available from
 ###################################### data.trait (in m); so no sd's and only obs per spp
+data.trait <- read.csv("./data/raw/DataNVS/nz_traits_130801.csv", , header = TRUE, 
+    stringsAsFactors = FALSE)
+data.trait <- data.trait[, -1]
+colnames(data.trait)[1] <- "sp"
+
 data.max.height <- data.frame(code = data.trait[["sp"]], Max.height.mean = log10(data.trait[["height.m"]]), 
     Max.height.sd = NA, Max.height.nobs = 1)
-# write.csv(data.max.height,file='./data/process/data.max.height.nz.csv') ## I
-# was planning to save processed data in that folder
+# write.csv(data.max.height,file='./data/process/data.max.height.nz.csv') 
 
 ################################################################ FORMAT INDIVIDUAL TREE DATA
 
 ## change unit and names of variables to be the same in all data for the tree
 data.nz$G <- 10 * (data.nz[["D1"]] - data.nz[["D0"]])/(data.nz[["t1"]] - data.nz[["t0"]])  ## diameter growth in mm per year
 data.nz$year <- (data.nz[["t1"]] - data.nz[["t0"]])  ## number of year between measurement
-data.nz$D <- data.nz[["D0"]]  ## diameter in mm convert to cm
+data.nz$D <- data.nz[["D0"]]  ## diameter in cm
 data.nz$dead <- as.numeric(is.na(data.nz[["D1"]]))  ## dummy variable for dead tree 0 alive 1 dead
-data.nz$sp <- data.nz$sp
-data.nz$plot <- data.nz$plid
-data.nz$htot <- rep(NA, length(data.nz[["sp"]]))  ## Max height is already available so have as missing
-data.nz$tree.id <- data.nz[["tag"]]
-data.nz$tree.id <- gsub("__", "_", data.nz$tree.id)
-data.nz$tree.id <- gsub("_", ".", data.nz$tree.id)  ## tree unique id
+data.nz$plot <- data.nz$plid; data.nz$plid <- NULL
+data.nz$htot <- rep(NA, nrow(data.nz))  ## Max height is already available so have as missing
+data.nz$tree.id <- rep(NA, nrow(data.nz)) 
+#data.nz$tree.id <- gsub("__", "_", data.nz$tree.id)
+#data.nz$tree.id <- gsub("_", ".", data.nz$tree.id)  ## tree unique id
+data.nz$weights <- 1/(pi*(0.5*data.nz$D/100)^2)
 
 ########################################## CHANGE COORDINATE SYSTEM DON'T KNOW THE EPSG CODE HERE change coordinates
 ########################################## system of Easting Northing to be in lat long WGS84
@@ -60,21 +66,14 @@ data.nz$Lat <- coordinates(data.sp2)[, "Northing"]
 rm(data.sp, data.sp2)
 
 ###################### ECOREGION merge greco to have no ecoregion with low number of observation
-data.nz <- merge(data.nz, data.frame(plot = data.plot$plid, data.plot[, 11:12]), 
-    by = "plot")
 table(data.nz$Broad)
 
 ###################### PERCENT DEAD
-perc.dead <- tapply(data.nz[["dead"]], INDEX = data.nz[["plot"]], FUN = function.perc.dead)
-# ## VARIABLE TO SELECT PLOT WITH NOT BIG DISTURBANCE KEEP OFTHER VARIABLES IF
-# AVAILABLE (disturbance record)
+perc.dead <- tapply(data.nz[["dead"]], INDEX = data.nz[["plot"]], FUN = function.perc.dead2)
 data.nz <- merge(data.nz, data.frame(plot = names(perc.dead), perc.dead = perc.dead), 
     by = "plot", sort = FALSE)
 
 ########################################################### PLOT SELECTION FOR THE ANALYSIS
-data.nz <- merge(data.nz, data.plot[, c(1, 8:10)], by = "plid")
-colnames(data.nz)[colnames(data.nz) %in% c("mat", "map")] <- c("MAT", "MAP")
-# colnames(data.nz)[names(data.nz) =='eco_codemerged' ] <- c('ecocode')
 vec.abio.var.names <- c("MAT", "MAP")
 vec.basic.var <- c("tree.id", "sp", "spname", "plot", "ecocode", "D", "G", "dead", 
     "year", "htot", "Lon", "Lat", "perc.dead")