-
Notifications
You must be signed in to change notification settings - Fork 0
/
FQ_VMS_analysis.R
449 lines (359 loc) · 28.4 KB
/
FQ_VMS_analysis.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
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
##################################
#Code for Funiculina quadrangularis fishing effects study as part of the Small Isles data report results
##Created by Rebecca Langton 08 September 2022
##Adapted from code by Philip Boulcott
##For sources of underlying data and abbreviations, see the report
##################################
########################################################################
# Analysis of the VMS FQ data
# 5 quadrats per station - approximately 30 stations (sometimes 25) per box
# The quadrat station design is used to increase the quadrat area given the density of FQ in the test areas and the fixed photographic area
########################################################################
# Section 1. Reading in the data
########################################################################
#set directories
rm(list=ls())
getwd()
dir<-substr(getwd(), 1, nchar(getwd())-25)
dir
#load libraries
library(pbkrtest) # Estimate p-values computed using a Kenward-Roger correction is not available in glmer - need this for family =poisson
#reads in the dataset with variables centred using the median as a reference value.
dtaS= read.csv("FQ_2017_dtaStation.csv", header=T) #this is a dataset that has been
## Now to produce a second simplified dataset called dtaSdrop
# Without the box which data exploration suggested was an outlier
dtaSdrop = dtaS[dtaS$fBox != c("VM21"),]
############################################################################################
#Section 2 - mean density
############################################################################################
(with(dtaS,tapply(Funiculina.quadrangularis, fBox, mean)))
# ordering it to help read through
sort(with(dtaS,tapply(Funiculina.quadrangularis, fBox, mean)))
names(dtaS)
# Writing out a csv that matches the densities figure in the manuscript - see WC_FQ_Survey map
FQDEN_TABLE_MANUSCRIPT = data.frame((with(dtaS,tapply(Funiculina.quadrangularis, fBox, mean))))
colnames(FQDEN_TABLE_MANUSCRIPT) = c("density")
FQDEN_TABLE_MANUSCRIPT$oldnames =row.names(FQDEN_TABLE_MANUSCRIPT)
FQDEN_TABLE_MANUSCRIPT$newnames = paste0("F",1:29)
FQDEN_TABLE_MANUSCRIPT$vms =(with(dtaS,tapply(SurfSAR, fBox, mean)))
FQDEN_TABLE_MANUSCRIPT
FQDEN_TABLE_MANUSCRIPT$Area=dtaS$fArea[match(FQDEN_TABLE_MANUSCRIPT$oldnames,dtaS$fBox)]
write.csv(FQDEN_TABLE_MANUSCRIPT, file= "FQ_2017_VMS_Manuscript_Densities.csv")
#This does produce the same table as that in the report
############################################################################################
# Section 3. Fitting the model and model selection
############################################################################################
# dropping out terms - using backwards selection of AIC
#uses full dataset
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
md10<- glmer(Funiculina.quadrangularis ~ cdepth + cvms + cmud + cgravel + cminsal +cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
summary (md10)
md10.D<- glmer(Funiculina.quadrangularis ~ cvms + cmud + cgravel + cminsal +cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md10.V<- glmer(Funiculina.quadrangularis ~ cdepth + cmud + cgravel + cminsal+cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md10.M<- glmer(Funiculina.quadrangularis ~ cdepth + cvms + cgravel + cminsal +cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md10.G<- glmer(Funiculina.quadrangularis ~ cdepth + cvms + cmud + cminsal +cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md10.S = glmer(Funiculina.quadrangularis ~ cdepth + cvms + cmud + cgravel +cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md10.SL = glmer(Funiculina.quadrangularis ~ cdepth + cvms + cmud + cgravel + cminsal +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md10.C = glmer(Funiculina.quadrangularis ~ cdepth + cvms + cmud + cgravel + cminsal+cslope + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md10.A<- glmer(Funiculina.quadrangularis ~ cdepth + cvms + cmud + cgravel + cminsal+cslope +ccurvature + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
AIC(md10,md10.D, md10.V, md10.M, md10.G, md10.S, md10.SL,md10.C,md10.A) #Gravel drops first
# df AIC
# md10 12 1210.270
# md10.D 11 1212.753
# md10.V 11 1211.193
# md10.M 11 1209.170
# md10.G 11 1208.625 *lowest
# md10.S 11 1210.720
# md10.SL 11 1215.143
# md10.C 11 1223.541
# md10.A 10 1221.506
summary(md10) # this is supported by the output summary for the maximal model
anova(md10,md10.G)
# Data: dtaS
# Models:
# md10.G: Funiculina.quadrangularis ~ cdepth + cvms + cmud + cminsal + cslope + ccurvature + fArea + (1 | fBox) + (1 | fStation.No:fBox)
# md10: Funiculina.quadrangularis ~ cdepth + cvms + cmud + cgravel + cminsal + cslope + ccurvature + fArea + (1 | fBox) + (1 | fStation.No:fBox)
# npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
# md10.G 11 1208.6 1262.0 -593.31 1186.6
# md10 12 1210.3 1268.5 -593.13 1186.3 0.3548 1 0.5514
md11<- glmer(Funiculina.quadrangularis ~ cdepth + cvms + cmud + cminsal+cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md11.D<- glmer(Funiculina.quadrangularis ~ cvms + cmud + cminsal+cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md11.V<- glmer(Funiculina.quadrangularis ~ cdepth + cmud + cminsal+cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md11.M<- glmer(Funiculina.quadrangularis ~ cdepth + cvms + cminsal+cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md11.S<- glmer(Funiculina.quadrangularis ~ cdepth + cvms + cmud +cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md11.SL<- glmer(Funiculina.quadrangularis ~ cdepth + cvms + cmud + cminsal+ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md11.C<- glmer(Funiculina.quadrangularis ~ cdepth + cvms + cmud + cminsal+cslope + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md11.A<- glmer(Funiculina.quadrangularis ~ cdepth + cvms + cmud + cminsal+cslope +ccurvature + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
summary(md11)
AIC(md11,md11.D, md11.V, md11.M, md11.S, md11.SL,md11.C,md11.A) # salinity drops out
#
# df AIC
# md11 11 1208.625
# md11.D 10 1210.990
# md11.V 10 1209.368
# md11.M 10 1209.704
# md11.S 10 1209.693 *lowest
# md11.SL 10 1214.910
# md11.C 10 1221.541
# md11.A 9 1220.923
anova(md11,md11.S) # chisq= 3.068 df = 1, pr = 0.08 for Salinty
#Data: dtaS
# Models:
# md11.S: Funiculina.quadrangularis ~ cdepth + cvms + cmud + cslope + ccurvature + fArea + (1 | fBox) + (1 | fStation.No:fBox)
# md11: Funiculina.quadrangularis ~ cdepth + cvms + cmud + cminsal + cslope + ccurvature + fArea + (1 | fBox) + (1 | fStation.No:fBox)
# npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
# md11.S 10 1209.7 1258.2 -594.85 1189.7
# md11 11 1208.6 1262.0 -593.31 1186.6 3.0687 1 0.07981 .
# ---
# Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
md12<- glmer(Funiculina.quadrangularis ~ cdepth + cvms + cmud +cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md12.D<- glmer(Funiculina.quadrangularis ~ cvms + cmud +cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md12.V<- glmer(Funiculina.quadrangularis ~ cdepth + cmud +cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md12.M<- glmer(Funiculina.quadrangularis ~ cdepth + cvms+cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md12.SL<- glmer(Funiculina.quadrangularis ~ cdepth + cvms + cmud +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md12.C<- glmer(Funiculina.quadrangularis ~ cdepth + cvms + cmud +cslope + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md12.A<- glmer(Funiculina.quadrangularis ~ cdepth + cvms + cmud +cslope +ccurvature + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
summary (md12)
AIC(md12,md12.D, md12.V, md12.M, md12.SL,md12.C,md12.A) # VMS drops out
# df AIC
# md12 10 1209.693
# md12.D 9 1211.521
# md12.V 9 1208.931 *lowest
# md12.M 9 1209.183
# md12.SL 9 1220.281
# md12.C 9 1221.743
# md12.A 8 1219.352
anova(md12,md12.V) # chisq= 1.238 df = 1, pr = 0.265 for VMS
# Data: dtaS
# Models:
# md12.V: Funiculina.quadrangularis ~ cdepth + cmud + cslope + ccurvature + fArea + (1 | fBox) + (1 | fStation.No:fBox)
# md12: Funiculina.quadrangularis ~ cdepth + cvms + cmud + cslope + ccurvature + fArea + (1 | fBox) + (1 | fStation.No:fBox)
# npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
# md12.V 9 1208.9 1252.6 -595.47 1190.9
# md12 10 1209.7 1258.2 -594.85 1189.7 1.2381 1 0.2658
md13 <- glmer(Funiculina.quadrangularis ~ cdepth + cmud +cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md13.D <- glmer(Funiculina.quadrangularis ~ cmud +cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md13.M <- glmer(Funiculina.quadrangularis ~ cdepth +cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md13.SL <- glmer(Funiculina.quadrangularis ~ cdepth + cmud +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md13.C <- glmer(Funiculina.quadrangularis ~ cdepth + cmud +cslope + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md13.A <- glmer(Funiculina.quadrangularis ~ cdepth + cmud +cslope +ccurvature + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaS, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
summary (md13)
AIC(md13,md13.D, md13.M, md13.SL,md13.C,md13.A) #
# df AIC
# md13 9 1208.931
# md13.D 8 1210.537
# md13.M 8 1209.819
# md13.SL 8 1219.995
# md13.C 8 1221.289
# md13.A 7 1218.362
anova(md13,md13.D) # chisq= 3.606 df = 1, pr = 0.057 for VMS
# Data: dtaS
# Models:
# md13.D: Funiculina.quadrangularis ~ cmud + cslope + ccurvature + fArea + (1 | fBox) + (1 | fStation.No:fBox)
# md13: Funiculina.quadrangularis ~ cdepth + cmud + cslope + ccurvature + fArea + (1 | fBox) + (1 | fStation.No:fBox)
# npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
# md13.D 8 1210.5 1249.3 -597.27 1194.5
# md13 9 1208.9 1252.6 -595.47 1190.9 3.606 1 0.05757 .
# ---
# Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#
# To find the P values and Chi sqrs of the above md13
summary (md13)
#Generalized linear mixed model fit by maximum likelihood (Laplace Approximation) ['glmerMod']
# Family: poisson ( log )
# Formula: Funiculina.quadrangularis ~ cdepth + cmud + cslope + ccurvature + fArea + (1 | fBox) + (1 | fStation.No:fBox)
# Data: dtaS
# Control: glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 1e+05))
#
# AIC BIC logLik deviance df.resid
# 1208.9 1252.6 -595.5 1190.9 934
#
# Scaled residuals:
# Min 1Q Median 3Q Max
# -1.2803 -0.4641 -0.1696 -0.0801 6.3170
#
# Random effects:
# Groups Name Variance Std.Dev.
# fStation.No:fBox (Intercept) 0.5419 0.7361
# fBox (Intercept) 1.7977 1.3408
# Number of obs: 943, groups: fStation.No:fBox, 943; fBox, 29
#
# Fixed effects:
# Estimate Std. Error z value Pr(>|z|)
# (Intercept) -4.26616 0.66720 -6.394 1.61e-10 ***
# cdepth -0.26299 0.13992 -1.880 0.060172 .
# cmud 0.26400 0.15107 1.748 0.080549 .
# cslope -0.48563 0.14481 -3.354 0.000798 ***
# ccurvature -0.16584 0.04456 -3.722 0.000198 ***
# fAreaSK 3.08094 0.90936 3.388 0.000704 ***
# fAreaWR 3.15672 0.93235 3.386 0.000710 ***
# ---
# Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#
# Correlation of Fixed Effects:
# (Intr) cdepth cmud cslope ccrvtr fAreSK
# cdepth 0.500
# cmud -0.325 -0.244
# cslope -0.147 -0.098 0.110
# ccurvature 0.225 0.187 -0.122 -0.027
# fAreaSK -0.797 -0.622 0.055 0.099 -0.180
# fAreaWR -0.797 -0.524 0.422 0.130 -0.190 0.638
anova(md13, md13.D) # chisq= 3.606 df = 1, pr = 0.058 for Depth ; negative relationship
# Models:
# md13.D: Funiculina.quadrangularis ~ cmud + cslope + ccurvature + fArea + (1 | fBox) + (1 | fStation.No:fBox)
# md13: Funiculina.quadrangularis ~ cdepth + cmud + cslope + ccurvature + fArea + (1 | fBox) + (1 | fStation.No:fBox)
# npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
# md13.D 8 1210.5 1249.3 -597.27 1194.5
# md13 9 1208.9 1252.6 -595.47 1190.9 3.606 1 0.05757 .
# ---
# Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
anova(md13, md13.M) # chisq= 2.888 df = 1, pr = 0.08 for Mud ; positive relationship
# Data: dtaS
# Models:
# md13.M: Funiculina.quadrangularis ~ cdepth + cslope + ccurvature + fArea + (1 | fBox) + (1 | fStation.No:fBox)
# md13: Funiculina.quadrangularis ~ cdepth + cmud + cslope + ccurvature + fArea + (1 | fBox) + (1 | fStation.No:fBox)
# npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
# md13.M 8 1209.8 1248.6 -596.91 1193.8
# md13 9 1208.9 1252.6 -595.47 1190.9 2.888 1 0.08924 .
# ---
# Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
anova(md13, md13.SL) # chisq= 4.371 df = 1, pr < 0.001 for Slope ; negative relationship
# Data: dtaS
# Models:
# md13.SL: Funiculina.quadrangularis ~ cdepth + cmud + ccurvature + fArea + (1 | fBox) + (1 | fStation.No:fBox)
# md13: Funiculina.quadrangularis ~ cdepth + cmud + cslope + ccurvature + fArea + (1 | fBox) + (1 | fStation.No:fBox)
# npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
# md13.SL 8 1220.0 1258.8 -602.00 1204.0
# md13 9 1208.9 1252.6 -595.47 1190.9 13.064 1 0.000301 ***
# ---
# Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
anova(md13, md13.C) # chisq= 14.357 df = 1, pr < 0.001 for Curvature ; positive relationship
# Data: dtaS
# Models:
# md13.C: Funiculina.quadrangularis ~ cdepth + cmud + cslope + fArea + (1 | fBox) + (1 | fStation.No:fBox)
# md13: Funiculina.quadrangularis ~ cdepth + cmud + cslope + ccurvature + fArea + (1 | fBox) + (1 | fStation.No:fBox)
# npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
# md13.C 8 1221.3 1260.1 -602.64 1205.3
# md13 9 1208.9 1252.6 -595.47 1190.9 14.357 1 0.0001512 ***
# ---
# Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
anova(md13, md13.A) # chisq= 13.430 df = 2, pr < 0.01 for Area ; SMI < Skye < WR
# Data: dtaS
# Models:
# md13.A: Funiculina.quadrangularis ~ cdepth + cmud + cslope + ccurvature + (1 | fBox) + (1 | fStation.No:fBox)
# md13: Funiculina.quadrangularis ~ cdepth + cmud + cslope + ccurvature + fArea + (1 | fBox) + (1 | fStation.No:fBox)
# npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
# md13.A 7 1218.4 1252.3 -602.18 1204.4
# md13 9 1208.9 1252.6 -595.47 1190.9 13.43 2 0.001212 **
# ---
# Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
############################################################################################
# Section 3. Fitting the model and model selection without outlier
############################################################################################
md14<- glmer(Funiculina.quadrangularis ~ cdepth + cvms + cmud + cgravel + cminsal +cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
summary (md14)
md14.D<- glmer(Funiculina.quadrangularis ~ cvms + cmud + cgravel + cminsal +cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md14.V<- glmer(Funiculina.quadrangularis ~ cdepth + cmud + cgravel + cminsal+cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md14.M<- glmer(Funiculina.quadrangularis ~ cdepth + cvms + cgravel + cminsal +cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md14.G<- glmer(Funiculina.quadrangularis ~ cdepth + cvms + cmud + cminsal +cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md14.S = glmer(Funiculina.quadrangularis ~ cdepth + cvms + cmud + cgravel +cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md14.SL = glmer(Funiculina.quadrangularis ~ cdepth + cvms + cmud + cgravel + cminsal +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md14.C = glmer(Funiculina.quadrangularis ~ cdepth + cvms + cmud + cgravel + cminsal+cslope + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md14.A<- glmer(Funiculina.quadrangularis ~ cdepth + cvms + cmud + cgravel + cminsal+cslope +ccurvature + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
AIC(md14,md14.D, md14.V, md14.M, md14.G, md14.S, md14.SL,md14.C,md14.A) #Gravel drops first - to give a lower AIC :
# d14 12 1062.837
# md14.D 11 1066.103
# md14.V 11 1061.766
# md14.M 11 1061.821
# md14.G 11 1061.196 *lowest
# md14.S 11 1062.644
# md14.SL 11 1065.697
# md14.C 11 1072.694
# md14.A 10 1074.394
summary(md14) # this is supported by the output summary for the maximal model
anova(md14,md14.G)
# Data: dtaSdrop
# Models:
# md14.G: Funiculina.quadrangularis ~ cdepth + cvms + cmud + cminsal + cslope + ccurvature + fArea + (1 | fBox) + (1 | fStation.No:fBox)
# md14: Funiculina.quadrangularis ~ cdepth + cvms + cmud + cgravel + cminsal + cslope + ccurvature + fArea + (1 | fBox) + (1 | fStation.No:fBox)
# npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
# md14.G 11 1061.2 1114.2 -519.60 1039.2
# md14 12 1062.8 1120.6 -519.42 1038.8 0.3585 1 0.5493
# dropping Gravel
md15<- glmer(Funiculina.quadrangularis ~ cdepth + cvms + cmud + cminsal+cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md15.D<- glmer(Funiculina.quadrangularis ~ cvms + cmud + cminsal+cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md15.V<- glmer(Funiculina.quadrangularis ~ cdepth + cmud + cminsal+cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md15.M<- glmer(Funiculina.quadrangularis ~ cdepth + cvms + cminsal+cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md15.S<- glmer(Funiculina.quadrangularis ~ cdepth + cvms + cmud +cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md15.SL<- glmer(Funiculina.quadrangularis ~ cdepth + cvms + cmud + cminsal+ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md15.C<- glmer(Funiculina.quadrangularis ~ cdepth + cvms + cmud + cminsal+cslope + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md15.A<- glmer(Funiculina.quadrangularis ~ cdepth + cvms + cmud + cminsal+cslope +ccurvature + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
AIC(md15,md15.D, md15.V, md15.M, md15.S, md15.SL,md15.C,md15.A) #VMS drops first - to give a lower AIC :
# df AIC
# md15 11 1061.196
# md15.D 10 1064.470
# md15.V 10 1059.943 *lowest
# md15.M 10 1062.765
# md15.S 10 1061.542
# md15.SL 10 1065.207
# md15.C 10 1070.743
# md15.A 9 1073.458
summary(md15) # this is supported by the output summary for the maximal model
anova(md15,md15.V) # chisq= 0.747, df = 1, pr = 0.38 for VMS
# Data: dtaSdrop
# Models:
# md15.V: Funiculina.quadrangularis ~ cdepth + cmud + cminsal + cslope + ccurvature + fArea + (1 | fBox) + (1 | fStation.No:fBox)
# md15: Funiculina.quadrangularis ~ cdepth + cvms + cmud + cminsal + cslope + ccurvature + fArea + (1 | fBox) + (1 | fStation.No:fBox)
# npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
# md15.V 10 1059.9 1108.1 -519.97 1039.9
# md15 11 1061.2 1114.2 -519.60 1039.2 0.7474 1 0.3873
#drop VMS
md16<- glmer(Funiculina.quadrangularis ~ cdepth + cmud + cminsal+cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md16.D<- glmer(Funiculina.quadrangularis ~ cmud + cminsal+cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md16.M<- glmer(Funiculina.quadrangularis ~ cdepth + cminsal+cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md16.S<- glmer(Funiculina.quadrangularis ~ cdepth + cmud + cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md16.SL<- glmer(Funiculina.quadrangularis ~ cdepth + cmud + cminsal +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md16.C<- glmer(Funiculina.quadrangularis ~ cdepth + cmud + cminsal+cslope + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md16.A<- glmer(Funiculina.quadrangularis ~ cdepth + cmud + cminsal+cslope +ccurvature + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
AIC(md16,md16.D,md16.M,md16.S,md16.SL,md16.C,md16.A)
# df AIC
# md16 10 1059.943
# md16.D 9 1062.998
# md16.M 9 1062.445
# md16.S 9 1059.650 *lowest
# md16.SL 9 1065.249
# md16.C 9 1069.623
# md16.A 8 1071.626
anova(md16,md16.S)
#drop salinity
# Data: dtaSdrop
# Models:
# md16.S: Funiculina.quadrangularis ~ cdepth + cmud + cslope + ccurvature + fArea + (1 | fBox) + (1 | fStation.No:fBox)
# md16: Funiculina.quadrangularis ~ cdepth + cmud + cminsal + cslope + ccurvature + fArea + (1 | fBox) + (1 | fStation.No:fBox)
# npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
# md16.S 9 1059.7 1103.0 -520.82 1041.7
# md16 10 1059.9 1108.1 -519.97 1039.9 1.7067 1 0.1914
md17<- glmer(Funiculina.quadrangularis ~ cdepth + cmud + cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md17.D<- glmer(Funiculina.quadrangularis ~ cmud + cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md17.M<- glmer(Funiculina.quadrangularis ~ cdepth + cslope +ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md17.SL<- glmer(Funiculina.quadrangularis ~ cdepth + cmud + ccurvature + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md17.C<- glmer(Funiculina.quadrangularis ~ cdepth + cmud + cslope + fArea + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
md17.A<- glmer(Funiculina.quadrangularis ~ cdepth + cmud + cslope +ccurvature + (1|fBox) + (1|fStation.No:fBox),family = poisson, data = dtaSdrop, glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 100000)))
AIC(md17,md17.D,md17.M,md17.SL,md17.C,md17.A)
# df AIC
# md17 9 1059.650 *lowest
# md17.D 8 1062.206
# md17.M 8 1060.569
# md17.SL 8 1067.530
# md17.C 8 1068.579
# md17.A 7 1069.808
anova(md17,md17.M)
# Data: dtaSdrop
# Models:
# md17.M: Funiculina.quadrangularis ~ cdepth + cslope + ccurvature + fArea + (1 | fBox) + (1 | fStation.No:fBox)
# md17: Funiculina.quadrangularis ~ cdepth + cmud + cslope + ccurvature + fArea + (1 | fBox) + (1 | fStation.No:fBox)
# npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
# md17.M 8 1060.6 1099.1 -522.28 1044.6
# md17 9 1059.7 1103.0 -520.82 1041.7 2.9193 1 0.08752 .
# ---
# Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1