Revision 29eea01f
Added by Benoit Parmentier over 9 years ago
| climate/research/oregon/interpolation/global_run_scalingup_mosaicing.R | ||
|---|---|---|
| 5 | 5 |
#Analyses, figures, tables and data are also produced in the script. |
| 6 | 6 |
#AUTHOR: Benoit Parmentier |
| 7 | 7 |
#CREATED ON: 04/14/2015 |
| 8 |
#MODIFIED ON: 05/25/2015
|
|
| 8 |
#MODIFIED ON: 05/27/2015
|
|
| 9 | 9 |
#Version: 4 |
| 10 | 10 |
#PROJECT: Environmental Layers project |
| 11 | 11 |
#COMMENTS: analyses for run 10 global analyses,all regions 1500x4500km and other tiles |
| ... | ... | |
| 69 | 69 |
return(out_dir) |
| 70 | 70 |
} |
| 71 | 71 |
|
| 72 |
sine_structure_fun <-function(x,T,phase,a,b,use_cos=FALSE){
|
|
| 73 |
|
|
| 74 |
#Create sine for a one dimensional series |
|
| 75 |
#Note that sine function uses radian unit. |
|
| 76 |
#a=amplitude |
|
| 77 |
#b=mean or amplitude 0 of the series |
|
| 78 |
#T= stands for period definition |
|
| 79 |
#phase=phase angle (in radian!!) |
|
| 80 |
#cos: use cosine instead of sine if TRUE |
|
| 81 |
|
|
| 82 |
if(use_cos==FALSE){
|
|
| 83 |
y <- a*sin((x*pi/T)+ phase) + b |
|
| 84 |
}else{
|
|
| 85 |
y <- a*cos((x*pi/T)+ phase) + b |
|
| 86 |
} |
|
| 87 |
return(y) |
|
| 88 |
} |
|
| 89 |
|
|
| 72 | 90 |
create_weights_fun <- function(i, list_param){
|
| 73 | 91 |
#This function generates weights from a point location on a raster layer. |
| 74 | 92 |
#Note that the weights are normatlized on 0-1 scale using max and min values. |
| 75 | 93 |
#Inputs: |
| 76 | 94 |
#lf: list of raster files |
| 77 |
#df_points: |
|
| 95 |
#df_points: reference points from which to compute distance |
|
| 96 |
#r_feature: reference features as raster image from which to compute distance from |
|
| 97 |
#methods: options available: use_sine_weights,use_edge,use_linear_weights |
|
| 78 | 98 |
#Outputs: |
| 79 |
#TODO: add options to generate weights from edges/reference image rather than reference centroids points... |
|
| 99 |
#raster list of weights and product of wegihts and inuts |
|
| 100 |
#TODO: |
|
| 101 |
# -use gdal proximity for large files and use_edge option |
|
| 102 |
# - add raster options |
|
| 103 |
# - improve efficiency |
|
| 104 |
# - change name options |
|
| 105 |
# |
|
| 80 | 106 |
############ |
| 81 | 107 |
|
| 82 | 108 |
lf <- list_param$lf |
| 83 |
df_centroids <- list_param$df_points |
|
| 84 |
feature_raster <- list_param$feature_raster |
|
| 85 |
use_edge <- list_param$use_edge |
|
| 109 |
df_points <- list_param$df_points |
|
| 110 |
r_feature <- list_param$r_feature #this should be change to a list |
|
| 111 |
padding <- TRUE #if padding true then make buffer around edges?? |
|
| 112 |
method <- list_param$method #differnt methods available to create weights |
|
| 113 |
#NAflag,file_format,out_suffix etc... |
|
| 86 | 114 |
out_dir_str <- list_param$out_dir_str |
| 87 | 115 |
|
| 88 | 116 |
####### START SCRIPT ##### |
| 89 | 117 |
|
| 90 |
r1 <- raster(lf[i]) #input image
|
|
| 118 |
r_in <- raster(lf[i]) #input image
|
|
| 91 | 119 |
|
| 92 | 120 |
set1f <- function(x){rep(NA, x)}
|
| 93 |
r_init <- init(r1, fun=set1f)
|
|
| 121 |
r_init <- init(r_in, fun=set1f)
|
|
| 94 | 122 |
|
| 95 |
if(!is.null(df_centroids)){
|
|
| 96 |
cell_ID <- cellFromXY(r_init,xy=df_centroids[i,]) |
|
| 97 |
r_init[cell_ID] <- df_centroids$ID[i] |
|
| 123 |
if(!is.null(r_feature)){
|
|
| 124 |
r_init <- r_feature |
|
| 98 | 125 |
} |
| 99 | 126 |
|
| 100 |
if(!is.null(feature_raster)){
|
|
| 101 |
cell_ID <- cellFromXY(r_init,xy=df_centroids[i,])
|
|
| 102 |
r_init[cell_ID] <- df_centroids$ID[i]
|
|
| 127 |
if(!is.null(df_points)){ #reference points as SPDF object
|
|
| 128 |
cell_ID <- cellFromXY(r_init,xy=df_points[i,])
|
|
| 129 |
r_init[cell_ID] <- df_points$ID[i]
|
|
| 103 | 130 |
} |
| 104 | 131 |
|
| 105 |
#change here to distance from edges... |
|
| 106 |
|
|
| 107 |
#if(use_edge==T){
|
|
| 108 |
# n_col <- ncol(r_init) |
|
| 109 |
# n_row <- nrow(r_init) |
|
| 110 |
# |
|
| 111 |
# #xfrom |
|
| 112 |
# r_init[1,1:n_col] <- 1 |
|
| 113 |
# r_init[n_row,1:n_col] <- 1 |
|
| 114 |
# r_init[1:n_row,1] <- 1 |
|
| 115 |
# r_init[1:n_row,n_col] <- 1 |
|
| 132 |
if(method=="use_sine_weights"){
|
|
| 133 |
#Generate spatial pattern 5: |
|
| 134 |
n_col <- ncol(r_init) |
|
| 135 |
n_row <- nrow(r_init) |
|
| 136 |
|
|
| 137 |
#u <- xFromCol(r_init,col=1:n_col) |
|
| 138 |
#add padding option later...buffer from a specific distance and tailling of at 0.1 |
|
| 139 |
u <- 1:n_col |
|
| 140 |
a<- 1 #amplitude in this case |
|
| 141 |
b<- 0 |
|
| 142 |
T<- n_col |
|
| 143 |
phase <- 0 |
|
| 144 |
use_cos <- FALSE |
|
| 145 |
ux <- sine_structure_fun(u,T,phase,a,b,use_cos) |
|
| 146 |
ux_rep <-rep(ux,time=n_row) |
|
| 147 |
r1 <-setValues(r_init,ux_rep) #note efficient in memory might need to revise this |
|
| 148 |
#plot(r) |
|
| 149 |
|
|
| 150 |
v <- 1:n_row |
|
| 151 |
a<- 1 #amplitude in this case |
|
| 152 |
b<- 0 |
|
| 153 |
T<- n_row |
|
| 154 |
phase <- 0 |
|
| 155 |
use_cos <- FALSE |
|
| 156 |
vx <- sine_structure_fun(v,T,phase,a,b,use_cos) |
|
| 157 |
vx_rep <- unlist((lapply(1:n_row,FUN=function(i){rep(vx[i],time=n_col)})))
|
|
| 158 |
|
|
| 159 |
r2 <-setValues(r_init,vx_rep) |
|
| 160 |
#plot(r2) |
|
| 161 |
|
|
| 162 |
r <- (r1+r2)*0.5 #combine patterns to make a elliptic surface, -0.5 could be modified |
|
| 163 |
#plot(r) |
|
| 164 |
} |
|
| 116 | 165 |
|
| 117 |
#} too slow |
|
| 166 |
#change here to distance from edges.. |
|
| 167 |
if(method=="use_edge"){ #does not work with large images
|
|
| 168 |
#change...use gdal |
|
| 169 |
n_col <- ncol(r_init) |
|
| 170 |
n_row <- nrow(r_init) |
|
| 171 |
|
|
| 172 |
#xfrom |
|
| 173 |
r_init[1,1:n_col] <- 1 |
|
| 174 |
r_init[n_row,1:n_col] <- 1 |
|
| 175 |
r_init[1:n_row,1] <- 1 |
|
| 176 |
r_init[1:n_row,n_col] <- 1 |
|
| 177 |
r_dist <- distance(r_init) |
|
| 178 |
min_val <- cellStats(r_dist,min) |
|
| 179 |
max_val <- cellStats(r_dist,max) |
|
| 180 |
r <- abs(r_dist - max_val)/ (max_val - min_val) |
|
| 181 |
} #too slow with R use http://www.gdal.org/gdal_proximity.html |
|
| 118 | 182 |
|
| 119 | 183 |
#plot(r_init) |
| 120 |
r_dist <- distance(r_init) |
|
| 121 |
min_val <- cellStats(r_dist,min) |
|
| 122 |
max_val <- cellStats(r_dist,max) |
|
| 123 |
r_dist_normalized <- abs(r_dist - max_val)/ (max_val - min_val) |
|
| 184 |
|
|
| 185 |
if(method=="use_linear_weights"){
|
|
| 186 |
# |
|
| 187 |
r_dist <- distance(r_init) |
|
| 188 |
min_val <- cellStats(r_dist,min) |
|
| 189 |
max_val <- cellStats(r_dist,max) |
|
| 190 |
r <- abs(r_dist - max_val)/ (max_val - min_val) |
|
| 191 |
} |
|
| 124 | 192 |
|
| 125 | 193 |
extension_str <- extension(lf[i]) |
| 126 | 194 |
raster_name_tmp <- gsub(extension_str,"",basename(lf[i])) |
| 127 |
raster_name <- file.path(out_dir_str,paste(raster_name_tmp,"_weights.tif",sep="")) |
|
| 128 |
writeRaster(r_dist_normalized, NAflag=NA_flag_val,filename=raster_name,overwrite=TRUE)
|
|
| 195 |
raster_name <- file.path(out_dir_str,paste(raster_name_tmp,"_",method,"_weights.tif",sep=""))
|
|
| 196 |
writeRaster(r, NAflag=NA_flag_val,filename=raster_name,overwrite=TRUE) |
|
| 129 | 197 |
|
| 130 |
r_var_prod <- r1*r_dist_normalized
|
|
| 131 |
raster_name_prod <- file.path(out_dir_str, paste(raster_name_tmp,"_prod_weights.tif",sep="")) |
|
| 198 |
r_var_prod <- r_in*r
|
|
| 199 |
raster_name_prod <- file.path(out_dir_str, paste(raster_name_tmp,"_",method,"_prod_weights.tif",sep=""))
|
|
| 132 | 200 |
writeRaster(r_var_prod, NAflag=NA_flag_val,filename=raster_name_prod,overwrite=TRUE) |
| 133 | 201 |
|
| 134 | 202 |
weights_obj <- list(raster_name,raster_name_prod) |
| ... | ... | |
| 244 | 312 |
|
| 245 | 313 |
y_var_name <- "dailyTmax" #PARAM1 |
| 246 | 314 |
interpolation_method <- c("gam_CAI") #PARAM2
|
| 247 |
out_suffix <- "mosaic_run10_1500x4500_global_analyses_05252015" #PARAM3
|
|
| 315 |
out_suffix <- "mosaic_run10_1500x4500_global_analyses_05272015" #PARAM3
|
|
| 248 | 316 |
out_dir <- in_dir #PARAM4 |
| 249 | 317 |
create_out_dir_param <- TRUE #PARAM 5 |
| 250 | 318 |
|
| ... | ... | |
| 315 | 383 |
out_dir_str <- out_dir |
| 316 | 384 |
lf_r_weights <- vector("list",length=length(lf_mosaic))
|
| 317 | 385 |
|
| 318 |
list_param_create_weights <- list(lf_mosaic,df_centroids,out_dir_str) |
|
| 319 |
names(list_param_create_weights) <- c("lf","df_points","out_dir_str")
|
|
| 386 |
#methods availbable:use_sine_weights,use_edge,use_linear_weights |
|
| 387 |
|
|
| 388 |
### First use linear weights |
|
| 389 |
method <- "use_linear_weights" |
|
| 390 |
df_points <- df_centroids |
|
| 391 |
#df_points <- NULL |
|
| 392 |
r_feature <- NULL |
|
| 393 |
|
|
| 394 |
list_param_create_weights <- list(lf_mosaic,df_points,r_feature,method,out_dir_str) |
|
| 395 |
names(list_param_create_weights) <- c("lf","df_points","r_feature","method","out_dir_str")
|
|
| 320 | 396 |
num_cores <- 11 |
| 321 | 397 |
|
| 322 | 398 |
#debug(create_weights_fun) |
| ... | ... | |
| 324 | 400 |
|
| 325 | 401 |
#This is the function creating the weights by tile. Distance from the centroids needs to be change from distance to |
| 326 | 402 |
#the edges...can use rows and columsn to set edges to 1 and 0 for the others. |
| 327 |
weights_obj_list <- mclapply(1:length(lf_mosaic),FUN=create_weights_fun,list_param=list_param_create_weights,mc.preschedule=FALSE,mc.cores = num_cores) |
|
| 403 |
linear_weights_obj_list <- mclapply(1:length(lf_mosaic),FUN=create_weights_fun,list_param=list_param_create_weights,mc.preschedule=FALSE,mc.cores = num_cores) |
|
| 404 |
|
|
| 405 |
list_linear_r_weights <- lapply(1:length(linear_weights_obj_list), FUN=function(i,x){x[[i]]$r_weights},x=linear_weights_obj_list)
|
|
| 406 |
list_linear_r_weights_prod <- lapply(1:length(linear_weights_obj_list), FUN=function(i,x){x[[i]]$r_weights_prod},x=linear_weights_obj_list)
|
|
| 407 |
|
|
| 408 |
### Second use sine weights |
|
| 409 |
method <- "use_sine_weights" |
|
| 410 |
#df_points <- df_centroids |
|
| 411 |
df_points <- NULL |
|
| 412 |
r_feature <- NULL |
|
| 413 |
|
|
| 414 |
list_param_create_weights <- list(lf_mosaic,df_points,r_feature,method,out_dir_str) |
|
| 415 |
names(list_param_create_weights) <- c("lf","df_points","r_feature","method","out_dir_str")
|
|
| 416 |
num_cores <- 11 |
|
| 417 |
|
|
| 418 |
#debug(create_weights_fun) |
|
| 419 |
weights_obj <- create_weights_fun(1,list_param=list_param_create_weights) |
|
| 328 | 420 |
|
| 329 |
list_r_weights <- lapply(1:length(weights_obj_list), FUN=function(i,x){x[[i]]$r_weights},x=weights_obj_list)
|
|
| 330 |
list_r_weights_prod <- lapply(1:length(weights_obj_list), FUN=function(i,x){x[[i]]$r_weights_prod},x=weights_obj_list)
|
|
| 421 |
#This is the function creating the weights by tile. Distance from the centroids needs to be change from distance to |
|
| 422 |
#the edges...can use rows and columsn to set edges to 1 and 0 for the others. |
|
| 423 |
sine_weights_obj_list <- mclapply(1:length(lf_mosaic),FUN=create_weights_fun,list_param=list_param_create_weights,mc.preschedule=FALSE,mc.cores = num_cores) |
|
| 424 |
|
|
| 425 |
list_sine_r_weights <- lapply(1:length(sine_weights_obj_list), FUN=function(i,x){x[[i]]$r_weights},x=sine_weights_obj_list)
|
|
| 426 |
list_sine_r_weights_prod <- lapply(1:length(sine_weights_obj_list), FUN=function(i,x){x[[i]]$r_weights_prod},x=sine_weights_obj_list)
|
|
| 427 |
|
|
| 428 |
## |
|
| 331 | 429 |
|
| 332 | 430 |
#Then scale on 1 to zero? or 0 to 1000 |
| 333 | 431 |
#e.g. for a specific pixel |
| ... | ... | |
| 337 | 435 |
#m_val= sum_val/weight_sum #mean value |
| 338 | 436 |
# |
| 339 | 437 |
|
| 438 |
############### |
|
| 439 |
### PART 3: prepare weightsfor mosaicing by matching extent ############ |
|
| 440 |
|
|
| 340 | 441 |
## Rasters tiles vary slightly in resolution, they need to be matched for the mosaic. Resolve issue in the |
| 341 | 442 |
#mosaic funciton using gdal_merge to compute a reference image to mach. |
| 342 | 443 |
#outrastnames <- "reg1_mosaic_weights.tif" |
| ... | ... | |
| 348 | 449 |
|
| 349 | 450 |
## Create raster image for original predicted images with matching resolution and extent to the mosaic (reference image) |
| 350 | 451 |
|
| 351 |
rast_ref <- file.path(out_dir,"avg.tif") |
|
| 352 |
lf_files <- unlist(list_r_weights) |
|
| 452 |
rast_ref <- file.path(out_dir,"avg.tif") #this is a the ref |
|
| 453 |
|
|
| 454 |
|
|
| 455 |
### First match weights from linear option |
|
| 456 |
lf_files <- unlist(list_linear_r_weights) |
|
| 353 | 457 |
|
| 354 | 458 |
list_param_raster_match <- list(lf_files,rast_ref,file_format,out_suffix,out_dir) |
| 355 | 459 |
names(list_param_raster_match) <- c("lf_files","rast_ref","file_format","out_suffix","out_dir_str")
|
| ... | ... | |
| 357 | 461 |
#debug(raster_match) |
| 358 | 462 |
#r_test <- raster(raster_match(1,list_param_raster_match)) |
| 359 | 463 |
|
| 360 |
list_weights_m <- mclapply(1:length(lf_files),FUN=raster_match,list_param=list_param_raster_match,mc.preschedule=FALSE,mc.cores = num_cores) |
|
| 464 |
list_linear_weights_m <- mclapply(1:length(lf_files),FUN=raster_match,list_param=list_param_raster_match,mc.preschedule=FALSE,mc.cores = num_cores)
|
|
| 361 | 465 |
|
| 362 |
lf_files <- unlist(list_r_weights_prod) |
|
| 466 |
lf_files <- unlist(list_linear_r_weights_prod)
|
|
| 363 | 467 |
list_param_raster_match <- list(lf_files,rast_ref,file_format,out_suffix,out_dir) |
| 364 | 468 |
names(list_param_raster_match) <- c("lf_files","rast_ref","file_format","out_suffix","out_dir_str")
|
| 365 | 469 |
|
| 366 | 470 |
num_cores <-11 |
| 367 |
list_weights_prod_m <- mclapply(1:length(lf_files),FUN=raster_match,list_param=list_param_raster_match,mc.preschedule=FALSE,mc.cores = num_cores) |
|
| 471 |
list_linear_weights_prod_m <- mclapply(1:length(lf_files),FUN=raster_match,list_param=list_param_raster_match,mc.preschedule=FALSE,mc.cores = num_cores) |
|
| 472 |
|
|
| 473 |
|
|
| 474 |
### Second match wegihts from sine option |
|
| 475 |
|
|
| 476 |
lf_files <- unlist(list_sine_r_weights) |
|
| 477 |
|
|
| 478 |
list_param_raster_match <- list(lf_files,rast_ref,file_format,out_suffix,out_dir) |
|
| 479 |
names(list_param_raster_match) <- c("lf_files","rast_ref","file_format","out_suffix","out_dir_str")
|
|
| 368 | 480 |
|
| 481 |
#debug(raster_match) |
|
| 482 |
#r_test <- raster(raster_match(1,list_param_raster_match)) |
|
| 483 |
|
|
| 484 |
list_sine_weights_m <- mclapply(1:length(lf_files),FUN=raster_match,list_param=list_param_raster_match,mc.preschedule=FALSE,mc.cores = num_cores) |
|
| 485 |
|
|
| 486 |
lf_files <- unlist(list_sine_r_weights_prod) |
|
| 487 |
list_param_raster_match <- list(lf_files,rast_ref,file_format,out_suffix,out_dir) |
|
| 488 |
names(list_param_raster_match) <- c("lf_files","rast_ref","file_format","out_suffix","out_dir_str")
|
|
| 489 |
|
|
| 490 |
num_cores <-11 |
|
| 491 |
list_sine_weights_prod_m <- mclapply(1:length(lf_files),FUN=raster_match,list_param=list_param_raster_match,mc.preschedule=FALSE,mc.cores = num_cores) |
|
| 492 |
|
|
| 493 |
#### Third use original images |
|
| 369 | 494 |
#macth file to mosaic extent using the original predictions |
| 370 | 495 |
lf_files <- lf_mosaic |
| 371 | 496 |
list_param_raster_match <- list(lf_files,rast_ref,file_format,out_suffix,out_dir) |
| ... | ... | |
| 374 | 499 |
list_pred_m <- mclapply(1:length(lf_files),FUN=raster_match,list_param=list_param_raster_match,mc.preschedule=FALSE,mc.cores = num_cores) |
| 375 | 500 |
|
| 376 | 501 |
##################### |
| 377 |
###### PART 3: compute the weighted mean with the mosaic function #####
|
|
| 502 |
###### PART 4: compute the weighted mean with the mosaic function #####
|
|
| 378 | 503 |
|
| 379 | 504 |
#get the list of weights into raster objects |
| 380 |
list_args_weights <- list_weights_m |
|
| 505 |
list_args_linear_weights <- list_linear_weights_m |
|
| 506 |
#list_args_weights <- (mixedsort(list.files(pattern="r_weights_m_.*.tif"))) |
|
| 507 |
list_args_linear_weights <- lapply(1:length(list_args_linear_weights), FUN=function(i,x){raster(x[[i]])},x=list_args_linear_weights)
|
|
| 508 |
|
|
| 509 |
#get the list of weights product into raster objects |
|
| 510 |
list_args_linear_weights_prod <- list_linear_weights_prod_m |
|
| 511 |
#list_args_weights_prod <- (mixedsort(list.files(pattern="r_weights_prod_m_.*.tif"))) |
|
| 512 |
list_args_linear_weights_prod <- lapply(1:length(list_args_linear_weights_prod), FUN=function(i,x){raster(x[[i]])},x=list_args_linear_weights_prod)
|
|
| 513 |
|
|
| 514 |
#get the list of sine weights into raster objects |
|
| 515 |
list_args_sine_weights <- list_sine_weights_m |
|
| 381 | 516 |
#list_args_weights <- (mixedsort(list.files(pattern="r_weights_m_.*.tif"))) |
| 382 |
list_args_weights <- lapply(1:length(list_args_weights), FUN=function(i,x){raster(x[[i]])},x=list_args_weights)
|
|
| 517 |
list_args_sine_weights <- lapply(1:length(list_args_sine_weights), FUN=function(i,x){raster(x[[i]])},x=list_args_sine_weights)
|
|
| 383 | 518 |
|
| 384 | 519 |
#get the list of weights product into raster objects |
| 385 |
list_args_weights_prod <- list_weights_prod_m
|
|
| 520 |
list_args_sine_weights_prod <- list_sine_weights_prod_m
|
|
| 386 | 521 |
#list_args_weights_prod <- (mixedsort(list.files(pattern="r_weights_prod_m_.*.tif"))) |
| 387 |
list_args_weights_prod <- lapply(1:length(list_args_weights_prod), FUN=function(i,x){raster(x[[i]])},x=list_args_weights_prod)
|
|
| 522 |
list_args_sine_weights_prod <- lapply(1:length(list_args_sine_weights_prod), FUN=function(i,x){raster(x[[i]])},x=list_args_sine_weights_prod)
|
|
| 388 | 523 |
|
| 389 | 524 |
#get the original predicted image to raster (matched previously to the mosaic extent) |
| 390 | 525 |
list_args_pred_m <- list_pred_m |
| 391 | 526 |
#list_args_pred_m <- (mixedsort(list.files(pattern="^gam_CAI.*.m_mosaic_run10_1500x4500_global_analyses_03252015.tif"))) |
| 392 | 527 |
list_args_pred_m <- lapply(1:length(list_args_pred_m), FUN=function(i,x){raster(x[[i]])},x=list_args_pred_m)
|
| 393 | 528 |
|
| 394 |
list_args_weights$fun <- "sum" #we want the sum to compute the weighted mean |
|
| 395 |
list_args_weights$na.rm <- TRUE |
|
| 529 |
list_args_linear_weights$fun <- "sum" #we want the sum to compute the weighted mean
|
|
| 530 |
list_args_linear_weights$na.rm <- TRUE
|
|
| 396 | 531 |
|
| 397 |
list_args_weights_prod$fun <- "sum" |
|
| 398 |
list_args_weights_prod$na.rm <- TRUE |
|
| 532 |
list_args_linear_weights_prod$fun <- "sum" |
|
| 533 |
list_args_linear_weights_prod$na.rm <- TRUE |
|
| 534 |
|
|
| 535 |
list_args_sine_weights$fun <- "sum" #we want the sum to compute the weighted mean |
|
| 536 |
list_args_sine_weights$na.rm <- TRUE |
|
| 537 |
|
|
| 538 |
list_args_sine_weights_prod$fun <- "sum" |
|
| 539 |
list_args_sine_weights_prod$na.rm <- TRUE |
|
| 399 | 540 |
|
| 400 | 541 |
list_args_pred_m$fun <- "mean" |
| 401 | 542 |
list_args_pred_m$na.rm <- TRUE |
| 402 | 543 |
|
| 403 | 544 |
#Mosaic files |
| 404 |
r_weights_sum <- do.call(mosaic,list_args_weights) #weights sum image mosaiced |
|
| 405 |
r_prod_sum <- do.call(mosaic,list_args_weights_prod) #weights sum product image mosacied |
|
| 545 |
r_linear_weights_sum <- do.call(mosaic,list_args_linear_weights) #weights sum image mosaiced |
|
| 546 |
r_linear_prod_sum <- do.call(mosaic,list_args_linear_weights_prod) #weights sum product image mosacied |
|
| 547 |
|
|
| 548 |
r_m_linear_weighted_mean <- r_linear_prod_sum/r_linear_weights_sum #this is the mosaic using weighted mean... |
|
| 406 | 549 |
|
| 407 |
r_m_weighted_mean <- r_prod_sum/r_weights_sum #this is the mosaic using weighted mean... |
|
| 550 |
r_sine_weights_sum <- do.call(mosaic,list_args_sine_weights) #weights sum image mosaiced |
|
| 551 |
r_sine_prod_sum <- do.call(mosaic,list_args_sine_weights_prod) #weights sum product image mosacied |
|
| 408 | 552 |
|
| 409 |
raster_name <- file.path(out_dir,paste("r_m_weighted_mean",out_suffix,".tif",sep=""))
|
|
| 410 |
writeRaster(r_m_weighted_mean, NAflag=NA_flag_val,filename=raster_name,overwrite=TRUE) |
|
| 553 |
r_m_sine_weighted_mean <- r_sine_prod_sum/r_sine_weights_sum #this is the mosaic using weighted mean... |
|
| 554 |
|
|
| 555 |
raster_name <- file.path(out_dir,paste("r_m_linear_weighted_mean_",out_suffix,".tif",sep=""))
|
|
| 556 |
writeRaster(r_m_linear_weighted_mean, NAflag=NA_flag_val,filename=raster_name,overwrite=TRUE) |
|
| 557 |
|
|
| 558 |
raster_name <- file.path(out_dir,paste("r_m_sine_weighted_mean_",out_suffix,".tif",sep=""))
|
|
| 559 |
writeRaster(r_m_sine_weighted_mean, NAflag=NA_flag_val,filename=raster_name,overwrite=TRUE) |
|
| 411 | 560 |
|
| 412 | 561 |
r_m_mean <- do.call(mosaic,list_args_pred_m) #this is unweighted mean from the predicted raster |
| 413 | 562 |
|
| 414 |
raster_name <- file.path(out_dir,paste("r_m_mean",out_suffix,".tif",sep=""))
|
|
| 563 |
raster_name <- file.path(out_dir,paste("r_m_mean_",out_suffix,".tif",sep=""))
|
|
| 415 | 564 |
writeRaster(r_m_mean, NAflag=NA_flag_val,filename=raster_name,overwrite=TRUE) #unweighted mean |
| 416 | 565 |
|
| 566 |
r_diff_weighted_mean <- r_m_linear_weighted_mean - r_m_sine_weighted_mean |
|
| 567 |
#r_diff_weighted_mean<-r_diff_weighted_meam |
|
| 568 |
|
|
| 569 |
r_diff_mean_linear <- r_m_mean - r_m_linear_weighted_mean |
|
| 570 |
r_diff_mean_sine <- r_m_mean - r_m_sine_weighted_mean |
|
| 571 |
|
|
| 572 |
##################### |
|
| 573 |
###### PART 5: Now plot of the weighted mean and unweighted mean with the mosaic function ##### |
|
| 574 |
|
|
| 417 | 575 |
res_pix <- 1200 |
| 418 | 576 |
col_mfrow <- 1 |
| 419 | 577 |
row_mfrow <- 0.8 |
| ... | ... | |
| 429 | 587 |
col_mfrow <- 1 |
| 430 | 588 |
row_mfrow <- 0.8 |
| 431 | 589 |
|
| 432 |
png(filename=paste("Figure2_weigthed_mean_for_region_",region_name,"_",out_suffix,".png",sep=""),
|
|
| 590 |
png(filename=paste("Figure2_linear_weigthed_mean_for_region_",region_name,"_",out_suffix,".png",sep=""),
|
|
| 433 | 591 |
width=col_mfrow*res_pix,height=row_mfrow*res_pix) |
| 434 | 592 |
|
| 435 |
plot(r_m_weighted_mean)
|
|
| 593 |
plot(r_m_linear_weighted_mean )
|
|
| 436 | 594 |
|
| 437 | 595 |
dev.off() |
| 438 | 596 |
|
| 597 |
res_pix <- 1200 |
|
| 598 |
col_mfrow <- 1 |
|
| 599 |
row_mfrow <- 0.8 |
|
| 600 |
|
|
| 601 |
png(filename=paste("Figure2_sine_weigthed_mean_for_region_",region_name,"_",out_suffix,".png",sep=""),
|
|
| 602 |
width=col_mfrow*res_pix,height=row_mfrow*res_pix) |
|
| 603 |
|
|
| 604 |
plot(r_m_sine_weighted_mean) |
|
| 605 |
|
|
| 606 |
dev.off() |
|
| 607 |
|
|
| 608 |
res_pix <- 1200 |
|
| 609 |
col_mfrow <- 1 |
|
| 610 |
row_mfrow <- 0.8 |
|
| 611 |
|
|
| 612 |
png(filename=paste("Figure2_diff_linear_sine_weigthed_mean_for_region_",region_name,"_",out_suffix,".png",sep=""),
|
|
| 613 |
width=col_mfrow*res_pix,height=row_mfrow*res_pix) |
|
| 614 |
|
|
| 615 |
plot(r_diff_weighted_mean) |
|
| 616 |
|
|
| 617 |
dev.off() |
|
| 618 |
|
|
| 619 |
res_pix <- 1200 |
|
| 620 |
col_mfrow <- 1 |
|
| 621 |
row_mfrow <- 0.8 |
|
| 622 |
|
|
| 623 |
png(filename=paste("Figure2_diff_mean_linear_for_region_",region_name,"_",out_suffix,".png",sep=""),
|
|
| 624 |
width=col_mfrow*res_pix,height=row_mfrow*res_pix) |
|
| 625 |
|
|
| 626 |
plot(r_diff_mean_linear) |
|
| 627 |
|
|
| 628 |
dev.off() |
|
| 629 |
|
|
| 630 |
res_pix <- 1200 |
|
| 631 |
col_mfrow <- 1 |
|
| 632 |
row_mfrow <- 0.8 |
|
| 633 |
|
|
| 634 |
png(filename=paste("Figure2_diff_mean_sine_for_region_",region_name,"_",out_suffix,".png",sep=""),
|
|
| 635 |
width=col_mfrow*res_pix,height=row_mfrow*res_pix) |
|
| 636 |
|
|
| 637 |
plot(r_diff_mean_sine) |
|
| 638 |
|
|
| 639 |
dev.off() |
|
| 439 | 640 |
|
| 440 | 641 |
##################### END OF SCRIPT ###################### |
| 441 | 642 |
|
Also available in: Unified diff
mosaicing test scaling up adding sine weights and cleaning up code with other options