-
Notifications
You must be signed in to change notification settings - Fork 6
/
LME_BiodivFacets_DetrendedStability.R
254 lines (169 loc) · 8.76 KB
/
LME_BiodivFacets_DetrendedStability.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
###################################################
# De-trended facets of biodiversity ~ stability #
###################################################
# 1. Species diversity
# 2. Species asynchrony
# 3. Phylogenetic diversity (MNTD)
# 4. Functional diversity (FDis)
# 5. Functional composition
library(lmerTest)
library(nlme)
library(dplyr)
library(AICcmodavg)
library(MuMIn)
library(ggplot2)
ICClme <- function(out) {
varests <- as.numeric(VarCorr(out)[1:2])
return(paste("ICC =", varests[1]/sum(varests)))
}
# Prepare data for analysis
stab<-read.delim("data.csv",sep=",",header=T)
stab<-filter(stab,Site!="BIODEPTH_GR") # should get rid of site where we didn't have good trait coverage
stab_4<-select(stab,Site,Study_length,UniqueID,SppN,eMPD,eMNTD,ePSE,sMPD,sMNTD,FDis4,FRic4,PCAdim1_4trts,SLA, LDMC, LeafN, LeafP,
Plot_TempStab,Plot_TempStab_Detrend, Plot_Biomassxbar, Plot_Biomasssd, Plot_Biomasssd_detrend=detrend_sd,
Gross_synchrony, mArid,sdAridity)
# for plots with ONLY 1 spp, we assume that a species #is perfectly synchronized with itself
stab_4$Gross_synchrony<-ifelse(is.na(stab_4$Gross_synchrony)==TRUE,1,stab_4$Gross_synchrony)
# convert synchrony metrics to different scale
stab_4$GrossAsynchrony_s<-stab_4$Gross_synchrony*-1
# further adjustments
stab_4$SppN<-as.numeric(stab_4$SppN)
stab_4$TS_lg2<-log(stab_4$Plot_TempStab,base=2)
stab_4$TS_detrend_lg2<-log(stab_4$Plot_TempStab_Detrend,base=2)
stab_4$lg2SppN <- log(stab_4$SppN,2)
stab_4$lg2_mArid <-log(stab_4$mArid,base=2)
# Filter out NAs for Asynchrony and FRic4
stab_444<-filter(stab_4, is.na(PlotAsynchrony_s)==FALSE)
stab_444<-filter(stab_444, is.na(FRic4)==FALSE)
#######################################################################
# Effects on ecosystem stability of individual biodiversity facets ##
#######################################################################
# Control list set up for LMM in nlme
cc<-lmeControl(opt="optim")
########################
## Species richness ####
########################
## Test multiple random effect structures; compare using AICc
Cand.set <- list( )
Cand.set[[1]]<-lme(TS_detrend_lg2~lg2SppN,random=~1+lg2SppN|Site,control=cc,data=stab_444)
Cand.set[[2]]<-lme(TS_detrend_lg2~lg2SppN,random=~1|Site,control=cc,data=stab_444)
Cand.set[[3]]<-lme(TS_detrend_lg2~lg2SppN,random=~1|Site/SppN,control=cc,data=stab_444)
Modnames <- paste("Mod", 1:length(Cand.set), sep = " ")
res.table <- aictab(cand.set = Cand.set, modnames = Modnames,second.ord = T)
res.table
plot(Cand.set[[best_model]])
qqnorm(Cand.set[[best_model]])
# Log ratio test for significance of main effect
big<-lme(TS_detrend_lg2~lg2SppN,random=~1+lg2SppN|Site,control=cc,data=stab_444,method="ML")
small<-lme(TS_detrend_lg2~1,random=~1+lg2SppN|Site,control=cc,data=stab_444,method="ML")
anova(big,small)
#final model summary
final<-lme(TS_detrend_lg2~lg2SppN,random=~1+lg2SppN|Site,control=cc,data=stab_444)
r.squaredGLMM(final)
intervals(final, which="fixed")
VarCorr(final)
ICClme(final)
#######################
# Species Asynchrony #
#######################
## Test multiple random effect structures; compare using AICc
Cand.set <- list( )
Cand.set[[1]]<-lme(TS_detrend_lg2~GrossAsynchrony_s,random=~1+GrossAsynchrony_s|Site,control=cc,data=stab_444)
Cand.set[[2]]<-lme(TS_detrend_lg2~GrossAsynchrony_s,random=~1+GrossAsynchrony_s|Site/SppN,control=cc,data=stab_444)
Cand.set[[3]]<-lme(TS_detrend_lg2~GrossAsynchrony_s,random=~1|Site,control=cc,data=stab_444)
Cand.set[[4]]<-lme(TS_detrend_lg2~GrossAsynchrony_s,random=~1|Site/SppN,control=cc,data=stab_444)
Cand.set[[5]]<-lme(TS_detrend_lg2~GrossAsynchrony_s,random=~1+lg2SppN|Site,control=cc,data=stab_444)
Cand.set[[6]]<-lme(TS_detrend_lg2~GrossAsynchrony_s,random=~1+lg2SppN*GrossAsynchrony_s|Site,control=cc,data=stab_444)
Cand.set[[7]]<-lme(TS_detrend_lg2~GrossAsynchrony_s,random=list(~1+lg2SppN+GrossAsynchrony_s|Site),control=cc,data=stab_444)
Modnames <- paste("Mod", 1:length(Cand.set), sep = " ")
res.table <- aictab(cand.set = Cand.set, modnames = Modnames,second.ord = T)
res.table
plot(Cand.set[[best_model]])
qqnorm(Cand.set[[best_model]])
### Log ratio test
big<-lme(TS_detrend_lg2~GrossAsynchrony_s,random=~1+lg2SppN*GrossAsynchrony_s|Site,control=cc,data=stab_444,method="ML")
small<-lme(TS_detrend_lg2~1,random=~1+lg2SppN*GrossAsynchrony_s|Site,control=cc,data=stab_444,method="ML")
anova(big,small)
#final model summary
final<-lme(TS_detrend_lg2~GrossAsynchrony_s,random=list(~1+lg2SppN*GrossAsynchrony_s|Site),control=cc,data=stab_444,method="REML")
r.squaredGLMM(final)
VarCorr(final)
ICClme(final)
###############################
# Phylogenetic diversity #####
##############################
#eMNTD
## Test multiple random effect structures; compare using AICc
Cand.set <- list( )
Cand.set[[1]]<-lme(TS_detrend_lg2~eMNTD,random=~1+eMNTD|Site,control=cc,data=stab_444)
Cand.set[[2]]<-lme(TS_detrend_lg2~eMNTD,random=~1+eMNTD|Site/SppN,control=cc,data=stab_444)
Cand.set[[3]]<-lme(TS_detrend_lg2~eMNTD,random=~1|Site,control=cc,data=stab_444)
Cand.set[[4]]<-lme(TS_detrend_lg2~eMNTD,random=~1|Site/SppN,control=cc,data=stab_444)
Cand.set[[5]]<-lme(TS_detrend_lg2~eMNTD,random=~1+lg2SppN|Site,control=cc,data=stab_444)
Cand.set[[6]]<-lme(TS_detrend_lg2~eMNTD,random=~1+lg2SppN*eMNTD|Site,control=cc,data=stab_444)
Cand.set[[7]]<-lme(TS_detrend_lg2~eMNTD,random=list(~1+lg2SppN+eMNTD|Site),control=cc,data=stab_444)
Modnames <- paste("Mod", 1:length(Cand.set), sep = " ")
res.table <- aictab(cand.set = Cand.set, modnames = Modnames,second.ord = T)
res.table
plot(Cand.set[[best_model]])
qqnorm(Cand.set[[best_model]])
### Log ratio test
big<-lme(TS_detrend_lg2~eMNTD,random=~1+lg2SppN*eMNTD|Site,control=cc,data=stab_444,method="ML")
small<-lme(TS_detrend_lg2~1,random=~1+lg2SppN*eMNTD|Site,control=cc,data=stab_444,method="ML")
anova(big,small)
#final model summary
final<-lme(TS_detrend_lg2~eMNTD,random=~1+lg2SppN*eMNTD|Site,control=cc,data=stab_444)
r.squaredGLMM(final)
ICClme(final)
VarCorr(final)
########################
# Functional diversity #
########################
#FDis
## Test multiple random effect structures; compare using AICc
Cand.set <- list( )
Cand.set[[1]]<-lme(TS_detrend_lg2~FDis4,random=~1+FDis4|Site,control=cc,data=stab_444)
Cand.set[[2]]<-lme(TS_detrend_lg2~FDis4,random=~1+FDis4|Site/SppN,control=cc,data=stab_444)
Cand.set[[3]]<-lme(TS_detrend_lg2~FDis4,random=~1|Site,control=cc,data=stab_444)
Cand.set[[4]]<-lme(TS_detrend_lg2~FDis4,random=~1|Site/SppN,control=cc,data=stab_444)
Cand.set[[5]]<-lme(TS_detrend_lg2~FDis4,random=~1+FDis4|Site,control=cc,data=stab_444)
Cand.set[[6]]<-lme(TS_detrend_lg2~FDis4,random=~1+lg2SppN*FDis4|Site,control=cc,data=stab_444)
Cand.set[[7]]<-lme(TS_detrend_lg2~FDis4,random=list(~1+lg2SppN+FDis4|Site),control=cc,data=stab_444)
Modnames <- paste("Mod", 1:length(Cand.set), sep = " ")
res.table <- aictab(cand.set = Cand.set, modnames = Modnames,second.ord = T)
res.table
plot(Cand.set[[7]])
qqnorm(Cand.set[[7]])
### Log ratio test
big<-lme(TS_detrend_lg2~FDis4,random=list(~1+lg2SppN+FDis4|Site),control=cc,data=stab_444,method="ML")
small<-lme(TS_detrend_lg2~1,random=list(~1+lg2SppN+FDis4|Site),control=cc,data=stab_444,method="ML")
anova(big,small)
#final model summary
final<-lme(TS_detrend_lg2~FDis4,random=list(~1+lg2SppN+FDis4|Site),control=cc,data=stab_444,method="REML")
r.squaredGLMM(final)
ICClme(final)
VarCorr(final)
#Functional composition
## Test multiple random effect structures; compare using AICc
Cand.set <- list( )
Cand.set[[1]]<-lme(TS_detrend_lg2~PCAdim1_4trts,random=~1+PCAdim1_4trts|Site,control=cc,data=stab_444)
Cand.set[[2]]<-lme(TS_detrend_lg2~PCAdim1_4trts,random=~1+PCAdim1_4trts|Site/SppN,control=cc,data=stab_444)
Cand.set[[3]]<-lme(TS_detrend_lg2~PCAdim1_4trts,random=~1|Site,control=cc,data=stab_444)
Cand.set[[4]]<-lme(TS_detrend_lg2~PCAdim1_4trts,random=~1|Site/SppN,control=cc,data=stab_444)
Cand.set[[5]]<-lme(TS_detrend_lg2~PCAdim1_4trts,random=~1+lg2SppN|Site,control=cc,data=stab_444)
Cand.set[[6]]<-lme(TS_detrend_lg2~PCAdim1_4trts,random=~1+lg2SppN*PCAdim1_4trts|Site,control=cc,data=stab_444)
Cand.set[[7]]<-lme(TS_detrend_lg2~PCAdim1_4trts,random=list(~1+lg2SppN+PCAdim1_4trts|Site),control=cc,data=stab_444)
Modnames <- paste("Mod", 1:length(Cand.set), sep = " ")
res.table <- aictab(cand.set = Cand.set, modnames = Modnames,second.ord = T)
res.table
plot(Cand.set[[best_model]])
qqnorm(Cand.set[[best_model]])
### Log ratio test
big<-lme(TS_detrend_lg2~PCAdim1_4trts,random=list(~1+lg2SppN+PCAdim1_4trts|Site),control=cc,data=stab_444,method="ML")
small<-lme(TS_detrend_lg2~1,random=list(~1+lg2SppN+PCAdim1_4trts|Site),control=cc,data=stab_444,method="ML")
anova(big,small)
#final
final<-lme(TS_detrend_lg2~PCAdim1_4trts,random=list(~1+lg2SppN+PCAdim1_4trts|Site),control=cc,data=stab_444,method="REML")
r.squaredGLMM(final)
ICClme(final)
VarCorr(final)