-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathLCM.sim.CC.2020.r
271 lines (222 loc) · 15.8 KB
/
LCM.sim.CC.2020.r
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
####################################################
#This is the wrapper for running the LCM model for
# Crozier, Burke, Chasco, Widener and Zabel 2020
# Iconic salmon populations face perilous challenges from climate change across their life cycle
# published in Communications Biology
####################################################
####################################################
set.seed(10000)
library(plyr)
# Model control parameters---------
#1.Select which model you want to run. Models are defined by which freshwater and marine covariates they include
#e.g., Model 1:
# FW: p2TsuFfall
# In-river: ersstArc.win + ersstWAcoast.sum
# Transport: ersstArc.win
model<-"Model1";filename<-"p2TsuFfall"; Fname="F.S3.ParrF"; FW.name="Ffall"
# model<-"Model2"; filename<-"p2Fsu"; Fname="F.S2.ParrSu"; FW.name="Fsu"
# model<-"Model3"; filename<-"p2TsuFfall"; Fname="F.S3.ParrF"; FW.name="Ffall"
arrayname<-paste0(model,".array")
load(file=paste("Input files/param",arrayname,sep="."),verbose=T) #param.array.1K
#2. Decide how many simulations,populations and climate scenarios you want to run and build list of scenarios
runs = 10
years = 75 #2015:2089
pop<-creeks<-"bearvalley"
# pop<-creeks<-c("bearvalley","big","camas","loon","marsh","secesh","sulphur","valley")
climate =c("stationary","4.5" , "8.5" )
climateshort =c("1","45" , "85" )
quant = c("25", "50", "75")
nsim<-length(pop)*length(model)*length(climate)*length(quant)#;nsim
scenario<-length(model)*length(climate)*length(quant);scenario
scenario.name<-vector("character",length=scenario)
cnt.scenario=0
for(l in 1:length(model)){
for(cc in 1:length(climate)){
for(qq in 1:length(quant)){
cnt.scenario=cnt.scenario+1
scenario.name[cnt.scenario]=paste(climate[cc],quant[qq],model[l],sep=".");scenario.name
}}}
#Input files-------------------
#FW
source("LCM.fxn.r") #functions for model relationships
popinit<-read.csv("Input files/Ninit.csv", header = T, row.names = 1)
hydro<-read.csv(file="Input files/climategrid_survival_80yr_2020_PA.csv",header=TRUE,row.names = NULL);hydro$gridYear<-rep(1:80,length=nrow(hydro))
FW<-read.csv(file="Input files/SALM.Qflow.csv",header=TRUE,row.names = NULL)#;head(FW)
FW<-FW[FW$year>=2015,]
scalefactors<-read.csv(file="Input files/scalefactors.corrected.May2019.csv",row.names=1)#;scalefactors
su<-read.csv("Input files/AprilJuneAndSurv_LoopsTransport.csv",row.names = NULL,header=TRUE)#;head(su)
su$LGRtemp<-round_any(su$LGrTempAJ,0.5)#;simtemp
su$LGRflow<-round_any(su$LGrFlowAJ,20)#;simflow
#SARs
if(model=="Model3") sarfile<-"Model3.SAR100.Rdata" else sarfile<-"Model1.Model2.SAR100.Rdata"
print(sarfile)
load(paste0("Input files/",sarfile),verbose=TRUE)
#collect parameters used in each run
param.marT<-as.data.frame(t(sarcoefT_rqvs["stationary","50",,]))
names(param.marT)<-paste("T",names(param.marT),sep=".")
param.marI<-as.data.frame(t(sarcoefI_rqvs["stationary","50",,]))
names(param.marI)<-paste("I",names(param.marI),sep=".")
# Set up temporary arrays to store data during model runs-----------
N.new = N = array(0,5)
surv.s1=surv.trib=surv.inriver=pt=pt.sar=pt.sup=surv.main=surv.s2=surv.s3=surv.sar=surv.sarI=surv.sarT=surv.sup=sup.spr.inriver=sup.spr.transport=sup.sum.inriver=sup.sum.transport=sup.spring=sup.summer=Spawners = Recruits = array(0,years) #indiv years within a run
# Arrays to save output-----------
spawner.array<-parrsmolt.array<-recruit.array<-array(0,dim=c(years,runs,length(pop),scenario),dimnames=list(2015:2089,paste0("sim",1:runs),pop,scenario.name))#;dim(spawner.array)
response<-c("yrQET50","Mean2020","Mean2040","Mean2060","Mean2080","Meanallyears")
allparam<-c( names(param.marI),names(param.marT),dimnames(param.array.1K)[[2]],response)# ;allparam
param.array<-array(0,dim=c(runs,length(allparam),length(pop),scenario),dimnames=list(paste0("sim",1:runs),allparam,pop,scenario.name))#;dim(param.array)
#START LOOPING THROUGH SIMULATIONS ========================
print(Sys.time())
#k=1;l=1;m=1;cc=1;p=1;j=1;qq=1
cnt=0
for(k in 1:length(pop)){ # loop across populations
cnt.scenario=0
print(paste("Pop",k,pop[k],Sys.time()))
ha= popinit[pop[k],"ha"]
for(cc in 1:length(climate)){
for(qq in 1:length(quant)){ #loop across quantiles of each climate scenario
cnt.scenario=cnt.scenario+1;print(scenario.name[cnt.scenario])
for (j in 1:runs){
jj<-sample(length(param.array.1K[,1,1]),1)
jj.sar<-sample(length(sarI_rqys[1,1,1,]),1)
param<-as.list(param.array.1K[jj,,pop[k]])#;unlist(param)
#collect parameters used in each run
param.marT<-as.data.frame(t(sarcoefT_rqvs[climate[cc],quant[qq],,jj.sar])) #;unlist(param.marT) #dimnames(sarcoefT_rqvs)
param.marI<-as.data.frame(t(sarcoefI_rqvs[climate[cc],quant[qq],,jj.sar])) #dimnames(sarBeta_frqvs)[[1]][2]
param.array[j,paste("T",names(param.marT),sep="."),k,scenario.name[cnt.scenario]]<-unlist(param.marT)
param.array[j,paste("I",names(param.marI),sep="."),k,scenario.name[cnt.scenario]]<-unlist(param.marI)
param.array[j,names(param),k,scenario.name[cnt.scenario]]<-unlist(param)
#Load raw T and F projections from TMB model
T1<- sc_envProj_rqvys[climate[cc],quant[qq],"T.S2.ParrSu",,jj.sar]#*env_sc["T.S2.ParrSu"]+env_mu["T.S2.ParrSu"]
F1<- sc_envProj_rqvys[climate[cc],quant[qq],Fname,,jj.sar]#*env_sc[Fname]+env_mu[Fname]
#Add trends
if(cc>1) T1<-T1+FW[,paste0("Tsu.Q",quant[qq],".RCP",climate[cc])]
if(cc>1) F1<-F1+FW[,paste0(FW.name,quant[qq],".RCP",climateshort[cc])]
#Scale for gompertz model coefficients
T1<-(T1-scalefactors["mean","T.S2.ParrSu"])/scalefactors["sd","T.S2.ParrSu"]
F1<-(F1-scalefactors["mean",Fname])/scalefactors["sd",Fname]
#Initialize all age classes
N[1:5] = c(0,0,0,0,0)
for (i in 1:5){
xx<-sample(1:dim(envProj_rqvys)[5],1)
N.new[1] = popinit[pop[k],"stationaryMeanSp"]*exp(param$p1+param$c1*log(popinit[pop[k],"stationaryMeanSp"]/ha))
if(filename=="p2Fsu") surv.trib[i] = exp(param$p2+param$c2*log(max(N[1]/ha,1))+param$BF*F1[i])
if(length(grep("p2Tsu",filename))>0) {surv.trib[i] = exp(param$p2+param$c2*log(max(N[1]/ha,1))+param$BT*T1[i]+param$BF*F1[i])}
if(length(grep("c2",filename))>0) {surv.trib[i] = exp(param$p2+param$c2*log(max(N[1]/ha,1))+param$BT*T1[i]*log(max(N[1]/ha,1))+param$BF*F1[i]*log(max(N[1]/ha,1)))}
surv.inriver[i]= hydro[hydro$gridYear==envProj_rqvys[climate[cc],quant[qq],"gridYear",i,xx] &
hydro$flow==envProj_rqvys[climate[cc],quant[qq],"gridFlow",i,xx]
& hydro$temp==envProj_rqvys[climate[cc],quant[qq],"gridTemp",i,xx],"surv"]
pt[i]=hydro[hydro$gridYear==envProj_rqvys[climate[cc],quant[qq],"gridYear",i,xx] &
hydro$flow==envProj_rqvys[climate[cc],quant[qq],"gridFlow",i,xx]
& hydro$temp==envProj_rqvys[climate[cc],quant[qq],"gridTemp",i,xx],"proptrans"]
surv.main[i] =S2.mainstem(pt=pt[i], s.inriver=surv.inriver[i])
pt.sar[i]<-pt[i]*0.98/((1-pt[i])*surv.inriver[i]+pt[i]*0.98)
N.new[2] = N[1]*surv.trib[i]*surv.main[i]
N.new[3] = N[2]*s3.fxn(pt=pt.sar[i],sarI=plogis(sarI_rqys[climate[cc],quant[qq],i,xx]),sarT=plogis(sarT_rqys[climate[cc],quant[qq],i,xx]),b3=param$b3,b4=param$b4,So=param$s0)$s3
N.new[4] = N[3]*(1-param$b3)*param$s0
N.new[5] = N[4]*(1-param$b4)*param$s0
N = N.new
};N
for (i in 1:years){
# Calculate effective spawners for recruit calculations
#In the first year, we do not have pt.sup yet, so we are just using the average survival of upstream migrants
if(i==1){
myvar="aprmayjunetempLGR";lgrtemp<-sc_envProj_rqvys[climate[cc],quant[qq],myvar,i,jj.sar]#*env_sc[myvar]+env_mu[myvar]
myvar="aprmayjuneflowLGR";lgrflow<-sc_envProj_rqvys[climate[cc],quant[qq],myvar,i,jj.sar]#*env_sc[myvar]+env_mu[myvar]
simtemp<-round_any(lgrtemp,0.5)#;simtemp
simflow<-round_any(lgrflow,20)#;simflow
su.Temp.matches<-su[which(abs(simtemp-(su$LGRtemp))==min(abs(simtemp-(su$LGRtemp)))),]#;su.Temp.matches
su.Flow.matches<-su.Temp.matches[which(abs(simflow-(su.Temp.matches$LGRflow))==min(abs(simflow-(su.Temp.matches$LGRflow)))),]#;su.Flow.matches
surv.sup[i]<-sample(su.Flow.matches$SpringSurv_BoToLG,1)
if(k %in% c(6,8)) surv.sup[i]<-sample(su.Flow.matches$SummerSurv_BoToLG,1)
spawners = (param$b4*N[4] + param$f5*N[5])*surv.sup[i]*0.9
}
#In subsequent years, we use output from SARs in the previous year to determine the precent of upstream migrants that were transported as juveniles
if(i>1){ spawners = (param$b4*N[4] + param$f5*N[5])*surv.sup[i-1]*0.9 }
#Survival in each age class
surv.s1[i]=exp(param$p1+param$c1*log(spawners/ha))
if(filename=="p2Fsu") surv.trib[i] = exp(param$p2+param$c2*log(max(N[1]/ha,1))+param$BF*F1[i])
if(length(grep("p2Tsu",filename))>0) {surv.trib[i] = exp(param$p2+param$c2*log(max(N[1]/ha,1))+param$BT*T1[i]+param$BF*F1[i])}
if(length(grep("c2",filename))>0) {surv.trib[i] = exp(param$p2+param$c2*log(max(N[1]/ha,1))+param$BT*T1[i]*log(max(N[1]/ha,1))+param$BF*F1[i]*log(max(N[1]/ha,1)))}
surv.inriver[i]= hydro[hydro$gridYear==envProj_rqvys[climate[cc],quant[qq],"gridYear",i,jj.sar] &
hydro$flow==envProj_rqvys[climate[cc],quant[qq],"gridFlow",i,jj.sar] &
hydro$temp==envProj_rqvys[climate[cc],quant[qq],"gridTemp",i,jj.sar],"surv"]
pt[i]= hydro[hydro$gridYear==envProj_rqvys[climate[cc],quant[qq],"gridYear",i,jj.sar] &
hydro$flow==envProj_rqvys[climate[cc],quant[qq],"gridFlow",i,jj.sar] &
hydro$temp==envProj_rqvys[climate[cc],quant[qq],"gridTemp",i,jj.sar],"proptrans"]
surv.main[i] =S2.mainstem(pt=pt[i], s.inriver=surv.inriver[i])
pt.sar[i]<-pt[i]*0.98/((1-pt[i])*surv.inriver[i]+pt[i]*0.98)
surv.s2[i] =surv.trib[i]*surv.main[i]
s3 = s3.fxn(pt=pt.sar[i],sarI=plogis(sarI_rqys[climate[cc],quant[qq],i,jj.sar]),sarT=plogis(sarT_rqys[climate[cc],quant[qq],i,jj.sar]),b3=param$b3,b4=param$b4,So=param$s0)
surv.s3[i] = s3$s3
surv.sarI[i] = s3$sarI
surv.sarT[i] = s3$sarT
surv.sar[i] = s3$sar
#tracking in-river and transported fish separately for upstream migration survival
pt.sup[i]<-pt.sar[i]*surv.sarT[i]/surv.sar[i]
myvar="aprmayjunetempLGR";lgrtemp<-sc_envProj_rqvys[climate[cc],quant[qq],myvar,i,jj.sar]#*env_sc[myvar]+env_mu[myvar]
myvar="aprmayjuneflowLGR";lgrflow<-sc_envProj_rqvys[climate[cc],quant[qq],myvar,i,jj.sar]#*env_sc[myvar]+env_mu[myvar]
simtemp<-round_any(lgrtemp,0.5)#;simtemp
simflow<-round_any(lgrflow,20)#;simflow
su.Temp.matches<-su[which(abs(simtemp-(su$LGRtemp))==min(abs(simtemp-(su$LGRtemp)))),]#;su.Temp.matches
su.Flow.matches<-su.Temp.matches[which(abs(simflow-(su.Temp.matches$LGRflow))==min(abs(simflow-(su.Temp.matches$LGRflow)))),]#;su.Flow.matches
xx<-sample(1:nrow(su.Flow.matches),size=1);xx
sup.spr.inriver[i]<-su.Flow.matches$Spr_Pred_BOLG_inriver[xx]
sup.spr.transport[i]<-su.Flow.matches$Spr_Pred_BOLG_transport[xx]
sup.sum.inriver[i]<-su.Flow.matches$Sum_Pred_BOLG_inriver[xx]
sup.sum.transport[i]<-su.Flow.matches$Sum_Pred_BOLG_transport[xx]
sup.spring[i]<-pt.sup[i]* sup.spr.transport[i]+(1-pt.sup[i])* sup.spr.inriver[i]
sup.summer[i]<-pt.sup[i]* sup.sum.transport[i]+(1-pt.sup[i])* sup.sum.inriver[i]
surv.sup[i]<-sup.spring[i]
if(k %in% c(6,8)) surv.sup[i]<-sup.summer[i]
#Note that surv.sup[i] will be used in next year's calculation of spawners, not this year's N[1]
N.new[1] = spawners*surv.s1[i]
N.new[2] = N[1]*surv.s2[i]
N.new[3] = N[2]*surv.s3[i]
N.new[4] = N[3]*(1-param$b3)*param$s0
N.new[5] = N[4]*(1-param$b4)*param$s0
N = N.new
# calculate spawners
Spawners[i] = (param$b4*N[4] + N[5])*surv.sup[i]*0.9
# calculate recruits referenced to brood year
if (i > 4 && i <= (years-1))
Recruits[i-4] = param$b4*N[4]*surv.sup[i]*0.9
if (i > 5)
Recruits[i-5] = Recruits[i-5] + N[5]*surv.sup[i]*0.9
} #end years
Spawners2<-Spawners[5:years]
spawner.array[,j,k,cnt.scenario]<-Spawners
parrsmolt.array[,j,k,cnt.scenario]<-surv.trib
recruit.array[,j,k,cnt.scenario]<-Recruits
} #end runs
#Store spawner and recruit matrices and name them by scenario
save(spawner.array,file=paste("output/spawner",filename,arrayname,sep="."))
save(param.array,file=paste("output/param",filename,arrayname,sep="."))
save(parrsmolt.array,file=paste("output/parrsmolt",filename,arrayname,sep="."))
save(recruit.array,file=paste("output/recruit",filename,arrayname,sep="."))
cnt=cnt+1
} #end quantiles qq
} #end climate cc
} #end populations
#Add response metrics to param.array-------------
years<-2015:2089 #dimnames(spawner.array)[[1]]
runs<-dim(spawner.array)[[2]]
pop<-dimnames(spawner.array)[[3]]
scenarios<-dimnames(spawner.array)[[4]]
spQET<-spawner.array
spQET[1:3,,,]<-NA
for(k in 1:length(pop)){
for(ss in 1:length(scenarios)){
for ( j in 1:runs){
spawners<-spawner.array[,j,k,scenarios[ss]];spawners
for (i in 4:75){ifelse (sum(spawners[(i-3):i]) < 200,spQET[i,j,k,ss]<-1,spQET[i,j,k,ss]<-0) }
param.array[j,"yrQET50",k,scenarios[ss]]<-ifelse(max(spQET[,j,k,scenarios[ss]],na.rm=TRUE)>0.5 ,
years[min(which(spQET[,j,k,scenarios[ss]]>0.5))] , 2115)
param.array[j,"Mean2020",k,scenarios[ss]]<-exp(mean(log(spawners[6:15])))
param.array[j,"Mean2040",k,scenarios[ss]]<-exp(mean(log(spawners[26:35])))
param.array[j,"Mean2060",k,scenarios[ss]]<-exp(mean(log(spawners[46:55])))
param.array[j,"Mean2080",k,scenarios[ss]]<-exp(mean(log(spawners[66:75])))
param.array[j,"Meanallyears",k,scenarios[ss]]<-exp(mean(log(spawners)))
} #end runs
}#end scenarios
}#end pop
save(param.array,file=paste("output/param",filename,arrayname,sep="."))