minor changes to NSW and Swiss in accordance with milestones

merge.data.NSW.R

@@ -52,24 +52,24 @@ data.traits <- read.csv("./data/raw/DataNSW/NSW_traits.csv", header = TRUE, stri
 
 ########################################## FORMAT INDIVIDUAL TREE DATA Each tree has at most 3 observations (from prelim
 ########################################## checks of the data)
-data.nsw$treeid <- apply(data.nsw[, 1:2], 1, paste, collapse = ".")
+data.nsw$treeid <- apply(data.nsw[, 1:2], 1, paste, collapse = "_")
 data.nsw2 <- data.frame(data.nsw[1, ], year1 = NA, year2 = NA, dbh1 = NA, dbh2 = NA)
 for (k in 1:length(unique(data.nsw$treeid))) {
     sub.datansw <- as.data.frame(data.nsw[which(data.nsw$treeid == unique(data.nsw$treeid)[k]), 
         ])
     if (nrow(sub.datansw) == 1) {
         data.nsw2 <- rbind(data.nsw2, data.frame(sub.datansw, year1 = sub.datansw$Date.of.measure[1], 
-            year2 = NA, dbh1 = sub.datansw$Dbh[1], dbh2 = NA))
+            year2 = NA, dbh1 = sub.datansw$Dbh[1], dbh2 = NA, stringsAsFactors = F))
     }
     if (nrow(sub.datansw) == 2) {
         data.nsw2 <- rbind(data.nsw2, data.frame(sub.datansw[1, ], year1 = sub.datansw$Date.of.measure[1], 
-            year2 = sub.datansw$Date.of.measure[2], dbh1 = sub.datansw$Dbh[1], dbh2 = sub.datansw$Dbh[2]))
+            year2 = sub.datansw$Date.of.measure[2], dbh1 = sub.datansw$Dbh[1], dbh2 = sub.datansw$Dbh[2], stringsAsFactors = F))
     }
     if (nrow(sub.datansw) == 3) {
         data.nsw2 <- rbind(data.nsw2, data.frame(sub.datansw[1, ], year1 = sub.datansw$Date.of.measure[1], 
-            year2 = sub.datansw$Date.of.measure[2], dbh1 = sub.datansw$Dbh[1], dbh2 = sub.datansw$Dbh[2]))
+            year2 = sub.datansw$Date.of.measure[2], dbh1 = sub.datansw$Dbh[1], dbh2 = sub.datansw$Dbh[2], stringsAsFactors = F))
         data.nsw2 <- rbind(data.nsw2, data.frame(sub.datansw[1, ], year1 = sub.datansw$Date.of.measure[2], 
-            year2 = sub.datansw$Date.of.measure[3], dbh1 = sub.datansw$Dbh[2], dbh2 = sub.datansw$Dbh[3]))
+            year2 = sub.datansw$Date.of.measure[3], dbh1 = sub.datansw$Dbh[2], dbh2 = sub.datansw$Dbh[3], stringsAsFactors = F))
     }
 }
 data.nsw2 <- data.nsw2[-1, ]
@@ -80,13 +80,11 @@ for (k in 9:12) data.nsw[, k] <- as.numeric(data.nsw[, k])
 ## change unit and names of variables to be the same in all data for the tree
 data.nsw$year <- (data.nsw$year2 - data.nsw$year1)  ## number of year between measurements
 data.nsw$G <- 10 * (data.nsw$dbh2 - data.nsw$dbh1)/(data.nsw$year)  ## diameter growth in mm per year
-## THERE ARE SOME ROWS WITH STRONG NEGATIVE GROWTH THAT YOU MIGHT WANT TO REMOVE
-head(data.nsw[order(data.nsw$G), ])
-
-data.nsw$D <- data.nsw[["dbh1"]]
-## diameter in cm
+#head(data.nsw[order(data.nsw$G), ])
+data.nsw$D <- data.nsw[["dbh1"]] ## diameter in cm
 data.nsw$dead <- rep(NA, nrow(data.nsw))  ## dummy variable for dead tree 0 alive 1 dead - MISSING
-data.nsw$sp <- as.character(data.nsw[["species"]])  ## species code - use the spp name as code
+data.nsw$sp <- as.numeric(factor(data.nsw[["species"]]))  ## species code - use the spp name as code
+data.nsw$sp.name <- data.nsw[["species"]]; data.nsw$species <- NULL
 data.nsw$plot <- as.character(data.nsw[["Plot"]])  ## plot code
 data.nsw$htot <- rep(NA, nrow(data.nsw))  ## height of tree in m - MISSING
 ### add plot weights for computation of competition index (in 1/m^2) - from the
@@ -103,11 +101,14 @@ data.nsw$weights[grep("BR", data.nsw$Plot)] <- 1/(60.4 * 60.4)
 data.nsw$weights[grep("END", data.nsw$Plot)] <- 1/(40 * 50)
 data.nsw$weights[grep("TND", data.nsw$Plot)] <- 1/(40 * 50)
 
-data.nsw$obs.id <- 1:nrow(data.nsw)
+data.nsw$obs.id <- 1:nrow(data.nsw) ## Should change this when census variable is created
 
 ###################### ECOREGION nsw has only 1 eco-region
 
 ###################### PERCENT DEAD NO DATA ON MORTALITY
+perc.dead <- tapply(data.nsw[["dead"]], INDEX = data.nsw[["plot"]], FUN = function.perc.dead2)
+data.nsw <- merge(data.nsw, data.frame(plot = names(perc.dead), perc.dead = perc.dead), 
+    by = "plot", sort = FALSE)
 
 ########################################### VARIABLES SELECTION FOR THE ANALYSIS
 vec.abio.var.names <- c("MAT", "MAP")  ## MISSING

merge.data.SWISS.R

-### MERGE Swiss DATA Edited by FH
+### MERGE Swiss DATA
 rm(list = ls())
 source("./R/format.function.R")
 source("./R/FUN.TRY.R")
@@ -46,14 +46,12 @@ data.swiss$Lat <- coordinates(data.sp2)[, "y"]
 ## points(data.swiss$x,data.swiss$y)
 ## rm(data.sp, data.sp2)
 
-
-
 ## Do not need to read in spp list as it is already available in data.swiss
 
 ###################################### MASSAGE TRAIT DATA Compute maximum height per species plus sd from observed
 ###################################### height to add variables to the traits data base Because we have two heights,
 ###################################### then take the max of the two heights and then bootstrap
-res.quant.boot <- t(sapply(levels(factor(data.swiss[["spcode"]])), FUN = f.quantile.boot, 
+res.quant.boot <- t(sapply(levels(factor(data.swiss[["sp"]])), FUN = f.quantile.boot, 
     R = 1000, x = log10(apply(data.swiss[, c("ht1", "ht2")], 1, max, na.rm = T)), 
     fac = factor(data.swiss[["spcode"]])))
 
@@ -65,33 +63,32 @@ write.csv(data.max.height,file='./data/process/data.max.height.swiss.csv')
 
 ########################################## FORMAT INDIVIDUAL TREE DATA change unit and names of variables to be the same
 ########################################## in all data for the tree
-data.swiss$G <- 10 * (data.swiss$dbh_diff)/data.swiss$year  ## diameter growth in mm per year - SOME EXTREMELY NEGATIVE HERE!
+data.swiss$G <- 10 * (data.swiss$dbh_diff)/data.swiss$year  ## diameter growth in mm per year
 data.swiss$D <- data.swiss[["dbh1"]]
-data.swiss$D[data.swiss$D == 0] <- NA
-## diameter in cm
-data.swiss$dead <- rep(NA, length(data.swiss[["dbh1"]]))  ## Mortality - MISSING
-data.swiss$plot <- data.swiss$siteid  ## plot code
+data.swiss$D[data.swiss$D == 0] <- NA ## diameter in cm
+data.swiss$dead <- rep(NA, nrow(data.swiss))  ## Mortality - MISSING
+data.swiss$plot.id <- data.swiss$siteid; data.swiss$siteid <- NULL ## plot code
 data.swiss$htot <- data.swiss$ht1  ## height of tree in m 
 data.swiss$census <- data.swiss$Inventid
-data.swiss$obs.id <- as.character(1:nrow(data.swiss))
+data.swiss$obs.id <- apply(data.swiss[,c("tree.id","Inventid")],1,paste,collapse="_")
 data.swiss$weights <- data.swiss$repfactor1/10000
 data.swiss$sp <- paste("sp.",data.swiss$sp,sep="")
+
 ###################### ECOREGION Ecoregion not available for swiss data
 data.swiss$ecocode <- rep("A", nrow(data.swiss))
+
 ###################### 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.swiss[["dead"]], INDEX = data.swiss[["plot"]], FUN = function.perc.dead)
+perc.dead <- tapply(data.swiss[["dead"]], INDEX = data.swiss[["plot"]], FUN = function.perc.dead2)
 # ## VARIABLE TO SELECT PLOT WITH NOT BIG DISTURBANCE KEEP OTHER VARIABLES IF
 # AVAILABLE (disturbance record)
 data.swiss <- merge(data.swiss, data.frame(plot = names(perc.dead), perc.dead = perc.dead, stringsAsFactors =FALSE), 
     by = "plot", sort = FALSE)
 
-########################################################### PLOT SELECTION FOR THE ANALYSIS Remove data with dead == 1
-## Nothing to remove
-
+########################################################### PLOT SELECTION FOR THE ANALYSIS
 data.climate <- read.dbf(file = "./data/raw/DataSwiss/LFI14_climate.dbf")
-data.climate <- data.climate[, c(1, 7, 15:19)]
-data.climate$MAP <- apply(data.climate[, 4:7], 1, sum)
+data.climate <- data.climate[, c("CLNR","swb_100","tave_68","prec_122","prec_35","prec_68","prec_911")]
+data.climate$MAP <- apply(data.climate[, c("prec_122","prec_35","prec_68","prec_911")], 1, sum)
 
 data.swiss <- merge(data.swiss, data.frame(siteid = data.climate$CLNR, swb = data.climate$swb_100, 
     MAT = data.climate$tave_68, MAP = data.climate$MAP, stringsAsFactors =FALSE), sort = F, all.x = T)
@@ -100,7 +97,6 @@ rm(data.climate)
 ############################################## 
 ############################################## 
 #################### GENERATE ONE OBJECT PER ECOREGION
-
 # vector of ecoregion name
 ecoregion.unique <- unique(data.swiss[["ecocode"]])