##################    MULTI SAMPLING GAM FUSION METHOD ASSESSMENT ####################################

############################ Merging LST and station data ##########################################

#This script interpolates tmax values using MODIS LST and GHCND station data                      

#interpolation area. It requires the text file of stations and a shape file of the study area.           

#Note that the projection for both GHCND and study area is lonlat WGS84.       

#Options to run this program are:

#1) Multisampling: vary the porportions of hold out and use random samples for each run

#2)Constant sampling: use the same sample over the runs

#3)over dates: run over for example 365 dates without mulitsampling

#4)use seed number: use seed if random samples must be repeatable

#5)GAM fusion: possibilty of running GAM+FUSION or GAM separately 

#AUTHOR: Benoit Parmentier                                                                        

#DATE: 12/27/2012                                                                                 

#PROJECT: NCEAS INPLANT: Environment and Organisms --TASK#363--                                   

###################################################################################################

###Loading R library and packages                                                      

library(gtools)                                         # loading some useful tools 

library(mgcv)                                           # GAM package by Simon Wood

library(sp)                                             # Spatial pacakge with class definition by Bivand et al.

library(spdep)                               # Spatial pacakge with methods and spatial stat. by Bivand et al.

library(rgdal)                               # GDAL wrapper for R, spatial utilities

library(gstat)                               # Kriging and co-kriging by Pebesma et al.

library(fields)                              # NCAR Spatial Interpolation methods such as kriging, splines

library(raster)                              # Hijmans et al. package for raster processing

library(rasterVis)

library(parallel)                            # Urbanek S. and Ripley B., package for multi cores & parralel processing

library(reshape)

library(plotrix)

### Parameters and argument

infile1<- "ghcn_or_tmax_covariates_06262012_OR83M.shp"             #GHCN shapefile containing variables for modeling 2010                 

infile2<-"list_365_dates_04212012.txt"

infile3<-"LST_dates_var_names.txt"                        #LST dates name

infile4<-"models_interpolation_05142012.txt"              #Interpolation model names

infile5<-"mean_day244_rescaled.rst"                       #Raster or grid for the locations of predictions

#infile6<-"lst_climatology.txt"

infile6<-"LST_files_monthly_climatology.txt"

inlistf<-"list_files_05032012.txt"                        #Stack of images containing the Covariates

path<-"/home/parmentier/Data/IPLANT_project/data_Oregon_stations_10242012_GAM"

setwd(path)

#Station location of the study area

stat_loc<-read.table(paste(path,"/","location_study_area_OR_0602012.txt",sep=""),sep=",", header=TRUE)

#GHCN Database for 1980-2010 for study area (OR) 

data3<-read.table(paste(path,"/","ghcn_data_TMAXy1980_2010_OR_0602012.txt",sep=""),sep=",", header=TRUE)

nmodels<-9   #number of models running

y_var_name<-"dailyTmax"

predval<-1

prop<-0.3             #Proportion of testing retained for validation   

#prop<-0.25

seed_number<- 100  #if seed zero then no seed?                                                                 #Seed number for random sampling

out_prefix<-"_365d_GAM_fus5_all_lstd_12302012"                #User defined output prefix

#out_prefix<-"_365d_GAM_12272012"                #User defined output prefix

bias_val<-0            #if value 1 then training data is used in the bias surface rather than the all monthly stations

bias_prediction<-1     #if value 1 then use GAM for the BIAS prediction otherwise GAM direct repdiction for y_var (daily tmax)

nb_sample<-1           #number of time random sampling must be repeated for every hold out proportion

prop_min<-0.3          #if prop_min=prop_max and step=0 then predicitons are done for the number of dates...

prop_max<-0.3

step<-0         

constant<-0             #if value 1 then use the same samples as date one for the all set of dates

#projection used in the interpolation of the study area

CRS_interp<-"+proj=lcc +lat_1=43 +lat_2=45.5 +lat_0=41.75 +lon_0=-120.5 +x_0=400000 +y_0=0 +ellps=GRS80 +units=m +no_defs";

CRS_locs_WGS84<-CRS("+proj=longlat +ellps=WGS84 +datum=WGS84 +towgs84=0,0,0") #Station coords WGS84

source("GAM_fusion_function_multisampling_12302012.R")

###################### START OF THE SCRIPT ########################

###Reading the station data and setting up for models' comparison

filename<-sub(".shp","",infile1)             #Removing the extension from file.

ghcn<-readOGR(".", filename)                 #reading shapefile 

CRS<-proj4string(ghcn)                       #Storing projection information (ellipsoid, datum,etc.)

mean_LST<- readGDAL(infile5)                 #Reading the whole raster in memory. This provides a grid for kriging

proj4string(mean_LST)<-CRS                   #Assigning coordinate information to prediction grid.

ghcn <- transform(ghcn,Northness = cos(ASPECT*pi/180)) #Adding a variable to the dataframe

ghcn <- transform(ghcn,Eastness = sin(ASPECT*pi/180))  #adding variable to the dataframe.

ghcn <- transform(ghcn,Northness_w = sin(slope*pi/180)*cos(ASPECT*pi/180)) #Adding a variable to the dataframe

ghcn <- transform(ghcn,Eastness_w = sin(slope*pi/180)*sin(ASPECT*pi/180))  #adding variable to the dataframe.

#Remove NA for LC and CANHEIGHT

ghcn$LC1[is.na(ghcn$LC1)]<-0

ghcn$LC3[is.na(ghcn$LC3)]<-0

ghcn$CANHEIGHT[is.na(ghcn$CANHEIGHT)]<-0

ghcn$LC4[is.na(ghcn$LC4)]<-0

ghcn$LC6[is.na(ghcn$LC6)]<-0

dates <-readLines(paste(path,"/",infile2, sep=""))

LST_dates <-readLines(paste(path,"/",infile3, sep=""))

models <-readLines(paste(path,"/",infile4, sep=""))

##Extracting the variables values from the raster files                                             

lines<-read.table(paste(path,"/",inlistf,sep=""), sep=" ")                  #Column 1 contains the names of raster files

inlistvar<-lines[,1]

inlistvar<-paste(path,"/",as.character(inlistvar),sep="")

covar_names<-as.character(lines[,2])                                         #Column two contains short names for covaraites

s_raster<- stack(inlistvar)                                                  #Creating a stack of raster images from the list of variables.

layerNames(s_raster)<-covar_names                                            #Assigning names to the raster layers

projection(s_raster)<-CRS

#stat_val<- extract(s_raster, ghcn3)                                          #Extracting values from the raster stack for every point location in coords data frame.

pos<-match("ASPECT",layerNames(s_raster)) #Find column with name "value"

r1<-raster(s_raster,layer=pos)             #Select layer from stack

pos<-match("slope",layerNames(s_raster)) #Find column with name "value"

r2<-raster(s_raster,layer=pos)             #Select layer from stack

N<-cos(r1*pi/180)

E<-sin(r1*pi/180)

Nw<-sin(r2*pi/180)*cos(r1*pi/180)   #Adding a variable to the dataframe

Ew<-sin(r2*pi/180)*sin(r1*pi/180)   #Adding variable to the dataframe.

pos<-match("LC1",layerNames(s_raster)) #Find column with name "value"

LC1<-raster(s_raster,layer=pos)             #Select layer from stack

s_raster<-dropLayer(s_raster,pos)

LC1[is.na(LC1)]<-0

pos<-match("LC3",layerNames(s_raster)) #Find column with name "value"

LC3<-raster(s_raster,layer=pos)             #Select layer from stack

s_raster<-dropLayer(s_raster,pos)

LC3[is.na(LC3)]<-0

pos<-match("LC4",layerNames(s_raster)) #Find column with name "value"

LC4<-raster(s_raster,layer=pos)             #Select layer from stack

s_raster<-dropLayer(s_raster,pos)

LC4[is.na(LC4)]<-0

pos<-match("LC6",layerNames(s_raster)) #Find column with name "value"

LC6<-raster(s_raster,layer=pos)             #Select layer from stack

s_raster<-dropLayer(s_raster,pos)

LC6[is.na(LC6)]<-0

LC_s<-stack(LC1,LC3,LC4,LC6)

layerNames(LC_s)<-c("LC1_forest","LC3_grass","LC4_crop","LC6_urban")

plot(LC_s)

pos<-match("CANHEIGHT",layerNames(s_raster)) #Find column with name "value"

CANHEIGHT<-raster(s_raster,layer=pos)             #Select layer from stack

s_raster<-dropLayer(s_raster,pos)

CANHEIGHT[is.na(CANHEIGHT)]<-0

pos<-match("ELEV_SRTM",layerNames(s_raster)) #Find column with name "ELEV_SRTM"

ELEV_SRTM<-raster(s_raster,layer=pos)             #Select layer from stack on 10/30

s_raster<-dropLayer(s_raster,pos)

ELEV_SRTM[ELEV_SRTM <0]<-NA

xy<-coordinates(r1)  #get x and y projected coordinates...

xy_latlon<-project(xy, CRS, inv=TRUE) # find lat long for projected coordinats (or pixels...)

lon<-raster(xy_latlon) #Transform a matrix into a raster object ncol=ncol(r1), nrow=nrow(r1))

ncol(lon)<-ncol(r1)

nrow(lon)<-nrow(r1)

extent(lon)<-extent(r1)

projection(lon)<-CRS  #At this stage this is still an empty raster with 536 nrow and 745 ncell 

lat<-lon

values(lon)<-xy_latlon[,1]

values(lat)<-xy_latlon[,2]

r<-stack(N,E,Nw,Ew,lon,lat,LC1,LC3,LC4,LC6, CANHEIGHT,ELEV_SRTM)

rnames<-c("Northness","Eastness","Northness_w","Eastness_w", "lon","lat","LC1","LC3","LC4","LC6","CANHEIGHT","ELEV_SRTM")

layerNames(r)<-rnames

s_raster<-addLayer(s_raster, r)

#s_sgdf<-as(s_raster,"SpatialGridDataFrame") #Conversion to spatial grid data frame

####### Preparing LST stack of climatology...

#l=list.files(pattern="mean_month.*rescaled.rst")

l <-readLines(paste(path,"/",infile6, sep=""))

molst<-stack(l)  #Creating a raster stack...

#setwd(old)

molst<-molst-273.16  #K->C          #LST stack of monthly average...

idx <- seq(as.Date('2010-01-15'), as.Date('2010-12-15'), 'month')

molst <- setZ(molst, idx)

layerNames(molst) <- month.abb

######  Preparing tables for model assessment: specific diagnostic/metrics

#Model assessment: specific diagnostics/metrics

results_AIC<- matrix(1,1,nmodels+3)  

results_GCV<- matrix(1,1,nmodels+3)

results_DEV<- matrix(1,1,nmodels+3)

#results_RMSE_f<- matrix(1,length(models)+3)

#Model assessment: general diagnostic/metrics 

results_RMSE <- matrix(1,1,nmodels+4)

results_MAE <- matrix(1,1,nmodels+4)

results_ME <- matrix(1,1,nmodels+4)       #There are 8+1 models

results_R2 <- matrix(1,1,nmodels+4)       #Coef. of determination for the validation dataset

results_RMSE_f<- matrix(1,1,nmodels+4)    #RMSE fit, RMSE for the training dataset

results_MAE_f <- matrix(1,1,nmodels+4)

######## Preparing monthly averages from the ProstGres database and extracting covarvariates from stack

# do this work outside of (before) this function

# to avoid making a copy of the data frame inside the function call

date1<-ISOdate(data3$year,data3$month,data3$day) #Creating a date object from 3 separate column

date2<-as.POSIXlt(as.Date(date1))

data3$date<-date2

d<-subset(data3,year>=2000 & mflag=="0" ) #Selecting dataset 2000-2010 with good quality: 193 stations

#May need some screeing??? i.e. range of temp and elevation...

d1<-aggregate(value~station+month, data=d, mean)  #Calculate monthly mean for every station in OR

id<-as.data.frame(unique(d1$station))     #Unique station in OR for year 2000-2010: 193 but 7 loss of monthly avg    

dst<-merge(d1, stat_loc, by.x="station", by.y="STAT_ID")   #Inner join all columns are retained

#This allows to change only one name of the data.frame

pos<-match("value",names(dst)) #Find column with name "value"

names(dst)[pos]<-c("TMax")

dst$TMax<-dst$TMax/10                #TMax is the average max temp for monthy data

#dstjan=dst[dst$month==9,]  #dst contains the monthly averages for tmax for every station over 2000-2010

#Extracting covariates from stack

coords<- dst[c('lon','lat')]              #Define coordinates in a data frame

coordinates(dst)<-coords                      #Assign coordinates to the data frame

proj4string(dst)<-CRS_locs_WGS84                  #Assign coordinates reference system in PROJ4 format

dst_month<-spTransform(dst,CRS(CRS_interp))     #Project from WGS84 to new coord. system

stations_val<-extract(s_raster,dst_month)  #extraction of the infomration at station location

stations_val<-as.data.frame(stations_val)

dst_extract<-cbind(dst_month,stations_val)

dst<-dst_extract

#Now clean and screen monthly values

dst_all<-dst

dst<-subset(dst,dst$TMax>-15 & dst$TMax<40)

dst<-subset(dst,dst$ELEV_SRTM>0) #This will drop two stations...or 24 rows

######### Preparing daily values for training and testing

#Screening for bad values: value is tmax in this case

#ghcn$value<-as.numeric(ghcn$value)

ghcn_all<-ghcn

ghcn_test<-subset(ghcn,ghcn$value>-150 & ghcn$value<400)

ghcn_test2<-subset(ghcn_test,ghcn_test$ELEV_SRTM>0)

ghcn<-ghcn_test2

#coords<- ghcn[,c('x_OR83M','y_OR83M')]

##Sampling: training and testing sites.

if (seed_number>0) {

  set.seed(seed_number)                        #Using a seed number allow results based on random number to be compared...

}

nel<-length(dates)

dates_list<-vector("list",nel) #list of one row data.frame

prop_range<-(seq(from=prop_min,to=prop_max,by=step))*100     #range of proportion to run

sn<-length(dates)*nb_sample*length(prop_range)               #Number of samples to run

for(i in 1:length(dates)){

  d_tmp<-rep(dates[i],nb_sample*length(prop_range)) #repeating same date

  s_nb<-rep(1:nb_sample,length(prop_range))         #number of random sample per proportion

  prop_tmp<-sort(rep(prop_range, nb_sample))

  tab_run_tmp<-cbind(d_tmp,s_nb,prop_tmp)

  dates_list[[i]]<-tab_run_tmp

}

sampling_dat<-as.data.frame(do.call(rbind,dates_list))

names(sampling_dat)<-c("date","run_samp","prop")

for(i in 2:3){            # start of the for loop #1

  sampling_dat[,i]<-as.numeric(as.character(sampling_dat[,i]))  

}

sampling_dat$date<- as.character(sampling_dat[,1])

#ghcn.subsets <-lapply(dates, function(d) subset(ghcn, date==d)) #this creates a list of 10 or 365 subsets dataset based on dates

ghcn.subsets <-lapply(as.character(sampling_dat$date), function(d) subset(ghcn, date==d)) #this creates a list of 10 or 365 subsets dataset based on dates

## adding choice of constant sample 

if (seed_number>0) {

  set.seed(seed_number)                        #Using a seed number allow results based on random number to be compared...

}

sampling<-vector("list",length(ghcn.subsets))

sampling_station_id<-vector("list",length(ghcn.subsets))

for(i in 1:length(ghcn.subsets)){

  n<-nrow(ghcn.subsets[[i]])

  prop<-(sampling_dat$prop[i])/100

  ns<-n-round(n*prop)   #Create a sample from the data frame with 70% of the rows

  nv<-n-ns              #create a sample for validation with prop of the rows

  ind.training <- sample(nrow(ghcn.subsets[[i]]), size=ns, replace=FALSE) #This selects the index position for 70% of the rows taken randomly

  ind.testing <- setdiff(1:nrow(ghcn.subsets[[i]]), ind.training)

  #Find the corresponding 

  data_sampled<-ghcn.subsets[[i]][ind.training,] #selected the randomly sampled stations

  station_id.training<-data_sampled$station     #selected id for the randomly sampled stations (115)

  #Save the information

  sampling[[i]]<-ind.training

  sampling_station_id[[i]]<- station_id.training

}

## Use same samples across the year...

if (constant==1){

  sampled<-sampling[[1]]

  data_sampled<-ghcn.subsets[[1]][sampled,] #selected the randomly sampled stations

  station_sampled<-data_sampled$station     #selected id for the randomly sampled stations (115)

  list_const_sampling<-vector("list",sn)

  list_const_sampling_station_id<-vector("list",sn)

  for(i in 1:sn){

    station_id.training<-intersect(station_sampled,ghcn.subsets[[i]]$station)

    ind.training<-match(station_id.training,ghcn.subsets[[i]]$station)

    list_const_sampling[[i]]<-ind.training

    list_const_sampling_station_id[[i]]<-station_id.training

  }

  sampling<-list_const_sampling 

  sampling_station_id<-list_const_sampling_station_id

}

######## Prediction for the range of dates and sampling data

#gam_fus_mod<-mclapply(1:length(dates), runGAMFusion,mc.preschedule=FALSE,mc.cores = 8) #This is the end bracket from mclapply(...) statement

#gam_fus_mod_s<-mclapply(1:1, runGAMFusion,mc.preschedule=FALSE,mc.cores = 1) #This is the end bracket from mclapply(...) statement

gam_fus_mod_s<-mclapply(1:length(ghcn.subsets), runGAMFusion,mc.preschedule=FALSE,mc.cores = 9) #This is the end bracket from mclapply(...) statement

#gam_fus_mod2<-mclapply(4:4, runGAMFusion,mc.preschedule=FALSE,mc.cores = 1) #This is the end bracket from mclapply(...) statement

save(gam_fus_mod_s,file= paste(path,"/","results2_fusion_Assessment_measure_all",out_prefix,".RData",sep=""))

## Plotting and saving diagnostic measures

tb<-gam_fus_mod_s[[1]][[3]][0,]  #empty data frame with metric table structure that can be used in rbinding...

tb_tmp<-gam_fus_mod_s #copy

for (i in 1:length(tb_tmp)){

  tmp<-tb_tmp[[i]][[3]]

  tb<-rbind(tb,tmp)

}

rm(tb_tmp)

for(i in 4:ncol(tb)){            # start of the for loop #1

  tb[,i]<-as.numeric(as.character(tb[,i]))  

}

metrics<-as.character(unique(tb$metric))            #Name of accuracy metrics (RMSE,MAE etc.)

tb_metric_list<-vector("list",length(metrics))

for(i in 1:length(metrics)){            # Reorganizing information in terms of metrics 

  metric_name<-paste("tb_",metrics[i],sep="")

  tb_metric<-subset(tb, metric==metrics[i])

  tb_metric<-cbind(tb_metric,sampling_dat[,2:3])

  assign(metric_name,tb_metric)

  tb_metric_list[[i]]<-tb_metric

}

tb_diagnostic<-do.call(rbind,tb_metric_list)

tb_diagnostic[["prop"]]<-as.factor(tb_diagnostic[["prop"]])

mod_pat<-glob2rx("mod*")   

mod_var<-grep(mod_pat,names(tb_diagnostic),value=TRUE) # using grep with "value" extracts the matching names         

t<-melt(tb_diagnostic,

        measure=mod_var, 

        id=c("dates","metric","prop"),

        na.rm=F)

avg_tb<-cast(t,metric+prop~variable,mean)

median_tb<-cast(t,metric+prop~variable,median)

avg_tb[["prop"]]<-as.numeric(as.character(avg_tb[["prop"]]))

avg_RMSE<-subset(avg_tb,metric=="RMSE")

sampling_obj<-list(sampling_dat=sampling_dat,training=sampling, training_id=sampling_station_id, tb=tb_diagnostic)

write.table(avg_tb, file= paste(path,"/","results2_fusion_Assessment_measure_avg_",out_prefix,".txt",sep=""), sep=",")

write.table(median_tb, file= paste(path,"/","results2_fusion_Assessment_measure_median_",out_prefix,".txt",sep=""), sep=",")

write.table(tb_diagnostic, file= paste(path,"/","results2_fusion_Assessment_measure",out_prefix,".txt",sep=""), sep=",")

write.table(tb, file= paste(path,"/","results2_fusion_Assessment_measure_all",out_prefix,".txt",sep=""), sep=",")

save(sampling_obj, file= paste(path,"/","results2_fusion_sampling_obj",out_prefix,".RData",sep=""))

#save(gam_fus_mod_s,file= paste(path,"/","results2_fusion_Assessment_measure_all",out_prefix,".RData",sep=""))

gam_fus_mod_obj<-list(gam_fus_mod=gam_fus_mod_s,sampling_obj=sampling_obj)

save(gam_fus_mod_obj,file= paste(path,"/","results_mod_obj_",out_prefix,".RData",sep=""))

#### END OF SCRIPT