inst/datasets/f45_g37/1850_DAtest/B1850CN/pIKS/readCF.R
In garciapintado/rdafCESM: Setup and Assimilation of CESM
# notes:
# I will round timestamp for paleoclimate analysis to months. All model IO will be in months as a basic timestep
# Thus corresponding POSIXt will always end in "01-01 00:00:00" and will be UTC time
# this is IKP scenario for paper on methodological KS schemes.
CF <- list()
CF$path <- list()
CF$path$data <- dsndat
CF$region <- region
CF$event <- event
CF$scn <- scn
CF$model <- MODEL
CF$m <- 10 # ensemble members - match number on control variables for local sensitivity filters
CF$useGA <- TRUE # use [univariate, marginal, ensemble-based] Gaussian anamorphosis for gauDA[c('HE','y'] & MCgpar
CF$useGAy <- FALSE # cesmDA: whether to conduct the (~x,~y*) transformation (Amezcua 2012 notation): i.e. use x ensebmle to marginally transform y
CF$MThroughP <- 24.0 * 365 # [y/d] -> [d/d] == [h/h] Model Throughput [-] taken from a previous run for this resolution & compset and PES - used later to estimate walltime to request computing resources
CF$mpi <- FALSE # mpi for R-DA
CF$mcmode <- 1 # 1 for MonteCarlo simulation | 0 for parameter inheritance
CF$inh <- list() # non-empty list for mcmode == 0
CF$makeTruth <- FALSE # integrate a 'true' member instead of an ensemble. TODO: remove
#CF$truthext <- '000' # truth with no hosing | label to append to CASE_0 to identify a 'true' run
CF$truthext <- 'h01' # truth with hosing at GIS ~0.0075 Sv
CF$it_restart <- 0 # 0 for startup runs | >= 1 for restart - 1 does not build but runs from step 1 and jumps into assimilation at it=1
CF$delete_ocn_daily <- TRUE # in case history files for ocn are produced - delete these?
# just for temporary check - hosing routines with no freshwater input - M$POP$imau_gis <- 0
#CF$staTStr <- "1850-01-01 00:00:00" # paste(envrun$RUN_STARTDATE,'00:00:00 GMT', sep=' ')
#CF$endTStr <- "1851-01-01 00:00:00" # total integration times - including DA/forecast steps
#CF$makeTruth <- TRUE # integrate a 'true' member instead of an ensemble
#CF$truthext <- 'h00' # truth with hosing at 0.0 everywhere Sv
CF$hpc <- list() # definition of resources requested to HLRN
CF$hpc$account_name <- 'hbk00059' # project account
CF$hpc$nodes <- 6 # ppn:24
# map time into POSIX second times for DA analyses
CF$staTStr <- "1850-01-01 00:00:00" # paste(envrun$RUN_STARTDATE,'00:00:00 GMT', sep=' ')
CF$endTStr <- "1910-01-01 00:00:00" # total integration times - including DA/forecast steps [just 60 years to compare with the ETKF previously truncated integrations]
CF$tdiff <- 0
CF$dtREST <- 12*10 # [nmonths] NULL for no interruption [non continuation runs between sequential assimilation steps]
CF$REST_N <- 12*10 # >= (and exact multiple of) dtREST
CF$staT <- as.POSIXct(CF$staTStr, tz='GMT')
CF$endT <- as.POSIXct(CF$endTStr, tz='GMT')
CF$staTM <- CF$staT + CF$tdiff # start from year 1000 - equivalent model time
CF$endTM <- CF$endT + CF$tdiff
# observations: definition of observation yKIND,yTYPE,yFMT, where yFMT %in% c('g','p','s')
# y$KIND$TYPE$g - time-dense [gauge-like]
# s - space-dense [satellite-like]
# p - space-dense multi-time [assimilated via augmentation - fractional assimilation]
CF$y_class <- list() # observations: generic structure
CF$y_class$fname <- '' # If empty string [''], it must be set in readOb.R
CF$y_class$nto <- Inf # maximum number of times - for p_gen always Inf: there is just 1 observation/location
CF$y_class$qc <- 'none' # QC method
CF$y_class$r <- NA # scalar observation variance
CF$y_class$twin <- Inf # [s] maximum assimilation window from present to past [Inf: all integration run]
CF$y_class$use <- TRUE # assimilate or not
CF$y_class$w <- FALSE # TRUE for weigthed observations
CF$yfnames <- c('y_MARGOloc_KIND_ocn_sst.R') # syntethic SST at MARGO locations
CF$gauDAfname <- file.path(CASEROOT_0,'DA','gauDA_MARGOpos_KIND_ocn_sst.rds')
#CF$yfnames <- c('y_szon.R','y_forceAlbedos.R')
#CF$yfnames <- c('y_PAGES2k_coral.R')
#CF$gauDAfname <- file.path(CASEROOT_0,'DA','gauDA_PAGES2K_coral.rds')
CF$IMAU_ON <- TRUE # activate/deactive hosing into POP2 via IMAU FW
#y <- list() # empty list for no observations
#y$KIND_atm_sft$proxy1$s <- list() # atm - surface temperature
#y$KIND_atm_sft$proxy1$s$use <- FALSE
ana_class <- list() # generic analysis item
ana_class$u <- TRUE # whether this item is updated by the analysis
ana_class$ll <- 0.0 # 0.0 for global filtering
ana_class$infac <- 1.0 # 1.0 for no inflation <-> 0.0 for maximum inflation [recovery of background covariance]
ana_class$min <- -Inf # no lower bound: used in transformation (anamorphosis) and post-asimilation checks
ana_class$max <- Inf # no upper bound: ditto
ana_class$pos <- c(NA,NA) # [NA,NA] => no localisation
ana_class$trf <- 'GA' # 'none' for no transformation | 'autoGA' triggers Gaussian Anamorphosis based on Wilk-Shapiro test | 'GA' for Gaussian Anamorphosis
# not used for local sensitivity filters (based on marginal perturbations)
# parameters to check in connection to:
# atm. co2 forcing
# cloud emmisivity
# horizontal transport of sensible heat & horizontal transport of latent heat
# precipitation rates (e.g. threshold of relative huidity to start precipitating clouds)
# something to discrimate between low and high level clouds.
# ice. sea albedo
# land. snow albedo
# ocn. vertical (background values in KPP) and horizontal (isopycnal in current models) diffusion (Gent equation) [what is turbulence closure?]
ana <- list() # variables to analyse
#ana$KIND_albicei <- ana_class
#ana$KIND_albicei$trf <- 'GA'
#ana$KIND_albicev <- ana_class
#ana$KIND_albicev$trf <- 'GA'
#ana$KIND_coral <- ana_class
ana$KIND_ocn_sst <- ana_class # gridded variable matching MARGO dataset
#ana$KIND_ocn_sst$trf <- 'GA'
#ana$KIND_ocn_sst$ll <- 5000*km
#ana$KIND_szon <- ana_class
#ana$KIND_szon$min <- 0.0
#ana$KIND_szon$trf <- 'GA'
# parameters model global parameters - must be synchronised with header_mc.R
ana$CAM.cldfrc_rhminh <- ana_class # [min,max] CAM4:high stable cloud critical relative humidity [truth] in CAM4 was 0.80 - range as Jacson_al2008
ana$CAM.cldfrc_rhminl <- ana_class # [min,max] CAM4:low stable cloud critical relative humidity [truth] in CAM4 was 0.91 - range as Jacson_al2008
ana$CAM.ch4vmr <- ana_class # [mu,sd] CAM: CH4 volume mixing ratio. Invariant surface value of CH4 if no time varying values are specified
ana$CAM.co2vmr <- ana_class # [mu,sd] CAM: CO2 volumne mixing ratio. Invariant surface value of CO2 if no time varying values are specified
ana$CAM.zmconv_c0_lnd <- ana_class # [min,max] as Yang_al2013 for CAM5 | truth 0.0035 | Deep convective precipitation efficiency over land [m^{-1}]
ana$CAM.zmconv_c0_ocn <- ana_class # [min,max] as Yang_al2013 for CAM5 | truth 0.0035 | Deep convective precipitation efficiency over ocean [m^{-1}]
ana$CAM.zmconv_ke <- ana_class # [min,max] as Yang_al2013 for CAM5 | truth 1.0E-06| Evaporation efficiency of precipitation [(kg m^{-2} s^{-1})^{-1/2}s^{-1}]
ana$CICE.albicei <- ana_class
ana$CICE.albicei$u <- FALSE
ana$CICE.albicev <- ana_class
ana$CICE.albicev$u <- FALSE
ana$POP2.hmix_gm_nml.ah <- ana_class
ana$POP2.ah_bolus <- ana_class # constrained to == POP2.hmix_gm_nml.ah
ana$POP2.bckgrnd_vdc1 <- ana_class
ana$POP2.bckgrnd_vdc2 <- ana_class
ana$POP2.bckgrnd_vdc2$u <- FALSE # constant parameterization along vertical
ana$POP2.imau_gis <- ana_class # reported as POP2.freshwater_gis
#ana$POP2.imau_gis$trf <- 'autoGA'
ana$CICE.albicei$min <- 0.0
ana$CICE.albicev$min <- 0.0
ana$CICE.albicei$max <- 1.0
ana$CICE.albicev$max <- 1.0
#This should be replaced by all variable names in POP2.
# Specific analysed variables are in vector POP2vnamesU declared in the main caller
POP2vnames <- c('FW','HMXL','IFRAC','PREC_F','SALT','TAUX','TAUY','TEMP','UVEL','VVEL','WVEL')
for (vname in POP2vnames) {
ana[[paste('POP2',vname,sep='.')]] <- ana_class
ana[[paste('POP2',vname,sep='.')]]$trf <- 'none'
#ana[[paste('POP2',vname,sep='.')]]$ll <- 5000*km
}
ana$POP2.SALT$min <- 0
ana$POP2.SALT$trf <- 'GA'
ana$POP2.PREC_F$min <- 0
ana$POP2.PREC_F$trf <- 'GA'
ana$POP2.HMXL$min <- 0
ana$POP2.HMXL$trf <- 'GA'
ana$POP2.IFRAC$min <- 0
ana$POP2.IFRAC$trf <- 'GA'
CF$ana <- ana; rm(ana)
CF$ana_class <- ana_class; rm(ana_class)
garciapintado/rdafCESM documentation built on July 18, 2019, 4:41 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
# notes:
# I will round timestamp for paleoclimate analysis to months. All model IO will be in months as a basic timestep
# Thus corresponding POSIXt will always end in "01-01 00:00:00" and will be UTC time
# this is IKP scenario for paper on methodological KS schemes.
CF <- list()
CF$path <- list()
CF$path$data <- dsndat
CF$region <- region
CF$event <- event
CF$scn <- scn
CF$model <- MODEL
CF$m <- 10 # ensemble members - match number on control variables for local sensitivity filters
CF$useGA <- TRUE # use [univariate, marginal, ensemble-based] Gaussian anamorphosis for gauDA[c('HE','y'] & MCgpar
CF$useGAy <- FALSE # cesmDA: whether to conduct the (~x,~y*) transformation (Amezcua 2012 notation): i.e. use x ensebmle to marginally transform y
CF$MThroughP <- 24.0 * 365 # [y/d] -> [d/d] == [h/h] Model Throughput [-] taken from a previous run for this resolution & compset and PES - used later to estimate walltime to request computing resources
CF$mpi <- FALSE # mpi for R-DA
CF$mcmode <- 1 # 1 for MonteCarlo simulation | 0 for parameter inheritance
CF$inh <- list() # non-empty list for mcmode == 0
CF$makeTruth <- FALSE # integrate a 'true' member instead of an ensemble. TODO: remove
#CF$truthext <- '000' # truth with no hosing | label to append to CASE_0 to identify a 'true' run
CF$truthext <- 'h01' # truth with hosing at GIS ~0.0075 Sv
CF$it_restart <- 0 # 0 for startup runs | >= 1 for restart - 1 does not build but runs from step 1 and jumps into assimilation at it=1
CF$delete_ocn_daily <- TRUE # in case history files for ocn are produced - delete these?
# just for temporary check - hosing routines with no freshwater input - M$POP$imau_gis <- 0
#CF$staTStr <- "1850-01-01 00:00:00" # paste(envrun$RUN_STARTDATE,'00:00:00 GMT', sep=' ')
#CF$endTStr <- "1851-01-01 00:00:00" # total integration times - including DA/forecast steps
#CF$makeTruth <- TRUE # integrate a 'true' member instead of an ensemble
#CF$truthext <- 'h00' # truth with hosing at 0.0 everywhere Sv
CF$hpc <- list() # definition of resources requested to HLRN
CF$hpc$account_name <- 'hbk00059' # project account
CF$hpc$nodes <- 6 # ppn:24
# map time into POSIX second times for DA analyses
CF$staTStr <- "1850-01-01 00:00:00" # paste(envrun$RUN_STARTDATE,'00:00:00 GMT', sep=' ')
CF$endTStr <- "1910-01-01 00:00:00" # total integration times - including DA/forecast steps [just 60 years to compare with the ETKF previously truncated integrations]
CF$tdiff <- 0
CF$dtREST <- 12*10 # [nmonths] NULL for no interruption [non continuation runs between sequential assimilation steps]
CF$REST_N <- 12*10 # >= (and exact multiple of) dtREST
CF$staT <- as.POSIXct(CF$staTStr, tz='GMT')
CF$endT <- as.POSIXct(CF$endTStr, tz='GMT')
CF$staTM <- CF$staT + CF$tdiff # start from year 1000 - equivalent model time
CF$endTM <- CF$endT + CF$tdiff
# observations: definition of observation yKIND,yTYPE,yFMT, where yFMT %in% c('g','p','s')
# y$KIND$TYPE$g - time-dense [gauge-like]
# s - space-dense [satellite-like]
# p - space-dense multi-time [assimilated via augmentation - fractional assimilation]
CF$y_class <- list() # observations: generic structure
CF$y_class$fname <- '' # If empty string [''], it must be set in readOb.R
CF$y_class$nto <- Inf # maximum number of times - for p_gen always Inf: there is just 1 observation/location
CF$y_class$qc <- 'none' # QC method
CF$y_class$r <- NA # scalar observation variance
CF$y_class$twin <- Inf # [s] maximum assimilation window from present to past [Inf: all integration run]
CF$y_class$use <- TRUE # assimilate or not
CF$y_class$w <- FALSE # TRUE for weigthed observations
CF$yfnames <- c('y_MARGOloc_KIND_ocn_sst.R') # syntethic SST at MARGO locations
CF$gauDAfname <- file.path(CASEROOT_0,'DA','gauDA_MARGOpos_KIND_ocn_sst.rds')
#CF$yfnames <- c('y_szon.R','y_forceAlbedos.R')
#CF$yfnames <- c('y_PAGES2k_coral.R')
#CF$gauDAfname <- file.path(CASEROOT_0,'DA','gauDA_PAGES2K_coral.rds')
CF$IMAU_ON <- TRUE # activate/deactive hosing into POP2 via IMAU FW
#y <- list() # empty list for no observations
#y$KIND_atm_sft$proxy1$s <- list() # atm - surface temperature
#y$KIND_atm_sft$proxy1$s$use <- FALSE
ana_class <- list() # generic analysis item
ana_class$u <- TRUE # whether this item is updated by the analysis
ana_class$ll <- 0.0 # 0.0 for global filtering
ana_class$infac <- 1.0 # 1.0 for no inflation <-> 0.0 for maximum inflation [recovery of background covariance]
ana_class$min <- -Inf # no lower bound: used in transformation (anamorphosis) and post-asimilation checks
ana_class$max <- Inf # no upper bound: ditto
ana_class$pos <- c(NA,NA) # [NA,NA] => no localisation
ana_class$trf <- 'GA' # 'none' for no transformation | 'autoGA' triggers Gaussian Anamorphosis based on Wilk-Shapiro test | 'GA' for Gaussian Anamorphosis
# not used for local sensitivity filters (based on marginal perturbations)
# parameters to check in connection to:
# atm. co2 forcing
# cloud emmisivity
# horizontal transport of sensible heat & horizontal transport of latent heat
# precipitation rates (e.g. threshold of relative huidity to start precipitating clouds)
# something to discrimate between low and high level clouds.
# ice. sea albedo
# land. snow albedo
# ocn. vertical (background values in KPP) and horizontal (isopycnal in current models) diffusion (Gent equation) [what is turbulence closure?]
ana <- list() # variables to analyse
#ana$KIND_albicei <- ana_class
#ana$KIND_albicei$trf <- 'GA'
#ana$KIND_albicev <- ana_class
#ana$KIND_albicev$trf <- 'GA'
#ana$KIND_coral <- ana_class
ana$KIND_ocn_sst <- ana_class # gridded variable matching MARGO dataset
#ana$KIND_ocn_sst$trf <- 'GA'
#ana$KIND_ocn_sst$ll <- 5000*km
#ana$KIND_szon <- ana_class
#ana$KIND_szon$min <- 0.0
#ana$KIND_szon$trf <- 'GA'
# parameters model global parameters - must be synchronised with header_mc.R
ana$CAM.cldfrc_rhminh <- ana_class # [min,max] CAM4:high stable cloud critical relative humidity [truth] in CAM4 was 0.80 - range as Jacson_al2008
ana$CAM.cldfrc_rhminl <- ana_class # [min,max] CAM4:low stable cloud critical relative humidity [truth] in CAM4 was 0.91 - range as Jacson_al2008
ana$CAM.ch4vmr <- ana_class # [mu,sd] CAM: CH4 volume mixing ratio. Invariant surface value of CH4 if no time varying values are specified
ana$CAM.co2vmr <- ana_class # [mu,sd] CAM: CO2 volumne mixing ratio. Invariant surface value of CO2 if no time varying values are specified
ana$CAM.zmconv_c0_lnd <- ana_class # [min,max] as Yang_al2013 for CAM5 | truth 0.0035 | Deep convective precipitation efficiency over land [m^{-1}]
ana$CAM.zmconv_c0_ocn <- ana_class # [min,max] as Yang_al2013 for CAM5 | truth 0.0035 | Deep convective precipitation efficiency over ocean [m^{-1}]
ana$CAM.zmconv_ke <- ana_class # [min,max] as Yang_al2013 for CAM5 | truth 1.0E-06| Evaporation efficiency of precipitation [(kg m^{-2} s^{-1})^{-1/2}s^{-1}]
ana$CICE.albicei <- ana_class
ana$CICE.albicei$u <- FALSE
ana$CICE.albicev <- ana_class
ana$CICE.albicev$u <- FALSE
ana$POP2.hmix_gm_nml.ah <- ana_class
ana$POP2.ah_bolus <- ana_class # constrained to == POP2.hmix_gm_nml.ah
ana$POP2.bckgrnd_vdc1 <- ana_class
ana$POP2.bckgrnd_vdc2 <- ana_class
ana$POP2.bckgrnd_vdc2$u <- FALSE # constant parameterization along vertical
ana$POP2.imau_gis <- ana_class # reported as POP2.freshwater_gis
#ana$POP2.imau_gis$trf <- 'autoGA'
ana$CICE.albicei$min <- 0.0
ana$CICE.albicev$min <- 0.0
ana$CICE.albicei$max <- 1.0
ana$CICE.albicev$max <- 1.0
#This should be replaced by all variable names in POP2.
# Specific analysed variables are in vector POP2vnamesU declared in the main caller
POP2vnames <- c('FW','HMXL','IFRAC','PREC_F','SALT','TAUX','TAUY','TEMP','UVEL','VVEL','WVEL')
for (vname in POP2vnames) {
ana[[paste('POP2',vname,sep='.')]] <- ana_class
ana[[paste('POP2',vname,sep='.')]]$trf <- 'none'
#ana[[paste('POP2',vname,sep='.')]]$ll <- 5000*km
}
ana$POP2.SALT$min <- 0
ana$POP2.SALT$trf <- 'GA'
ana$POP2.PREC_F$min <- 0
ana$POP2.PREC_F$trf <- 'GA'
ana$POP2.HMXL$min <- 0
ana$POP2.HMXL$trf <- 'GA'
ana$POP2.IFRAC$min <- 0
ana$POP2.IFRAC$trf <- 'GA'
CF$ana <- ana; rm(ana)
CF$ana_class <- ana_class; rm(ana_class)
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.