inst/scripts/99.abundance_estimation_stmv.r
In jae0/snowcrab: Snow crab assessment suite using aegis* and associated projects and data streams.
year.assessment = 2019
p = bio.snowcrab::load.environment( year.assessment=year.assessment )
# 11 hrs with these settings
ncpus = parallel::detectCores()
stmv_clusters = list(
scale=rep("localhost", ncpus),
interpolate=rep("localhost", ncpus)
)
p = snowcrab_stmv(
p=p,
project_class="stmv",
stmv_variables=list(Y="snowcrab.large.males_abundance"),
selection=list(
type = "biomass",
biologicals=list(
sex=0, # male
mat=1, # do not use maturity status in groundfish data as it is suspect ..
spec_bio=bio.taxonomy::taxonomy.recode( from="spec", to="parsimonious", tolookup=2526 ),
len= c( 95, 200 )/10, # mm -> cm ; aegis_db in cm
ranged_data="len"
),
survey=list(
data.source = c("snowcrab"),
yr = p$yrs # time frame for comparison specified above
)
),
DATA = 'snowcrab_stmv( p=p, DS="stmv_inputs" )',
stmv_global_modelengine ="gam",
stmv_global_family = gaussian(link="log"),
stmv_local_modelengine = "twostep",
stmv_twostep_time = "gam",
stmv_twostep_space = "fft",
stmv_fft_filter="matern", # matern, krige (very slow), lowpass, lowpass_matern
stmv_gam_optimizer = c("outer", "bfgs") ,
stmv_distance_statsgrid = 3, # resolution (km) of data aggregation (i.e. generation of the ** statistics ** ),
stmv_distance_scale = c( 25, 35, 45 ), #likely must be over 30km, so 50 +/- 20km, should likely match the setting in ~ line 256
stmv_clusters = stmv_clusters
)
if (0) {
p$stmv_global_modelformula = formula( paste(
' snowcrab.large.males_abundance',
' ~ s( t, k=3, bs="ts") + s( tsd, k=3, bs="ts") + s( degreedays, k=3, bs="ts") ',
' + s( log(z), k=3, bs="ts") + s( log(dZ), k=3, bs="ts") + s( log(ddZ), k=3, bs="ts") ',
' + s( substrate.grainsize, k=3, bs="ts") ',
' + s(pca1, bs="ts") + s(pca2, bs="ts") '
))
p$stmv_global_modelformula = formula( paste(
' snowcrab.large.males_abundance',
' ~ s( t, k = 4, bs="ts") + s( tmax, k = 4, bs="ts") + s( degreedays, k = 4, bs="ts") ',
' + s( log(z), k=4, bs="ts") + s( log(dZ), k=4, bs="ts") + s( log(ddZ), k=4, bs="ts") ',
' + s( substrate.grainsize, k=4, bs="ts") + s(pca1, k=4, bs="ts") + s(pca2, k=4, bs="ts") '
))
p = stmv_variablelist(p=p) # decompose into covariates, etc
o = snowcrab_stmv(p=p, DS="input_data" ) # create fields for
global_model = gam(
formula=p$stmv_global_modelformula,
family=p$stmv_global_family,
data = o,
weights=o$wt,
optimizer= p$stmv_gam_optimizer,
na.action="na.omit"
)
}
#Run the following line if you want to use maptools rather than GADMTools for mapping coastline
# p$DATA = 'snowcrab_stmv( p=p, DS="stmv_inputs", coastline_source="mapdata.coastPolygon" )'
# range( INP$snowcrab.large.males_abundance )
# [1] 14.3 6675.0
# o = snowcrab_stmv(p=p, DS="stmv_inputs" ) # create fields for
#--------------------------------------------------------------
# Run the process
#--------------------------------------------------------------
stmv( p=p, runmode=c("globalmodel", "interpolate" ) ) # for a clean start
snowcrab_stmv( p=p, DS="predictions.redo" ) # warp predictions to other grids (if any)
snowcrab_stmv( p=p, DS="stmv.stats.redo" ) # warp stats to other grids (if any)
snowcrab_stmv( p=p, DS="complete.redo" )
snowcrab_stmv( p=p, DS="baseline.redo" )
snowcrab_stmv( p=p, DS="map.all" )
global_model = stmv_global_model( p=p, DS="global_model")
summary( global_model )
par(mar=c(1,1,1,1)) #change plot margins for Rstudio
plot(global_model)
# 2018 results
# Family: gaussian
# Link function: log
#
# Formula:
# snowcrab.large.males_abundance ~ s(t, k=3, bs="ts") + s(tsd,
# k=3, bs="ts") + s(tmax, k=3, bs="ts") + s(degreedays,
# k=3, bs="ts") + s(log(z), k=3, bs="ts") + s(log(dZ),
# k=3, bs="ts") + s(log(ddZ), k=3, bs="ts") + s(substrate.grainsize,
# k=3, bs="ts") + s(pca1, k=3, bs="ts") + s(pca2, k=3,
# bs="ts")
#
# Parametric coefficients:
# Estimate Std. Error t value Pr(>|t|)
# (Intercept) 6.7468 0.0204 330 <2e-16
#
# Approximate significance of smooth terms:
# edf Ref.df F p-value
# s(t) 1.91 2 73.0 < 2e-16
# s(tsd) 2.00 2 14.0 8.3e-07
# s(tmax) 1.91 2 11.7 1.4e-06
# s(degreedays) 1.36 2 45.4 < 2e-16
# s(log(z)) 1.91 2 174.7 < 2e-16
# s(log(dZ)) 1.93 2 20.9 3.8e-10
# s(log(ddZ)) 1.95 2 61.5 < 2e-16
# s(substrate.grainsize) 1.99 2 51.0 < 2e-16
# s(pca1) 2.00 2 100.4 < 2e-16
# s(pca2) 1.94 2 95.3 < 2e-16
#
# R-sq.(adj) = 0.232 Deviance explained = 23.4%
# GCV = 6198.2 Scale est. = 6182 n = 7640
# -------------------------------------------------------------------------------------
# STEP TWO commercial presence /absence
#--------------------------------------------------------------------------------------
# presence-absence
# this takes about 40 hrs ... and 5-6 GB /process
# year.assessment = 2018
p = bio.snowcrab::load.environment( year.assessment=year.assessment )
ncpus = parallel::detectCores()
p = bio.snowcrab::snowcrab_parameters(
p=p,
project_class="stmv",
stmv_variables=list(Y="snowcrab.large.males_presence_absence"),
selection=list(
type = "presence_absence",
biologicals = list(
sex=0, # male
mat=1, # do not use maturity status in groundfish data as it is suspect ..
spec_bio=bio.taxonomy::taxonomy.recode( from="spec", to="parsimonious", tolookup=2526 ),
len= c( 95, 200 )/10, # mm -> cm ; aegis_db in cm
ranged_data="len"
),
survey=list(
data.source = c("snowcrab", "groundfish"), # add groundfish data too
drop.unreliable.zeros.groundfish.data=TRUE, # esp from 1970 to 1999 measurement of invertebrates was sporatic .. zero-values are dropped as they are unreliable
yr = p$yrs # time frame for comparison specified above
)
),
DATA = 'snowcrab_stmv( p=p, DS="stmv_inputs" )',
aegis_project_datasources = c("speciescomposition" ), # c("speciescomposition", "speciesarea", "sizespectrum", "condition", "metabolism", "biochem")
stmv_global_family = binomial( link="logit" ),
stmv_global_modelengine ="gam",
stmv_local_modelengine = "twostep",
stmv_twostep_time = "gam",
stmv_twostep_space = "fft",
stmv_fft_filter="matern", # matern, krige (very slow), lowpass, lowpass_matern
stmv_gam_optimizer=c("outer", "bfgs") ,
stmv_distance_statsgrid = 3, # resolution (km) of data aggregation (i.e. generation of the ** statistics ** ),
stmv_distance_scale = c( 25, 35, 45 ), #likely must be over 30km, so 50 +/- 20km, should likely match the setting in ~ line 256
stmv_clusters = list( scale=rep("localhost", ncpus), interpolate=rep("localhost", ncpus) )
)
if (0) {
p$stmv_global_modelformula = formula( paste(
' snowcrab.large.males_abundance',
' ~ s( t, k=3, bs="ts") + s( tsd, k=3, bs="ts") + s( degreedays, k=3, bs="ts") ',
' + s( t, tsd, degreedays, bs="ts") ',
' + s( log(z), k=3, bs="ts") + s( log(dZ), k=3, bs="ts") + s( log(ddZ), k=3, bs="ts") ',
' + s( log(z), log(dZ), log(ddZ), bs="ts") ',
' + s( substrate.grainsize, k=3, bs="ts") ',
' + s(pca1, k=3, bs="ts") + s(pca2, k=3, bs="ts") + s(pca1, pca2, k=8, bs="ts") '
))
p$stmv_global_modelformula = formula( paste(
' snowcrab.large.males_abundance',
' ~ s( t, k = 4, bs="ts") + s( tmax, k = 4, bs="ts") + s( degreedays, k = 4, bs="ts") ',
' + s( log(z), k=4, bs="ts") + s( log(dZ), k=4, bs="ts") + s( log(ddZ), k=4, bs="ts") ',
' + s( substrate.grainsize, k=4, bs="ts") + s(pca1, k=4, bs="ts") + s(pca2, k=4, bs="ts") '
))
o = snowcrab_stmv(p=p, DS="input_data" ) # create fields for
global_model = gam(
formula=p$stmv_global_modelformula,
family=p$stmv_global_family,
data = o,
weights=o$wt,
optimizer= p$stmv_gam_optimizer,
na.action="na.omit"
)
}
# p$DATA = 'snowcrab_stmv( p=p, DS="stmv_inputs", coastline_source="mapdata.coastPolygon" )'
# o = snowcrab_stmv(p=p, DS="stmv_inputs" ) # create fields for
stmv( p=p, runmode=c("globalmodel", "interpolate" ) ) # no global_model and force a clean restart
snowcrab_stmv( p=p, DS="predictions.redo" ) # warp predictions to other grids
snowcrab_stmv( p=p, DS="stmv.stats.redo" ) # warp stats to other grids
snowcrab_stmv( p=p, DS="complete.redo" )
snowcrab_stmv( p=p, DS="baseline.redo" )
snowcrab_stmv( p=p, DS="map.all" )
global_model = stmv_global_model( p=p, DS="global_model")
summary( global_model )
par(mar=c(1,1,1,1)) #change plot margins for Rstudio
plot(global_model, all.terms=TRUE, trans=bio.snowcrab::inverse.logit, seWithMean=TRUE, jit=TRUE, rug=TRUE )
# Family: binomial
# Link function: logit
#
# Formula:
# snowcrab.large.males_presence_absence ~ s(t, k=3, bs="ts") +
# s(tsd, k=3, bs="ts") + s(tmax, k=3, bs="ts") + s(degreedays,
# k=3, bs="ts") + s(log(z), k=3, bs="ts") + s(log(dZ),
# k=3, bs="ts") + s(log(ddZ), k=3, bs="ts") + s(substrate.grainsize,
# k=3, bs="ts") + s(pca1, k=3, bs="ts") + s(pca2, k=3,
# bs="ts")
#
# Parametric coefficients:
# Estimate Std. Error z value Pr(>|z|)
# (Intercept) 0.0302 0.0349 0.87 0.39
#
# Approximate significance of smooth terms:
# edf Ref.df Chi.sq p-value
# s(t) 1.99 2 336.7 < 2e-16
# s(tsd) 2.00 2 35.3 2.0e-08
# s(tmax) 1.98 2 59.9 7.0e-14
# s(degreedays) 1.99 2 188.4 < 2e-16
# s(log(z)) 2.00 2 1282.7 < 2e-16
# s(log(dZ)) 1.99 2 68.7 9.1e-16
# s(log(ddZ)) 1.98 2 224.4 < 2e-16
# s(substrate.grainsize) 1.98 2 394.4 < 2e-16
# s(pca1) 2.00 2 932.6 < 2e-16
# s(pca2) 2.00 2 1291.8 < 2e-16
#
# R-sq.(adj) = 0.62 Deviance explained = 56.2%
# UBRE = -0.58807 Scale est. = 1 n = 36031
#
# collect all predictions into a single file and return:
# year.assessment=2017
#------------------------------------------------------------------------------
# STEP THREE- Predict biomass by weighting +/- with probabilities
#------------------------------------------------------------------------------
p = bio.snowcrab::load.environment( year.assessment=year.assessment )
p = snowcrab_stmv(
p=p,
project_class="stmv",
stmv_variables = list(Y="snowcrab.large.males_abundance"),
selection=list(
type = "biomass",
biologicals=list(
sex=0, # male
mat=1, # do not use maturity status in groundfish data as it is suspect ..
spec_bio=bio.taxonomy::taxonomy.recode( from="spec", to="parsimonious", tolookup=2526 ),
len= c( 95, 200 )/10, # mm -> cm ; aegis_db in cm
ranged_data="len"
),
survey=list(
data.source = c("snowcrab"),
yr = p$yrs, # time frame for comparison specified above
drop.unreliable.zeros.groundfish.data=TRUE # esp from 1970 to 1999 measurement of invertebrates was sporatic .. zero-values are dropped as they are unreliable
)
)
)
interpolation.db( DS="fishable.biomass.redo", p=p ) # combine habitat and abundance info and map
K = interpolation.db( DS="fishable.biomass.timeseries", p=p )
if (0){
str(K)
table.view( K )
plot( total ~ yr, K[K$region=="cfanorth", ], type="b")
plot( total ~ yr, K[K$region=="cfasouth", ], type="b")
plot( total ~ yr, K[K$region=="cfa4x", ], type="b")
}
interpolation.db( DS="fishable.biomass.map", p=p )
figure.timeseries.snowcrab.habitat(p=p) # /bio.data/bio.snowcrab/assessments/2016/timeseries/interpolated/snowcrab.habitat.sa.png
figure.timeseries.snowcrab.habitat.temperatures(p=p) # /bio.data/bio.snowcrab/assessments/2016/timeseries/interpolated/mean.bottom.temp.snowcrab.habitat.png
### --------- prediction success:
set = snowcrab_stmv(p=p, DS="input_data" )
S = set[ , c("plon", "plat") ]
ii = array_map( "xy->1", S, gridparams=p$gridparams )
bs = bathymetry_db(p=p, DS="baseline")
bb = array_map( "xy->1", bs, gridparams=p$gridparams )
im = match( ii, bb )
it = match( set$yr, p$yrs )
bm = interpolation.db( DS="fishable.biomass", p=p )
spred = bm$m[cbind(im, it)] # approximate match (ignoring seasonality)
summary ( lm(log(spred)~log(snowcrab.large.males_abundance), data=set, na.actio="na.omit" ) )
plot(log(spred)~log(snowcrab.large.males_abundance), data=set )
cor(log(spred),log(set$snowcrab.large.males_abundance), use="complete.obs")
# Call:
# lm(formula = log(spred) ~ log(snowcrab.large.males_abundance),
# data = set, na.action = "na.omit")
#
# Residuals:
# Min 1Q Median 3Q Max
# -2.3039 -0.5507 0.0589 0.6056 2.1379
#
# Coefficients:
# Estimate Std. Error t value Pr(>|t|)
# (Intercept) 5.17806 0.03824 135.4 <2e-16
# log(snowcrab.large.males_abundance) 0.19035 0.00599 31.8 <2e-16
#
# Residual standard error: 0.795 on 5030 degrees of freedom
# (2223 observations deleted due to missingness)
# Multiple R-squared: 0.167, Adjusted R-squared: 0.167
# F-statistic: 1.01e+03 on 1 and 5030 DF, p-value: <2e-16
#
# R> plot(log(spred)~log(snowcrab.large.males_abundance), data=set )
# R> cor(log(spred),log(set$snowcrab.large.males_abundance), use="complete.obs")
# [1] 0.4269
# determine presence absence(Y) and weighting(wt)
# set$weekno = floor(set$julian / 365 * 52) + 1
# set$dyear = floor(set$julian / 365 ) + 1
## END
# figures and tables related to fishery indices
#Pick whichever year reference below is correct (most often year.assessment...-1)
if (!exists("year.assessment")) {
year.assessment=lubridate::year(Sys.Date())
year.assessment=lubridate::year(Sys.Date()) - 1
}
p = bio.snowcrab::load.environment( year.assessment=year.assessment )
# ------------------------------------------
# Time-series: all interpolated data estimated from interpolated analysis
# BZ R0.mass is likely the only variable required
figure.interpolated.results( p, outdir=file.path( p$annual.results, "timeseries", "interpolated" ), alt.zero.y=T )
# ------------------------------------------
# Time-series: immature male numerical abundance estimates from interpolated analysis (broken down by instar)
figure.immature.male( p, outdir=file.path( p$annual.results, "timeseries", "interpolated" ) )
# ------------------------------------------
# Time-series: immature male skip-moulter numerical abundance estimates from interpolated analysis (broken down by instar)
figure.immature.skipmoulter.male( p, outdir=file.path( p$annual.results, "timeseries", "interpolated" ) )
# ------------------------------------------
# Time-series: mature CC12 male numerical abundance estimates from interpolated analysis (broken down by instar)
figure.mature.CC12.male( p, outdir=file.path( p$annual.results, "timeseries", "interpolated" ) )
# ------------------------------------------
# Time-series: mature CC34 male numerical abundance estimates from interpolated analysis (broken down by instar)
figure.mature.CC34.male( p, outdir=file.path( p$annual.results, "timeseries", "interpolated" ) )
# ------------------------------------------
# Time-series: mature CC5 male numerical abundance estimates from interpolated analysis (broken down by instar)
figure.mature.CC5.male( p, outdir=file.path( p$annual.results, "timeseries", "interpolated" ) )
# ------------------------------------------
# Time-series: Exploitation rate
figure.exploitation.rate( p, outdir=file.path( p$annual.results, "timeseries", "interpolated" ) )
figure.exploitation.rate( p, outdir=file.path( p$annual.results, "timeseries", "interpolated" ), CFA4X.exclude.year.assessment=F )
# ------------------------------------------
# Time-series: Generic plots of all interpolated data
# figure.timeseries.interpolated( p, outdir=file.path( p$annual.results, "timeseries", "interpolated" ) )
# ------------------------------------------
# Time-series: Habitat variations (surface area of snow crab habitat)
# p$model.type = "gam.full" # choose method for habitat model :
# p$model.type = "gam.simple" # choose method for habitat model : no longer used
p$habitat.threshold.quantile = 0.05 # quantile at which to consider zero-valued abundance
p$prediction_dyear = 9/12 # predict for ~ Sept 1
figure.timeseries.snowcrab.habitat( p=p)
# ------------------------------------------
# Time-series: Habitat variations (mean temperature of snow crab habitat)
figure.timeseries.snowcrab.habitat.temperatures( p=p)
# ------------------------------------------
# Fecundity estimated indirectly via total number of females of primi and muli and applying mean egg production from allometric estimate
fecundity.indirect( p=p, outdir=file.path( p$annual.results, "timeseries", "survey" ) )
# ------------------------------------------
# Fecundity estimated directly via interpolation and individual-based fecundity estimate
figure.timeseries.fecundity( p=p, outdir=file.path( p$annual.results, "timeseries", "interpolated" ) )
# Tables for the SSR/RESDOC
# ---------------------------------------------
# Tables of CC from trawl surveys > 95mmCW
outvars = c("yr", "region", "vars", "total", "lbound", "ubound")
yy = c(1999:p$year.assessment)
rr = c("cfasouth", "cfanorth", "cfa4x")
vv = c( "totno.male.com.CC1", "totno.male.com.CC2", "totno.male.com.CC3",
"totno.male.com.CC4", "totno.male.com.CC5" )
p$runindex = list(y=yy, v=vv )
L = interpolation.db( DS="interpolation.simulation", p=p )
L$vars = as.character(L$vars)
for (y in yy) {
for (r in rr) {
i = which( L$yr==y & L$region==r )
# isum = sum(L$total[i], na.rm=T)
L[i, c("total", "lbound", "ubound")] =round( L[i, c("total", "lbound", "ubound")], 4)
}}
scale.factor = 1e2
L[, c("total", "lbound", "ubound")] = L[, c("total", "lbound", "ubound")] * scale.factor
L$vars = factor(x=L$vars, levels=vv, ordered=T)
Mn = xtabs(formula=total~yr+vars, data=L[which(L$region=="cfanorth"),], drop.unused.levels = T ) / scale.factor
dimnames(Mn)$vars = c(1:5)
Ms = xtabs(formula=total~yr+vars, data=L[which(L$region=="cfasouth"),], drop.unused.levels = T ) / scale.factor
dimnames(Ms)$vars = c(1:5)
Mx = xtabs(formula=total~yr+vars, data=L[which(L$region=="cfa4x"),], drop.unused.levels = T ) / scale.factor
dimnames(Mx)$vars = c(1:5)
(Mnp = round(Mn/rowSums(Mn)*100,2))
(Msp = round(Ms/rowSums(Ms)*100,2))
(Mxp = round(Mx/rowSums(Mx)*100,2))
latex( Mnp, file="", title="", label="table.CC.north.trawl", rowlabel="Year", cgroup="Carapace condition", na.blank=T, caption="Carapace condition of crab larger than 95 mm CW (percent by number) over time for N-ENS from trawl surveys. The transition from a spring to a fall survey occurred in 2002/2003.")
latex(Msp, file="", title="", label="table.CC.south.trawl", rowlabel="Year", cgroup="Carapace condition", na.blank=T, caption="Carapace condition of crab larger than 95 mm CW (percent by number) over time for S-ENS from trawl surveys. The transition from a spring to a fall survey occurred in 2002/2003.")
latex(Mxp, file="", title="", label="table.CC.4x.trawl", rowlabel="Year", cgroup="Carapace condition", na.blank=T, caption="Carapace condition (percent by number) over time for CFA 4X from trawl surveys.")
# ---------------------------------------------
# Tables of fishable biomass from trawl surveys
outvars = c("yr", "region", "vars", "total", "lbound", "ubound")
yy = c(1999:p$year.assessment)
rr = c("cfasouth", "cfanorth", "cfa4x")
vv = c( "totmass.male.com", "R0.mass", "R0.no", "R1.no", "R2.no", "R3.no", "totno.male.imm", "totno.male.com.CC1to2", "totno.male.com.CC3to4",
"totno.male.com.CC5" )
p$runindex = list(y=yy, v=vv )
L = interpolation.db( DS="interpolation.simulation", p=p )
# L = K[K$yr %in% yy & K$vars %in% vv & K$region %in% rr, outvars]
L$vars = as.character(L$vars)
for (y in yy) {
for (r in rr) {
i = which( L$yr==y & L$region==r )
# isum = sum(L$total[i], na.rm=T)
L[i, c("total", "lbound", "ubound")] =round( L[i, c("total", "lbound", "ubound")], 4)
}}
scale.factor = 1
L[, c("total", "lbound", "ubound")] = L[, c("total", "lbound", "ubound")] * scale.factor
L$vars = factor(x=L$vars, levels=vv, ordered=T)
Mn = xtabs(formula=total~yr+vars, data=L[which(L$region=="cfanorth"),], drop.unused.levels = T ) / scale.factor
dimnames(Mn)$vars = vv
Ms = xtabs(formula=total~yr+vars, data=L[which(L$region=="cfasouth"),], drop.unused.levels = T ) / scale.factor
dimnames(Ms)$vars = vv
Mx = xtabs(formula=total~yr+vars, data=L[which(L$region=="cfa4x"),], drop.unused.levels = T ) / scale.factor
dimnames(Mx)$vars = vv
options(width=240)
options(digits=1)
Mn
Ms
Mx
latex(Mn, file="", title="", label="table.stats.north.trawl", rowlabel="Year", cgroup="Various statistics", na.blank=T, caption="Various statistics of crab for N-ENS from trawl surveys. ")
latex(Ms, file="", title="", label="table.stats.south.trawl", rowlabel="Year", cgroup="Various statistics", na.blank=T, caption="Various statistics for S-ENS from trawl surveys. ")
latex(Mx, file="", title="", label="table.stats.4x.trawl", rowlabel="Year", cgroup="Various statistics", na.blank=T, caption="Various statistics for CFA 4X from trawl surveys.")
jae0/snowcrab documentation built on Nov. 6, 2024, 10:13 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.
year.assessment = 2019
p = bio.snowcrab::load.environment( year.assessment=year.assessment )
# 11 hrs with these settings
ncpus = parallel::detectCores()
stmv_clusters = list(
scale=rep("localhost", ncpus),
interpolate=rep("localhost", ncpus)
)
p = snowcrab_stmv(
p=p,
project_class="stmv",
stmv_variables=list(Y="snowcrab.large.males_abundance"),
selection=list(
type = "biomass",
biologicals=list(
sex=0, # male
mat=1, # do not use maturity status in groundfish data as it is suspect ..
spec_bio=bio.taxonomy::taxonomy.recode( from="spec", to="parsimonious", tolookup=2526 ),
len= c( 95, 200 )/10, # mm -> cm ; aegis_db in cm
ranged_data="len"
),
survey=list(
data.source = c("snowcrab"),
yr = p$yrs # time frame for comparison specified above
)
),
DATA = 'snowcrab_stmv( p=p, DS="stmv_inputs" )',
stmv_global_modelengine ="gam",
stmv_global_family = gaussian(link="log"),
stmv_local_modelengine = "twostep",
stmv_twostep_time = "gam",
stmv_twostep_space = "fft",
stmv_fft_filter="matern", # matern, krige (very slow), lowpass, lowpass_matern
stmv_gam_optimizer = c("outer", "bfgs") ,
stmv_distance_statsgrid = 3, # resolution (km) of data aggregation (i.e. generation of the ** statistics ** ),
stmv_distance_scale = c( 25, 35, 45 ), #likely must be over 30km, so 50 +/- 20km, should likely match the setting in ~ line 256
stmv_clusters = stmv_clusters
)
if (0) {
p$stmv_global_modelformula = formula( paste(
' snowcrab.large.males_abundance',
' ~ s( t, k=3, bs="ts") + s( tsd, k=3, bs="ts") + s( degreedays, k=3, bs="ts") ',
' + s( log(z), k=3, bs="ts") + s( log(dZ), k=3, bs="ts") + s( log(ddZ), k=3, bs="ts") ',
' + s( substrate.grainsize, k=3, bs="ts") ',
' + s(pca1, bs="ts") + s(pca2, bs="ts") '
))
p$stmv_global_modelformula = formula( paste(
' snowcrab.large.males_abundance',
' ~ s( t, k = 4, bs="ts") + s( tmax, k = 4, bs="ts") + s( degreedays, k = 4, bs="ts") ',
' + s( log(z), k=4, bs="ts") + s( log(dZ), k=4, bs="ts") + s( log(ddZ), k=4, bs="ts") ',
' + s( substrate.grainsize, k=4, bs="ts") + s(pca1, k=4, bs="ts") + s(pca2, k=4, bs="ts") '
))
p = stmv_variablelist(p=p) # decompose into covariates, etc
o = snowcrab_stmv(p=p, DS="input_data" ) # create fields for
global_model = gam(
formula=p$stmv_global_modelformula,
family=p$stmv_global_family,
data = o,
weights=o$wt,
optimizer= p$stmv_gam_optimizer,
na.action="na.omit"
)
}
#Run the following line if you want to use maptools rather than GADMTools for mapping coastline
# p$DATA = 'snowcrab_stmv( p=p, DS="stmv_inputs", coastline_source="mapdata.coastPolygon" )'
# range( INP$snowcrab.large.males_abundance )
# [1] 14.3 6675.0
# o = snowcrab_stmv(p=p, DS="stmv_inputs" ) # create fields for
#--------------------------------------------------------------
# Run the process
#--------------------------------------------------------------
stmv( p=p, runmode=c("globalmodel", "interpolate" ) ) # for a clean start
snowcrab_stmv( p=p, DS="predictions.redo" ) # warp predictions to other grids (if any)
snowcrab_stmv( p=p, DS="stmv.stats.redo" ) # warp stats to other grids (if any)
snowcrab_stmv( p=p, DS="complete.redo" )
snowcrab_stmv( p=p, DS="baseline.redo" )
snowcrab_stmv( p=p, DS="map.all" )
global_model = stmv_global_model( p=p, DS="global_model")
summary( global_model )
par(mar=c(1,1,1,1)) #change plot margins for Rstudio
plot(global_model)
# 2018 results
# Family: gaussian
# Link function: log
#
# Formula:
# snowcrab.large.males_abundance ~ s(t, k=3, bs="ts") + s(tsd,
# k=3, bs="ts") + s(tmax, k=3, bs="ts") + s(degreedays,
# k=3, bs="ts") + s(log(z), k=3, bs="ts") + s(log(dZ),
# k=3, bs="ts") + s(log(ddZ), k=3, bs="ts") + s(substrate.grainsize,
# k=3, bs="ts") + s(pca1, k=3, bs="ts") + s(pca2, k=3,
# bs="ts")
#
# Parametric coefficients:
# Estimate Std. Error t value Pr(>|t|)
# (Intercept) 6.7468 0.0204 330 <2e-16
#
# Approximate significance of smooth terms:
# edf Ref.df F p-value
# s(t) 1.91 2 73.0 < 2e-16
# s(tsd) 2.00 2 14.0 8.3e-07
# s(tmax) 1.91 2 11.7 1.4e-06
# s(degreedays) 1.36 2 45.4 < 2e-16
# s(log(z)) 1.91 2 174.7 < 2e-16
# s(log(dZ)) 1.93 2 20.9 3.8e-10
# s(log(ddZ)) 1.95 2 61.5 < 2e-16
# s(substrate.grainsize) 1.99 2 51.0 < 2e-16
# s(pca1) 2.00 2 100.4 < 2e-16
# s(pca2) 1.94 2 95.3 < 2e-16
#
# R-sq.(adj) = 0.232 Deviance explained = 23.4%
# GCV = 6198.2 Scale est. = 6182 n = 7640
# -------------------------------------------------------------------------------------
# STEP TWO commercial presence /absence
#--------------------------------------------------------------------------------------
# presence-absence
# this takes about 40 hrs ... and 5-6 GB /process
# year.assessment = 2018
p = bio.snowcrab::load.environment( year.assessment=year.assessment )
ncpus = parallel::detectCores()
p = bio.snowcrab::snowcrab_parameters(
p=p,
project_class="stmv",
stmv_variables=list(Y="snowcrab.large.males_presence_absence"),
selection=list(
type = "presence_absence",
biologicals = list(
sex=0, # male
mat=1, # do not use maturity status in groundfish data as it is suspect ..
spec_bio=bio.taxonomy::taxonomy.recode( from="spec", to="parsimonious", tolookup=2526 ),
len= c( 95, 200 )/10, # mm -> cm ; aegis_db in cm
ranged_data="len"
),
survey=list(
data.source = c("snowcrab", "groundfish"), # add groundfish data too
drop.unreliable.zeros.groundfish.data=TRUE, # esp from 1970 to 1999 measurement of invertebrates was sporatic .. zero-values are dropped as they are unreliable
yr = p$yrs # time frame for comparison specified above
)
),
DATA = 'snowcrab_stmv( p=p, DS="stmv_inputs" )',
aegis_project_datasources = c("speciescomposition" ), # c("speciescomposition", "speciesarea", "sizespectrum", "condition", "metabolism", "biochem")
stmv_global_family = binomial( link="logit" ),
stmv_global_modelengine ="gam",
stmv_local_modelengine = "twostep",
stmv_twostep_time = "gam",
stmv_twostep_space = "fft",
stmv_fft_filter="matern", # matern, krige (very slow), lowpass, lowpass_matern
stmv_gam_optimizer=c("outer", "bfgs") ,
stmv_distance_statsgrid = 3, # resolution (km) of data aggregation (i.e. generation of the ** statistics ** ),
stmv_distance_scale = c( 25, 35, 45 ), #likely must be over 30km, so 50 +/- 20km, should likely match the setting in ~ line 256
stmv_clusters = list( scale=rep("localhost", ncpus), interpolate=rep("localhost", ncpus) )
)
if (0) {
p$stmv_global_modelformula = formula( paste(
' snowcrab.large.males_abundance',
' ~ s( t, k=3, bs="ts") + s( tsd, k=3, bs="ts") + s( degreedays, k=3, bs="ts") ',
' + s( t, tsd, degreedays, bs="ts") ',
' + s( log(z), k=3, bs="ts") + s( log(dZ), k=3, bs="ts") + s( log(ddZ), k=3, bs="ts") ',
' + s( log(z), log(dZ), log(ddZ), bs="ts") ',
' + s( substrate.grainsize, k=3, bs="ts") ',
' + s(pca1, k=3, bs="ts") + s(pca2, k=3, bs="ts") + s(pca1, pca2, k=8, bs="ts") '
))
p$stmv_global_modelformula = formula( paste(
' snowcrab.large.males_abundance',
' ~ s( t, k = 4, bs="ts") + s( tmax, k = 4, bs="ts") + s( degreedays, k = 4, bs="ts") ',
' + s( log(z), k=4, bs="ts") + s( log(dZ), k=4, bs="ts") + s( log(ddZ), k=4, bs="ts") ',
' + s( substrate.grainsize, k=4, bs="ts") + s(pca1, k=4, bs="ts") + s(pca2, k=4, bs="ts") '
))
o = snowcrab_stmv(p=p, DS="input_data" ) # create fields for
global_model = gam(
formula=p$stmv_global_modelformula,
family=p$stmv_global_family,
data = o,
weights=o$wt,
optimizer= p$stmv_gam_optimizer,
na.action="na.omit"
)
}
# p$DATA = 'snowcrab_stmv( p=p, DS="stmv_inputs", coastline_source="mapdata.coastPolygon" )'
# o = snowcrab_stmv(p=p, DS="stmv_inputs" ) # create fields for
stmv( p=p, runmode=c("globalmodel", "interpolate" ) ) # no global_model and force a clean restart
snowcrab_stmv( p=p, DS="predictions.redo" ) # warp predictions to other grids
snowcrab_stmv( p=p, DS="stmv.stats.redo" ) # warp stats to other grids
snowcrab_stmv( p=p, DS="complete.redo" )
snowcrab_stmv( p=p, DS="baseline.redo" )
snowcrab_stmv( p=p, DS="map.all" )
global_model = stmv_global_model( p=p, DS="global_model")
summary( global_model )
par(mar=c(1,1,1,1)) #change plot margins for Rstudio
plot(global_model, all.terms=TRUE, trans=bio.snowcrab::inverse.logit, seWithMean=TRUE, jit=TRUE, rug=TRUE )
# Family: binomial
# Link function: logit
#
# Formula:
# snowcrab.large.males_presence_absence ~ s(t, k=3, bs="ts") +
# s(tsd, k=3, bs="ts") + s(tmax, k=3, bs="ts") + s(degreedays,
# k=3, bs="ts") + s(log(z), k=3, bs="ts") + s(log(dZ),
# k=3, bs="ts") + s(log(ddZ), k=3, bs="ts") + s(substrate.grainsize,
# k=3, bs="ts") + s(pca1, k=3, bs="ts") + s(pca2, k=3,
# bs="ts")
#
# Parametric coefficients:
# Estimate Std. Error z value Pr(>|z|)
# (Intercept) 0.0302 0.0349 0.87 0.39
#
# Approximate significance of smooth terms:
# edf Ref.df Chi.sq p-value
# s(t) 1.99 2 336.7 < 2e-16
# s(tsd) 2.00 2 35.3 2.0e-08
# s(tmax) 1.98 2 59.9 7.0e-14
# s(degreedays) 1.99 2 188.4 < 2e-16
# s(log(z)) 2.00 2 1282.7 < 2e-16
# s(log(dZ)) 1.99 2 68.7 9.1e-16
# s(log(ddZ)) 1.98 2 224.4 < 2e-16
# s(substrate.grainsize) 1.98 2 394.4 < 2e-16
# s(pca1) 2.00 2 932.6 < 2e-16
# s(pca2) 2.00 2 1291.8 < 2e-16
#
# R-sq.(adj) = 0.62 Deviance explained = 56.2%
# UBRE = -0.58807 Scale est. = 1 n = 36031
#
# collect all predictions into a single file and return:
# year.assessment=2017
#------------------------------------------------------------------------------
# STEP THREE- Predict biomass by weighting +/- with probabilities
#------------------------------------------------------------------------------
p = bio.snowcrab::load.environment( year.assessment=year.assessment )
p = snowcrab_stmv(
p=p,
project_class="stmv",
stmv_variables = list(Y="snowcrab.large.males_abundance"),
selection=list(
type = "biomass",
biologicals=list(
sex=0, # male
mat=1, # do not use maturity status in groundfish data as it is suspect ..
spec_bio=bio.taxonomy::taxonomy.recode( from="spec", to="parsimonious", tolookup=2526 ),
len= c( 95, 200 )/10, # mm -> cm ; aegis_db in cm
ranged_data="len"
),
survey=list(
data.source = c("snowcrab"),
yr = p$yrs, # time frame for comparison specified above
drop.unreliable.zeros.groundfish.data=TRUE # esp from 1970 to 1999 measurement of invertebrates was sporatic .. zero-values are dropped as they are unreliable
)
)
)
interpolation.db( DS="fishable.biomass.redo", p=p ) # combine habitat and abundance info and map
K = interpolation.db( DS="fishable.biomass.timeseries", p=p )
if (0){
str(K)
table.view( K )
plot( total ~ yr, K[K$region=="cfanorth", ], type="b")
plot( total ~ yr, K[K$region=="cfasouth", ], type="b")
plot( total ~ yr, K[K$region=="cfa4x", ], type="b")
}
interpolation.db( DS="fishable.biomass.map", p=p )
figure.timeseries.snowcrab.habitat(p=p) # /bio.data/bio.snowcrab/assessments/2016/timeseries/interpolated/snowcrab.habitat.sa.png
figure.timeseries.snowcrab.habitat.temperatures(p=p) # /bio.data/bio.snowcrab/assessments/2016/timeseries/interpolated/mean.bottom.temp.snowcrab.habitat.png
### --------- prediction success:
set = snowcrab_stmv(p=p, DS="input_data" )
S = set[ , c("plon", "plat") ]
ii = array_map( "xy->1", S, gridparams=p$gridparams )
bs = bathymetry_db(p=p, DS="baseline")
bb = array_map( "xy->1", bs, gridparams=p$gridparams )
im = match( ii, bb )
it = match( set$yr, p$yrs )
bm = interpolation.db( DS="fishable.biomass", p=p )
spred = bm$m[cbind(im, it)] # approximate match (ignoring seasonality)
summary ( lm(log(spred)~log(snowcrab.large.males_abundance), data=set, na.actio="na.omit" ) )
plot(log(spred)~log(snowcrab.large.males_abundance), data=set )
cor(log(spred),log(set$snowcrab.large.males_abundance), use="complete.obs")
# Call:
# lm(formula = log(spred) ~ log(snowcrab.large.males_abundance),
# data = set, na.action = "na.omit")
#
# Residuals:
# Min 1Q Median 3Q Max
# -2.3039 -0.5507 0.0589 0.6056 2.1379
#
# Coefficients:
# Estimate Std. Error t value Pr(>|t|)
# (Intercept) 5.17806 0.03824 135.4 <2e-16
# log(snowcrab.large.males_abundance) 0.19035 0.00599 31.8 <2e-16
#
# Residual standard error: 0.795 on 5030 degrees of freedom
# (2223 observations deleted due to missingness)
# Multiple R-squared: 0.167, Adjusted R-squared: 0.167
# F-statistic: 1.01e+03 on 1 and 5030 DF, p-value: <2e-16
#
# R> plot(log(spred)~log(snowcrab.large.males_abundance), data=set )
# R> cor(log(spred),log(set$snowcrab.large.males_abundance), use="complete.obs")
# [1] 0.4269
# determine presence absence(Y) and weighting(wt)
# set$weekno = floor(set$julian / 365 * 52) + 1
# set$dyear = floor(set$julian / 365 ) + 1
## END
# figures and tables related to fishery indices
#Pick whichever year reference below is correct (most often year.assessment...-1)
if (!exists("year.assessment")) {
year.assessment=lubridate::year(Sys.Date())
year.assessment=lubridate::year(Sys.Date()) - 1
}
p = bio.snowcrab::load.environment( year.assessment=year.assessment )
# ------------------------------------------
# Time-series: all interpolated data estimated from interpolated analysis
# BZ R0.mass is likely the only variable required
figure.interpolated.results( p, outdir=file.path( p$annual.results, "timeseries", "interpolated" ), alt.zero.y=T )
# ------------------------------------------
# Time-series: immature male numerical abundance estimates from interpolated analysis (broken down by instar)
figure.immature.male( p, outdir=file.path( p$annual.results, "timeseries", "interpolated" ) )
# ------------------------------------------
# Time-series: immature male skip-moulter numerical abundance estimates from interpolated analysis (broken down by instar)
figure.immature.skipmoulter.male( p, outdir=file.path( p$annual.results, "timeseries", "interpolated" ) )
# ------------------------------------------
# Time-series: mature CC12 male numerical abundance estimates from interpolated analysis (broken down by instar)
figure.mature.CC12.male( p, outdir=file.path( p$annual.results, "timeseries", "interpolated" ) )
# ------------------------------------------
# Time-series: mature CC34 male numerical abundance estimates from interpolated analysis (broken down by instar)
figure.mature.CC34.male( p, outdir=file.path( p$annual.results, "timeseries", "interpolated" ) )
# ------------------------------------------
# Time-series: mature CC5 male numerical abundance estimates from interpolated analysis (broken down by instar)
figure.mature.CC5.male( p, outdir=file.path( p$annual.results, "timeseries", "interpolated" ) )
# ------------------------------------------
# Time-series: Exploitation rate
figure.exploitation.rate( p, outdir=file.path( p$annual.results, "timeseries", "interpolated" ) )
figure.exploitation.rate( p, outdir=file.path( p$annual.results, "timeseries", "interpolated" ), CFA4X.exclude.year.assessment=F )
# ------------------------------------------
# Time-series: Generic plots of all interpolated data
# figure.timeseries.interpolated( p, outdir=file.path( p$annual.results, "timeseries", "interpolated" ) )
# ------------------------------------------
# Time-series: Habitat variations (surface area of snow crab habitat)
# p$model.type = "gam.full" # choose method for habitat model :
# p$model.type = "gam.simple" # choose method for habitat model : no longer used
p$habitat.threshold.quantile = 0.05 # quantile at which to consider zero-valued abundance
p$prediction_dyear = 9/12 # predict for ~ Sept 1
figure.timeseries.snowcrab.habitat( p=p)
# ------------------------------------------
# Time-series: Habitat variations (mean temperature of snow crab habitat)
figure.timeseries.snowcrab.habitat.temperatures( p=p)
# ------------------------------------------
# Fecundity estimated indirectly via total number of females of primi and muli and applying mean egg production from allometric estimate
fecundity.indirect( p=p, outdir=file.path( p$annual.results, "timeseries", "survey" ) )
# ------------------------------------------
# Fecundity estimated directly via interpolation and individual-based fecundity estimate
figure.timeseries.fecundity( p=p, outdir=file.path( p$annual.results, "timeseries", "interpolated" ) )
# Tables for the SSR/RESDOC
# ---------------------------------------------
# Tables of CC from trawl surveys > 95mmCW
outvars = c("yr", "region", "vars", "total", "lbound", "ubound")
yy = c(1999:p$year.assessment)
rr = c("cfasouth", "cfanorth", "cfa4x")
vv = c( "totno.male.com.CC1", "totno.male.com.CC2", "totno.male.com.CC3",
"totno.male.com.CC4", "totno.male.com.CC5" )
p$runindex = list(y=yy, v=vv )
L = interpolation.db( DS="interpolation.simulation", p=p )
L$vars = as.character(L$vars)
for (y in yy) {
for (r in rr) {
i = which( L$yr==y & L$region==r )
# isum = sum(L$total[i], na.rm=T)
L[i, c("total", "lbound", "ubound")] =round( L[i, c("total", "lbound", "ubound")], 4)
}}
scale.factor = 1e2
L[, c("total", "lbound", "ubound")] = L[, c("total", "lbound", "ubound")] * scale.factor
L$vars = factor(x=L$vars, levels=vv, ordered=T)
Mn = xtabs(formula=total~yr+vars, data=L[which(L$region=="cfanorth"),], drop.unused.levels = T ) / scale.factor
dimnames(Mn)$vars = c(1:5)
Ms = xtabs(formula=total~yr+vars, data=L[which(L$region=="cfasouth"),], drop.unused.levels = T ) / scale.factor
dimnames(Ms)$vars = c(1:5)
Mx = xtabs(formula=total~yr+vars, data=L[which(L$region=="cfa4x"),], drop.unused.levels = T ) / scale.factor
dimnames(Mx)$vars = c(1:5)
(Mnp = round(Mn/rowSums(Mn)*100,2))
(Msp = round(Ms/rowSums(Ms)*100,2))
(Mxp = round(Mx/rowSums(Mx)*100,2))
latex( Mnp, file="", title="", label="table.CC.north.trawl", rowlabel="Year", cgroup="Carapace condition", na.blank=T, caption="Carapace condition of crab larger than 95 mm CW (percent by number) over time for N-ENS from trawl surveys. The transition from a spring to a fall survey occurred in 2002/2003.")
latex(Msp, file="", title="", label="table.CC.south.trawl", rowlabel="Year", cgroup="Carapace condition", na.blank=T, caption="Carapace condition of crab larger than 95 mm CW (percent by number) over time for S-ENS from trawl surveys. The transition from a spring to a fall survey occurred in 2002/2003.")
latex(Mxp, file="", title="", label="table.CC.4x.trawl", rowlabel="Year", cgroup="Carapace condition", na.blank=T, caption="Carapace condition (percent by number) over time for CFA 4X from trawl surveys.")
# ---------------------------------------------
# Tables of fishable biomass from trawl surveys
outvars = c("yr", "region", "vars", "total", "lbound", "ubound")
yy = c(1999:p$year.assessment)
rr = c("cfasouth", "cfanorth", "cfa4x")
vv = c( "totmass.male.com", "R0.mass", "R0.no", "R1.no", "R2.no", "R3.no", "totno.male.imm", "totno.male.com.CC1to2", "totno.male.com.CC3to4",
"totno.male.com.CC5" )
p$runindex = list(y=yy, v=vv )
L = interpolation.db( DS="interpolation.simulation", p=p )
# L = K[K$yr %in% yy & K$vars %in% vv & K$region %in% rr, outvars]
L$vars = as.character(L$vars)
for (y in yy) {
for (r in rr) {
i = which( L$yr==y & L$region==r )
# isum = sum(L$total[i], na.rm=T)
L[i, c("total", "lbound", "ubound")] =round( L[i, c("total", "lbound", "ubound")], 4)
}}
scale.factor = 1
L[, c("total", "lbound", "ubound")] = L[, c("total", "lbound", "ubound")] * scale.factor
L$vars = factor(x=L$vars, levels=vv, ordered=T)
Mn = xtabs(formula=total~yr+vars, data=L[which(L$region=="cfanorth"),], drop.unused.levels = T ) / scale.factor
dimnames(Mn)$vars = vv
Ms = xtabs(formula=total~yr+vars, data=L[which(L$region=="cfasouth"),], drop.unused.levels = T ) / scale.factor
dimnames(Ms)$vars = vv
Mx = xtabs(formula=total~yr+vars, data=L[which(L$region=="cfa4x"),], drop.unused.levels = T ) / scale.factor
dimnames(Mx)$vars = vv
options(width=240)
options(digits=1)
Mn
Ms
Mx
latex(Mn, file="", title="", label="table.stats.north.trawl", rowlabel="Year", cgroup="Various statistics", na.blank=T, caption="Various statistics of crab for N-ENS from trawl surveys. ")
latex(Ms, file="", title="", label="table.stats.south.trawl", rowlabel="Year", cgroup="Various statistics", na.blank=T, caption="Various statistics for S-ENS from trawl surveys. ")
latex(Mx, file="", title="", label="table.stats.4x.trawl", rowlabel="Year", cgroup="Various statistics", na.blank=T, caption="Various statistics for CFA 4X from trawl surveys.")
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.