#### Assessing relevance of MODIS cloud data for SDM

library(dismo)

##  Species

sp=c("Spondias radlkoferi","Tapirira mexicana","Lozanella enantiophylla")

spl=strsplit(sp," ")

tsp=spl[[3]]

td=gbif(genus=tsp[1],species=tsp[2],geo=TRUE, sp=FALSE, removeZeros=FALSE, download=TRUE, getAlt=TRUE, start=1, end=NULL, feedback=3)

td

library(texreg)

#cld=read.csv("data/NDP026D/cld.csv")

#cldy=read.csv("data/NDP026D/cldy.csv")

#clda=read.csv("data/NDP026D/clda.csv")

#cld$month2=factor(cld$month,labels=month.name)

#cldys=cldy

#coordinates(cldys)=c("lon","lat")

#projection(cldys)=CRS("+proj=longlat +ellps=WGS84 +datum=WGS84 +no_defs")

mod09s=brick("data/cloud_yseasmean.nc",varname="CF");names(mod09s)=c("DJF","MAM","JJA","SON")

mod09a=brick("data/cloud_mean.nc",varname="CF");names(mod09a)="Mean Annual Cloud Frequency"

mod09min=brick("data/cloud_min.nc",varname="CFmin")

mod09max=brick("data/cloud_max.nc",varname="CFmax")

mod09sd=brick("data/cloud_std.nc",varname="CFsd")

cldm$difm=cldm$mod09-cldm$cld_all

cldm$difs=cldm$mod09sd+cldm$cldsd_all

#clda$dif=clda$mod09-clda$cld

library(rgdal)

#coast=getRgshhsMap("/mnt/data/jetzlab/Data/environ/global/gshhg/gshhs_h.b", xlim = NULL, ylim = NULL, level = 4) 

land=readShapePoly("/mnt/data/jetzlab/Data/environ/global/gshhg/GSHHS_shp/c/GSHHS_c_L1.shp",force_ring=TRUE)

projection(land)="+proj=longlat +ellps=WGS84 +datum=WGS84 +no_defs"

CP <- as(extent(-180, 180, -60, 84), "SpatialPolygons")

proj4string(CP) <- CRS(proj4string(land))

coast=as(land[land$area>50,],"SpatialLines")

## Clip the map

land <- gIntersection(land, CP, byid=F)

coast <- gIntersection(coast, CP, byid=F)

#### get biome data

##make spatial object

cldms=cldm

coordinates(cldms)=c("lon","lat")

projection(cldms)=CRS("+proj=longlat +ellps=WGS84 +datum=WGS84 +no_defs")

if(!file.exists("data/teow/biomes.shp")){

    teow=readOGR("/mnt/data/jetzlab/Data/environ/global/teow/official/","wwf_terr_ecos")

    teow=teow[teow$BIOME<90,]

    biome=unionSpatialPolygons(teow,teow$BIOME, threshold=5)

    biomeid=read.csv("/mnt/data/jetzlab/Data/environ/global/teow/official/biome.csv",stringsAsFactors=F)

    biome=SpatialPolygonsDataFrame(biome,data=biomeid)

    writeOGR(biome,"data/teow","biomes",driver="ESRI Shapefile")

}

biome=readOGR("data/teow/","biomes")

projection(biome)=projection(st)

st$biome=over(st,biome,returnList=F)$BiomeID

cldm$biome=st$biome[match(cldm$StaID,st$id)]

## get stratified sample of points from biomes for illustration

if(!file.exists("output/biomesamplepoints.csv")){

    n_biomesamples=1000

    library(multicore)

    biomesample=do.call(rbind.data.frame,mclapply(1:length(biome),function(i)

        data.frame(biome=biome$BiomeID[i],coordinates(spsample(biome[i,],n=n_biomesamples,type="stratified",nsig=2)))))

    write.csv(biomesample,"output/biomesamplepoints.csv",row.names=F)

}

biomesample=read.csv("output/biomesamplepoints.csv")

coordinates(biomesample)=c("x1","x2")

#biomesample=extract(biomesample,mod09c)

#pdf("output/Figures.pdf",width=11,height=8.5)

png("output/Figures_%03d.png",width=5000,height=4000,res=600,pointsize=36,bg="transparent")

res=1e5

greg=list(ylim=c(-60,84),xlim=c(-180,180))

  margin=F,maxpixels=res,ylab="",xlab="",useRaster=T,ylim=greg$ylim)+

## Mean annual with validation stations

levelplot(mod09a,col.regions=colr(n),cuts=100,at=seq(0,100,len=100),colorkey=list(space="bottom",adj=1),

  margin=F,maxpixels=res,ylab="",xlab="",useRaster=T,ylim=greg$ylim)+

## Seasonal Means

levelplot(mod09s,col.regions=colr(n),cuts=100,at=seq(0,100,len=100),colorkey=list(space="bottom",adj=2),

  margin=F,maxpixels=res,ylab="",xlab="",useRaster=T,ylim=greg$ylim)+

  layer(sp.lines(coast,col="black"),under=F)

## Monthly Means

levelplot(mod09c,col.regions=colr(n),cuts=100,at=seq(0,100,len=100),colorkey=list(space="bottom",adj=1),

  margin=F,maxpixels=res,ylab="Latitude",xlab="Longitude",useRaster=T,ylim=greg$ylim)+

  layer(sp.lines(coast,col="black"),under=F)

#- Monthly minimum

#- Monthly maximum

#- STDEV or Min-Max

#p_mac=levelplot(mod09a,col.regions=colr(n),cuts=99,at=seq(0,100,length=100),margin=F,maxpixels=1e5,colorkey=list(space="bottom",height=.75),xlab="",ylab="",useRaster=T)+

#      layer(sp.lines(coast,col="black"),under=F)

#p_min=levelplot(mod09min,col.regions=colr(n),cuts=99,margin=F,maxpixels=1e5,colorkey=list(space="bottom",height=.75),useRaster=T)

#p_max=levelplot(mod09max,col.regions=colr(n),cuts=99,margin=F,maxpixels=1e5,colorkey=list(space="bottom",height=.75),useRaster=T)

#p_sd=levelplot(mod09sd,col.regions=colr(n),cuts=99,at=seq(0,100,length=100),margin=F,maxpixels=1e5,colorkey=list(space="bottom",height=.75),useRaster=T)

#p3=c("Mean Cloud Frequency (%)"=p_mac,"Max Cloud Frequency (%)"=p_max,"Min Cloud Frequency (%)"=p_min,"Cloud Frequency Variability (SD)"=p_sd,x.same=T,y.same=T,merge.legends=T,layout=c(2,2))

#print(p3)

bgr=function(x,n=100,br=0,c1=c("darkblue","blue","grey"),c2=c("grey","red","purple")){

    at=unique(c(seq(min(x,na.rm=T),max(x,na.rm=T),len=n)))

    bg=colorRampPalette(c1)

    gr=colorRampPalette(c2)

    return(list(at=at,col=c(bg(sum(at<br)),gr(sum(at>=br)))))

}

colat=bgr(cldm$difm)

phist=histogram(cldm$difm,breaks=colat$at,border=NA,col=colat$col,xlim=c(-50,40),type="count",xlab="Difference (MOD09-NDP026D)")#,seq(0,1,len=6),na.rm=T)

pmap=xyplot(lat~lon|month2,data=cldm,groups=cut(cldm$difm,rev(colat$at)),

       par.settings=list(superpose.symbol=list(col=colat$col)),pch=16,cex=.25,

       auto.key=F,#list(space="right",title="Difference\n(MOD09-NDP026D)",cex.title=1),asp=1,

       ylab="Latitude",xlab="Longitude")+

  layer(sp.lines(coast,col="black",lwd=.1),under=F)

print(phist,position=c(0,.75,1,1),more=T)

print(pmap,position=c(0,0,1,.78))

hmcols=colorRampPalette(c("grey","blue","red","purple"))

tr=c(0,80)

xyplot(cld_all~mod09|month2,data=cldm,panel=function(x,y,subscripts){

  print(max(qat))

#  panel.abline(0,1,col="black",lwd=2)

  panel.abline(lm(y ~ x),col="black",lwd=2)

              legend= list(right = list(fun = colkey)))#+ layer(panel.abline(0,1,col="black",lwd=2))

## for(i in 1:2){

##  p1=xyplot(cld~mod09|month2,data=cldm,cex=.2,pch=16,subscripts=T,ylab="NDP Mean Cloud Amount",xlab="MOD09 Cloud Frequency (%)")+

##   layer(panel.lines(1:100,predict(lm(y~x),newdata=data.frame(x=1:100)),col="green"))+

##   layer(panel.lines(1:100,predict(lm(y~x+I(x^2)),newdata=data.frame(x=1:100)),col="blue"))+

##   layer(panel.abline(0,1,col="red"))

##     if(i==2){

##      p1=p1+layer(panel.segments(mod09[subscripts],cld[subscripts]-cldsd[subscripts],mod09[subscripts],cld[subscripts]+cldsd[subscripts],subscripts=subscripts,col="grey"),data=cldm,under=T,magicdots=T)

##      p1=p1+layer(panel.segments(mod09[subscripts]-mod09sd[subscripts],cld[subscripts],mod09[subscripts]+mod09sd[subscripts],cld[subscripts],subscripts=subscripts,col="grey"),data=cldm,under=T,magicdots=T) 

##        }

## print(p1)

## }

bwplot(lulcc~difm,data=cldm,horiz=T,xlab="Difference (MOD09-Observed)",varwidth=T,notch=T)+layer(panel.abline(v=0))

#library(BayesFactor)

#coplot(difm~month2|lulcc,data=cldm[!is.na(cldm$dif)&!is.na(cldm$lulcc),],panel = panel.smooth)

## Validation table construction

summary(lm(cld_all~mod09+lat,data=cldm))

## assess latitude bias

cldm$abslat=abs(cldm$lat)

cldm$absdif=abs(cldm$difm)

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

### validation by biome

bs=biome$BiomeID

mod_bs=lapply(bs,function(bs) lm(cld_all~mod09,data=cldm[cldm$biome==bs,]))

names(mod_bs)=biome$Biome

screenreg(mod_bs,digits=2,single.row=F,custom.model.names=names(mod_bs))

#lm_all=lm(cld_all~mod09,data=cldm[!is.na(cldm$cld),])

lm_mod=lm(cld~mod09+biome,data=cldm)

screenreg(lm_mod,digits=2,single.row=F)

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

## assess temporal stability

## spatialy subset data to stations at least 10km apart

st2=remove.duplicates(st,zero=10)

## Subset data

## drop missing observations

cldm.t=cldm[!is.na(cldm$cld_all)&!is.na(cldm$mod09)&!is.na(cldm$biome),]

cldm.t=cldm.t[cldm.t$lat>=-60,]

#  make sure all stations have all mod09 data

stdrop=names(which(tapply(cldm.t$month,cldm.t$StaID,length)!=12))

cldm.t=cldm.t[!cldm.t$StaID%in%stdrop,]

# Keep only stations at least 10km apart 

cldm.t=cldm.t[cldm.t$StaID%in%st2$id,]

## Subset to only some months, if desired

#cldm.t=cldm.t[cldm.t$month%in%1:3,]

## Select Knots

knots=spsample(land,500,type="regular")

                                        #  reshape data

m.cld=cast(cldm.t,StaID+lat+lon+biome~month,value="cld_all");colnames(m.cld)[-(1:4)]=paste("cld.",colnames(m.cld)[-(1:4)],sep="")

m.mod09=cast(cldm.t,StaID~month,value="mod09");colnames(m.mod09)[-1]=paste("mod09.",colnames(m.mod09)[-1],sep="")

mdata=cbind(m.cld,m.mod09)

## cast to 

coords <- as.matrix(m.cld[,c("lon","lat")])#as.matrix(ne.temp[,c("UTMX", "UTMY")]/1000)

max.d <- max(iDist(coords))

##make symbolic model formula statement for each month

mods <- lapply(paste(paste(paste("cld.",1:N.t,sep=''),paste("mod09.",1:N.t,sep=''),sep='~'),"",sep=""), as.formula)

tlm=model.matrix(lm(mods[[1]],data=mdata))

N.t <- ncol(m.mod09)-1 ##number of months

n <- nrow(m.cld) ##number of observation per months

p <- ncol(tlm) #number of regression parameters in each month

starting <- list("beta"=rep(0,N.t*p), "phi"=rep(3/(0.5*max.d), N.t),

                 "sigma.sq"=rep(2,N.t), "tau.sq"=rep(1, N.t),

                 "sigma.eta"=diag(rep(0.01, p)))

tuning <- list("phi"=rep(5, N.t))

priors <- list("beta.0.Norm"=list(rep(0,p), diag(1000,p)),

               "phi.Unif"=list(rep(3/(0.9*max.d), N.t), rep(3/(0.05*max.d), N.t)),

               "sigma.sq.IG"=list(rep(2,N.t), rep(10,N.t)),

               "tau.sq.IG"=list(rep(2,N.t), rep(5,N.t)),

               "sigma.eta.IW"=list(2, diag(0.001,p)))

cov.model <- "exponential"

## Run the model

n.samples <- 100

m.1=spDynLM(mods,data=mdata,coords=coords,knots=coordinates(knots),n.samples=n.samples,starting=starting,tuning=tuning,priors=priors,cov.model=cov.model,get.fitted=T,n.report=25)

## summarize

burn.in <- floor(0.75*n.samples)

quant <- function(x){quantile(x, prob=c(0.5, 0.025, 0.975))}

beta <- apply(m.1$p.beta.samples[burn.in:n.samples,], 2, quant)

beta.0 <- beta[,grep("Intercept", colnames(beta))]

beta.1 <- beta[,grep("mod09", colnames(beta))]

library(texreg)

 extract.lm <- function(model) {

     s <- summary(model)

     names <- rownames(s$coef)

     co <- s$coef[, 1]

     se <- s$coef[, 2]

     pval <- s$coef[, 4]

     rs <- s$r.squared

     n <- nobs(model)

     rmse=sqrt(mean((residuals(s)^2)))

     gof <- c(rs, rmse, n)

     gof.names <- c("R-Squared","RMSE","n")

     tr <- createTexreg(coef.names = names, coef = co, se = se, 

                        pvalues = pval, gof.names = gof.names, gof = gof)

     return(tr)

 }

setMethod("extract", signature = className("lm", "stats"),definition = extract.lm)

forms=c("cld~mod09+month2+lat")

lm_all=lm(cld_all~mod09+lat,data=cldm[!is.na(cldm$cld),])

### Compare two time periods

lm_all1=lm(cld_all~mod09,data=cldm[!is.na(cldm$cld),])

lm_mod=lm(cld~mod09,data=cldm)

mods=list("1970-2000 complete"=lm_all1,"2000-2009"=lm_mod)

screenreg(mods,digits=2,single.row=T,custom.model.names=names(mods))

htmlreg(mods,file = "output/tempstab.doc",

        custom.model.names = names(mods),

        single.row = T, inline.css = FALSE,doctype = TRUE, html.tag = TRUE, head.tag = TRUE, body.tag = TRUE)

abslm=lm(absdif~abslat:I(abslat^2),data=cldm)

plot(absdif~abslat,data=cldm,cex=.25,pch=16)

lines(0:90,predict(abslm,newdata=data.frame(abslat=0:90),type="response"),col="red")

bf=anovaBF(dif~lulcc+month2,data=cldm[!is.na(cldm$dif)&!is.na(cldm$lulcc),])

ch=posterior(bf, iterations = 1000)

summary(bf)

plot(bf)

lulctl=ddply(cldm,c("month","lulc"),function(x) c(count=nrow(x),rmse=sqrt(mean((x$mod09-x$cld)^2,na.rm=T))))

lulctl=lulctl[!is.na(lulctl$lulc),]

lulctl$lulcc=as.factor(IGBP$class[match(lulctl$lulc,IGBP$ID)])

lulctl= ddply(cldm,c("lulc"),function(x) c(count=nrow(x),mean=paste(mean(x$difm,na.rm=T)," (",sd(x$difm,na.rm=T),")",sep=""),rmse=sqrt(mean((x$difm)^2,na.rm=T))))

lulctl$lulcc=as.factor(IGBP$class[match(lulctl$lulc,IGBP$ID)]

print(xtable(lulctl[,c("lulcc","count","mean","rmse")],digits=1),"html")

lulcrmse.q=round(do.call(rbind,apply(lulcrmse,1,function(x) data.frame(Min=min(x,na.rm=T),Mean=mean(x,na.rm=T),Max=max(x,na.rm=T),SD=sd(x,na.rm=T)))),1)#quantile,c(0.025,0.5,.975),na.rm=T)),1)

lulcrmse.q=lulcrmse.q[order(lulcrmse.q$Mean,decreasing=T),]

lulcrmse.q

bgyr=colorRampPalette(c("blue","green","yellow","red"))

levelplot(rmse~month*lulcc,data=lulcrmsel,col.regions=bgyr(1000),at=quantile(lulcrmsel$rmse,seq(0,1,len=100),na.rm=T))

cld$Amt=cld$Amt/100

## calculate means and sds for full record (1970-2009)

Nobsthresh=20 #minimum number of observations to include 

      month=x$month[1],

      StaID=x$StaID[1],

      cld_all=mean(x$Amt[x$Nobs>=Nobsthresh],na.rm=T),  # full record

      cldsd_all=sd(x$Amt[x$Nobs>=Nobsthresh],na.rm=T),

      cld=mean(x$Amt[x$YR>=2000&x$Nobs>=Nobsthresh],na.rm=T), #only MODIS epoch

      cldsd=sd(x$Amt[x$YR>=2000&x$Nobs>=Nobsthresh],na.rm=T))}))

mod09st=read.csv("valid.csv",header=F)[,-c(1)]

colnames(mod09st)=c(names(mod09),"id","type")

mod09stl=melt(mod09st,id.vars=c("id","type"))

mod09stl=cast(mod09stl,id+year+month~type,value="value")

cldm$mod09=mod09stl$mean[match(paste(cldm$StaID,cldm$month),paste(mod09stl$id,as.numeric(mod09stl$month)))]

cldm$mod09sd=mod09stl$sd[match(paste(cldm$StaID,cldm$month),paste(mod09stl$id,as.numeric(mod09stl$month)))]

file.remove("tmp/mod09_2000-01-15.nc")

system(paste("cdo -O  mergetime -setrtomiss,-32768,-1 ",paste(list.files(tempdir,pattern="mod09.*.nc$",full=T),collapse=" ")," data/cloud_monthly.nc"))

system(paste("cdo -O  timmean data/cloud_monthly.nc  data/cloud_mean.nc"))

## Seasonal Means

system(paste("cdo  -f nc4c -O -yseasmean data/cloud_monthly.nc data/cloud_yseasmean.nc"))

system(paste("cdo  -f nc4c -O -yseasstd data/cloud_monthly.nc data/cloud_yseasstd.nc"))

    system("cdo -f nc4c  timmin data/cloud_ymonmean.nc data/cloud_min.nc")

    system("cdo -f nc4c  timmax data/cloud_ymonmean.nc data/cloud_max.nc")

# Regressions through time by season

s=c("DJF","MAM","JJA","SON")

system(paste("cdo  -f nc4c -O regres -selseas,",s[1]," data/cloud_monthly.nc data/slope_",s[1],".nc &",sep=""))

system(paste("cdo  -f nc4c -O regres -selseas,",s[2]," data/cloud_monthly.nc data/slope_",s[2],".nc &",sep=""))

system(paste("cdo  -f nc4c -O regres -selseas,",s[3]," data/cloud_monthly.nc data/slope_",s[3],".nc &",sep=""))

system(paste("cdo  -f nc4c -O regres -selseas,",s[4]," data/cloud_monthly.nc data/slope_",s[4],".nc &",sep=""))