-
Notifications
You must be signed in to change notification settings - Fork 0
/
03_SEVT_Effect_of_Spatial_Covariates_on_Return_Levels.R
521 lines (422 loc) · 24 KB
/
03_SEVT_Effect_of_Spatial_Covariates_on_Return_Levels.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
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
setwd("C:/Users/user/Desktop/EVT")
rm(list=ls())
#Load packages
library(dplyr)
library(tidyr)
library(SpatialExtremes)
library(extRemes)
#Load output from 01_SEVT_Max_Stable_Models first
##################################################################################
### To obtain the per unit change in the spatial covariate of interest: ###
### First, impose a unit change in the spatial covariate of interest, then use ###
### the locCoeff for all spatial covariates and matrix multiply with the new ###
### spatial covariate of interest(+0.01) to obtain the new parameters ###
### (location, scale, shape) for each location (for each X and Y). ###
### Finally, inverse the qgev distribution to get the new return levels. ###
##################################################################################
### Best model with lowest deviance: ms4.1t_brown ###
########################################################
#### Spatial covariates with positive coefficients #####
########################################################
## Freshwater surfaces
## impervious Surfaces
## Vegetation with human management (with tree canopy)
## Population size
## Median building age
###################################
###### FRESHWATER SURFACES ########
###################################
loc.form.4.1 <- y ~ xcoord + ls_v1
scale.form.4.1 <- y ~ xcoord + ycoord + ls_v1
shape.form.4.1 <- y ~ 1
data <- ms4.1t_brown$coord
loc.dsgnmat.ms4.1t_brown <- modeldef(data, loc.form.4.1)$dsgn.mat
scale.dsgnmat.ms4.1t_brown <- modeldef(data, scale.form.4.1)$dsgn.mat
shape.dsgnmat.ms4.1t_brown <- modeldef(data, shape.form.4.1)$dsgn.mat
####to get loc.pred ####
unit_change = 0.01 #impose a change of 1%
unit_change = 0.05 #impose a change of 5%
unit_change = 0.1 #impose a change of 10%
View(loc.dsgnmat.ms4.1t_brown)
loc.design.ms4.1t_brown = cbind(loc.dsgnmat.ms4.1t_brown[,c(1:2)],
loc.dsgnmat.ms4.1t_brown[, 3] + unit_change, #add 0.01 to col 3 freshwater
loc.dsgnmat.ms4.1t_brown[,c(4:11)])
colnames(loc.design.ms4.1t_brown)[3] = "freshwater"
View(loc.design.ms4.1t_brown)
loc_coef.ms4.1t_brown = t(ms4.1t_brown$param) #11 coefficients
View(loc_coef.ms4.1t_brown)
loc_coef.ms4.1t_brown = loc_coef.ms4.1t_brown[,c(3:13)] #extract columns 4 to 12
loc_coef_pred.ms4.1t_brown = as.matrix(loc.design.ms4.1t_brown) %*% loc_coef.ms4.1t_brown #matrix multiply (1618x1)
####to get scale.pred ####
View(scale.dsgnmat.ms4.1t_brown)
scale.design.ms4.1t_brown = cbind(scale.dsgnmat.ms4.1t_brown[,c(1:3)],
scale.dsgnmat.ms4.1t_brown[, 4] + unit_change, #add 0.01 to col 4 freshwater
scale.dsgnmat.ms4.1t_brown[,c(5:12)])
colnames(scale.design.ms4.1t_brown)[4] = "freshwater"
View(scale.design.ms4.1t_brown)
scale_coef.ms4.1t_brown = t(ms4.1t_brown$param) #11 coefficients
View(scale_coef.ms4.1t_brown)
scale_coef.ms4.1t_brown = scale_coef.ms4.1t_brown[,c(14:25)]
scale_coef_pred.ms4.1t_brown = as.matrix(scale.design.ms4.1t_brown) %*% scale_coef.ms4.1t_brown
####to get shape.pred #### constant
shape.value.ms4.1t_brown = ms4.1t_brown$param["shapeCoeff1"]
#Invert qgev distribution to produce new 30-year return levels, requires location, scale and shape
ret.per=30
for (T in ret.per) ret.lev.ms4.1t_brown <- qgev(1 - 1/T, loc_coef_pred.ms4.1t_brown,
scale_coef_pred.ms4.1t_brown, shape.value.ms4.1t_brown)
predicted_ms4.1t_brown_1 <- as.data.frame(predicted_ms4.1t_brown)
fw_change = ret.lev.ms4.1t_brown - predicted_ms4.1t_brown_1$Q30
fw_percent_change = ((ret.lev.ms4.1t_brown - predicted_ms4.1t_brown_1$Q30)/predicted_ms4.1t_brown_1$Q30)*100
summary(fw_change) #0.3 increase in return levels for a 10% increase in freshwater surfaces
summary(fw_percent_change) #mean of 0.7% for a 10% increase in freshwater surfaces
### Plot per unit change effect of freshwater on ret levels ###
library(rgeos)
library(rgdal)
library(tmap)
library(grid)
#Load the hexagons shapefile into R
spat <- readOGR(dsn="C:/Users/user/Desktop/EVT/hexagonsData", layer="hexagonsData")
#Change in return levels (per unit change in spatial cov): freshwater of BEST MODEL
ret.lev.ms4.1t_brown = as.data.frame(ret.lev.ms4.1t_brown)
predicted_ms4.1t_brown_df_rl_fw <- cbind(df_xy, ret.lev.ms4.1t_brown)
predicted_ms4.1t_brown_df_rl_fw["V1"][predicted_ms4.1t_brown_df_rl_fw["V1"]<0] <- 0 #force 0 for neg values
predicted_ms4.1t_brown_df_rl_fw$V1<- round(predicted_ms4.1t_brown_df_rl_fw$V1,
0)
colnames(predicted_ms4.1t_brown_df_rl_fw)[4] = "Return Levels"
ms4.1t_brown_rl_fw = merge(spat, predicted_ms4.1t_brown_df_rl_fw, by.x="cell_id", by.y="cell_id")
#1 unit change
ms4.1t_brown_retlev_fw_1 = tm_shape(ms4.1t_brown_rl_fw) +
tm_fill("Return Levels", palette = "Blues", style="fixed",title="", colorNA=grey(0.9),
breaks = c(13, 35, 45, 55, 66, 92),
textNA = "None recorded")+
tm_layout(legend.text.size = 1, legend.title.size = 1.5,
title="A",title.size = 2, frame=F, legend.show=F)
ms4.1t_brown_retlev_fw_1
#5 unit change
ms4.1t_brown_retlev_fw_5 = tm_shape(ms4.1t_brown_rl_fw) +
tm_fill("Return Levels", palette = "Blues", style="fixed",title="", colorNA=grey(0.9),
breaks = c(13, 35, 45, 55, 66, 92),
textNA = "None recorded")+
tm_layout(legend.text.size = 1, legend.title.size = 1.5,
title="B",title.size = 2, frame=F, legend.show=F)
ms4.1t_brown_retlev_fw_5
#10 unit change
ms4.1t_brown_retlev_fw_10 = tm_shape(ms4.1t_brown_rl_fw) +
tm_fill("Return Levels", palette = "Blues", style="fixed",title="", colorNA=grey(0.9),
breaks = c(13, 35, 45, 55, 66, 92), #0, 12, 30, 55, 64, 92
textNA = "None recorded")+
tm_layout(legend.text.size = 1, legend.title.size = 1.5,
title="C",title.size = 2, frame=F, legend.show=T)
ms4.1t_brown_retlev_fw_10
###################################
###### IMPERVIOUS SURFACES ########
###################################
####to get loc.pred ####
unit_change = 0.01
unit_change = 0.05
unit_change = 0.1
View(loc.dsgnmat.ms4.1t_brown)
loc.design.ms4.1t_brown = cbind(loc.dsgnmat.ms4.1t_brown[,c(1:3)],
loc.dsgnmat.ms4.1t_brown[, 4] + unit_change, #add 0.01 to col 4
loc.dsgnmat.ms4.1t_brown[,c(5:11)])
colnames(loc.design.ms4.1t_brown)[4] = "impervious"
View(loc.design.ms4.1t_brown)
loc_coef.ms4.1t_brown = t(ms4.1t_brown$param) #9 coefficients
View(loc_coef.ms4.1t_brown)
loc_coef.ms4.1t_brown = loc_coef.ms4.1t_brown[,c(3:13)] #extract columns 3 to 11
loc_coef_pred.ms4.1t_brown = as.matrix(loc.design.ms4.1t_brown) %*% loc_coef.ms4.1t_brown #matrix multiply (1618x1)
####to get scale.pred ####
View(scale.dsgnmat.ms4.1t_brown)
scale.design.ms4.1t_brown = cbind(scale.dsgnmat.ms4.1t_brown[,c(1:4)],
scale.dsgnmat.ms4.1t_brown[, 5] + unit_change, #add 0.01 to col 4 freshwater
scale.dsgnmat.ms4.1t_brown[,c(6:12)])
colnames(scale.design.ms4.1t_brown)[5] = "impervious"
View(scale.design.ms4.1t_brown)
scale_coef.ms4.1t_brown = t(ms4.1t_brown$param) #11 coefficients
View(scale_coef.ms4.1t_brown)
scale_coef.ms4.1t_brown = scale_coef.ms4.1t_brown[,c(14:25)]
scale_coef_pred.ms4.1t_brown = as.matrix(scale.design.ms4.1t_brown) %*% scale_coef.ms4.1t_brown
####to get shape.pred #### constant
shape.value.ms4.1t_brown = ms4.1t_brown$param["shapeCoeff1"]
#Invert qgev distribution to produce new 30-year return levels, requires location, scale and shape
ret.per=30
for (T in ret.per) ret.lev.ms4.1t_brown <- qgev(1 - 1/T, loc_coef_pred.ms4.1t_brown,
scale_coef_pred.ms4.1t_brown, shape.value.ms4.1t_brown)
predicted_ms4.1t_brown_1 <- as.data.frame(predicted_ms4.1t_brown)
imp_change = ret.lev.ms4.1t_brown - predicted_ms4.1t_brown_1$Q30
imp_percent_change = ((ret.lev.ms4.1t_brown - predicted_ms4.1t_brown_1$Q30)/predicted_ms4.1t_brown_1$Q30)*100
summary(imp_change) #1.6 increase in return levels for a 10% increase in impervious surfaces
summary(imp_percent_change) #3.3% mean change for a 10% increase
### Plot per unit change effect of impervious surfaces on return levels ###
#Change in return levels (per unit change in spatial cov: impervious surfaces of BEST MODEL
ret.lev.ms4.1t_brown = as.data.frame(ret.lev.ms4.1t_brown)
predicted_ms4.1t_brown_df_rl_imp <- cbind(df_xy, ret.lev.ms4.1t_brown)
#View(predicted_ms4.1t_brown_df_rl_imp)
predicted_ms4.1t_brown_df_rl_imp["V1"][predicted_ms4.1t_brown_df_rl_imp["V1"]<0] <- 0 #force 0 for neg values
predicted_ms4.1t_brown_df_rl_imp$V1<- round(predicted_ms4.1t_brown_df_rl_imp$V1,
0)
colnames(predicted_ms4.1t_brown_df_rl_imp)[4] = "Return Levels"
ms4.1t_brown_rl_imp = merge(spat, predicted_ms4.1t_brown_df_rl_imp, by.x="cell_id", by.y="cell_id")
#1 unit change
ms4.1t_brown_retlev_imp_1 = tm_shape(ms4.1t_brown_rl_imp) +
tm_fill("Return Levels", palette = "BuPu", style="fixed",title="", colorNA=grey(0.9),
breaks = c(13, 35, 45, 55, 66, 94),
textNA = "None recorded")+
tm_layout(legend.text.size = 1, legend.title.size = 1.5,
title="D",title.size = 2, frame=F, legend.show=F)
ms4.1t_brown_retlev_imp_1
#5 unit change
ms4.1t_brown_retlev_imp_5 = tm_shape(ms4.1t_brown_rl_imp) +
tm_fill("Return Levels", palette = "BuPu", style="fixed",title="", colorNA=grey(0.9),
breaks = c(13, 35, 45, 55, 66, 94),
textNA = "None recorded")+
tm_layout(legend.text.size = 1, legend.title.size = 1.5,
title="E",title.size = 2, frame=F, legend.show=F)
ms4.1t_brown_retlev_imp_5
#10 unit change
ms4.1t_brown_retlev_imp_10 = tm_shape(ms4.1t_brown_rl_imp) +
tm_fill("Return Levels", palette = "BuPu", style="fixed",title="", colorNA=grey(0.9),
breaks = c(13, 35, 45, 55, 66, 94),
textNA = "None recorded"
)+
tm_layout(legend.text.size = 1, legend.title.size = 1.5,
title="F",title.size = 2, frame=F, legend.show=T)
ms4.1t_brown_retlev_imp_10
##########################################################
### VEGETATION WITH HUMAN MANGEMENT (WITH TREE CANOPY) ###
##########################################################
####to get loc.pred ####
unit_change = 0.01
unit_change = 0.05
unit_change = 0.1
View(loc.dsgnmat.ms4.1t_brown)
loc.design.ms4.1t_brown = cbind(loc.dsgnmat.ms4.1t_brown[,c(1:5)],
loc.dsgnmat.ms4.1t_brown[, 6] + unit_change,
loc.dsgnmat.ms4.1t_brown[,c(7:11)])
colnames(loc.design.ms4.1t_brown)[6] = "veghmtc"
View(loc.design.ms4.1t_brown)
loc_coef.ms4.1t_brown = t(ms4.1t_brown$param) #9 coefficients
View(loc_coef.ms4.1t_brown)
loc_coef.ms4.1t_brown = loc_coef.ms4.1t_brown[,c(3:13)] #extract columns 3 to 11
loc_coef_pred.ms4.1t_brown = as.matrix(loc.design.ms4.1t_brown) %*% loc_coef.ms4.1t_brown #matrix multiply (1618x1)
####to get scale.pred ####
View(scale.dsgnmat.ms4.1t_brown)
scale.design.ms4.1t_brown = cbind(scale.dsgnmat.ms4.1t_brown[,c(1:6)],
scale.dsgnmat.ms4.1t_brown[, 7]
+ unit_change, scale.dsgnmat.ms4.1t_brown[,c(8:12)] )
colnames(scale.design.ms4.1t_brown)[7] = "veghmtc"
View(scale.design.ms4.1t_brown)
scale_coef.ms4.1t_brown = t(ms4.1t_brown$param) #11 coefficients
View(scale_coef.ms4.1t_brown)
scale_coef.ms4.1t_brown = scale_coef.ms4.1t_brown[,c(14:25)]
scale_coef_pred.ms4.1t_brown = as.matrix(scale.design.ms4.1t_brown) %*% scale_coef.ms4.1t_brown
####to get shape.pred #### constant
shape.value.ms4.1t_brown = ms4.1t_brown$param["shapeCoeff1"]
#Invert qgev distribution to produce new 30-year return levels, requires location, scale and shape
ret.per=30
for (T in ret.per) ret.lev.ms4.1t_brown <- qgev(1 - 1/T,
loc_coef_pred.ms4.1t_brown,
scale_coef_pred.ms4.1t_brown,
shape.value.ms4.1t_brown)
predicted_ms4.1t_brown_1 <- as.data.frame(predicted_ms4.1t_brown)
veghmtc_change = ret.lev.ms4.1t_brown - predicted_ms4.1t_brown_1$Q30
veghmtc_percent_change = ((ret.lev.ms4.1t_brown - predicted_ms4.1t_brown_1$Q30)/predicted_ms4.1t_brown_1$Q30)*100
summary(veghmtc_change) #0.7 increase in return levels for a 10% increase in veghmtc surfaces
summary(veghmtc_percent_change) #mean of 1.4% for a 10% increase in veghmtc surfaces
### Plot per unit change effect of veghmtc on ret levels ###
#Change in return levels (per unit change in spatial cov: veghmtc of BEST MODEL
ret.lev.ms4.1t_brown = as.data.frame(ret.lev.ms4.1t_brown)
predicted_ms4.1t_brown_df_rl_veghmtc <- cbind(df_xy, ret.lev.ms4.1t_brown)
predicted_ms4.1t_brown_df_rl_veghmtc["V1"][predicted_ms4.1t_brown_df_rl_veghmtc["V1"]<0] <- 0 #force 0 for neg values
predicted_ms4.1t_brown_df_rl_veghmtc$V1<- round(predicted_ms4.1t_brown_df_rl_veghmtc$V1,
0)
colnames(predicted_ms4.1t_brown_df_rl_veghmtc)[4] = "Return Levels"
ms4.1t_brown_rl_veghmtc = merge(spat, predicted_ms4.1t_brown_df_rl_veghmtc, by.x="cell_id", by.y="cell_id")
#1 unit change
ms4.1t_brown_retlev_veghmtc_1 = tm_shape(ms4.1t_brown_rl_veghmtc) +
tm_fill("Return Levels", palette = "BuGn", style="fixed",title="", colorNA=grey(0.9),
breaks = c(13, 35, 45, 55, 65, 93),
textNA = "None recorded")+
tm_layout(legend.text.size = 1, legend.title.size = 1.5,
title="G",title.size = 2, frame=F, legend.show=F)
ms4.1t_brown_retlev_veghmtc_1
#5 unit change
ms4.1t_brown_retlev_veghmtc_5 = tm_shape(ms4.1t_brown_rl_veghmtc) +
tm_fill("Return Levels", palette = "BuGn", style="fixed",title="", colorNA=grey(0.9),
breaks = c(13, 35, 45, 55, 65, 93),
textNA = "None recorded")+
tm_layout(legend.text.size = 1, legend.title.size = 1.5,
title="H",title.size = 2, frame=F, legend.show=F)
ms4.1t_brown_retlev_veghmtc_5
#10 unit change
ms4.1t_brown_retlev_veghmtc_10 = tm_shape(ms4.1t_brown_rl_veghmtc) +
tm_fill("Return Levels", palette = "BuGn", style="fixed",title="", colorNA=grey(0.9),
breaks = c(13, 35, 45, 55, 65, 93),
textNA = "None recorded"
)+
tm_layout(legend.text.size = 1, legend.title.size = 1.5,
title="I",title.size = 2, frame=F, legend.show=T)
ms4.1t_brown_retlev_veghmtc_10
########################
### TOTAL POPULATION ###
########################
####to get loc.pred ####
unit_change = 0.01
unit_change = 0.05
unit_change = 0.1
View(loc.dsgnmat.ms4.1t_brown)
loc.design.ms4.1t_brown = cbind(loc.dsgnmat.ms4.1t_brown[,c(1:9)],
loc.dsgnmat.ms4.1t_brown[, 10]*(1+unit_change),
loc.dsgnmat.ms4.1t_brown[,c(11)])
colnames(loc.design.ms4.1t_brown)[10] = "totalpop"
View(loc.design.ms4.1t_brown)
loc_coef.ms4.1t_brown = t(ms4.1t_brown$param) #9 coefficients
View(loc_coef.ms4.1t_brown)
loc_coef.ms4.1t_brown = loc_coef.ms4.1t_brown[,c(3:13)] #extract columns 3 to 11
loc_coef_pred.ms4.1t_brown = as.matrix(loc.design.ms4.1t_brown) %*% loc_coef.ms4.1t_brown #matrix multiply (1618x1)
####to get scale.pred ####
View(scale.dsgnmat.ms4.1t_brown)
scale.design.ms4.1t_brown = cbind(scale.dsgnmat.ms4.1t_brown[,c(1:10)],
scale.dsgnmat.ms4.1t_brown[, 11]*(1+unit_change),
scale.dsgnmat.ms4.1t_brown[,c(12)] )
colnames(scale.design.ms4.1t_brown)[11] = "totalpop"
View(scale.design.ms4.1t_brown)
scale_coef.ms4.1t_brown = t(ms4.1t_brown$param) #11 coefficients
View(scale_coef.ms4.1t_brown)
scale_coef.ms4.1t_brown = scale_coef.ms4.1t_brown[,c(14:25)]
scale_coef_pred.ms4.1t_brown = as.matrix(scale.design.ms4.1t_brown) %*% scale_coef.ms4.1t_brown
####to get shape.pred #### constant
shape.value.ms4.1t_brown = ms4.1t_brown$param["shapeCoeff1"]
#Invert qgev distribution to produce new 30-year return levels, requires location, scale and shape
ret.per=30
for (T in ret.per) ret.lev.ms4.1t_brown <- qgev(1 - 1/T,
loc_coef_pred.ms4.1t_brown,
scale_coef_pred.ms4.1t_brown,
shape.value.ms4.1t_brown)
predicted_ms4.1t_brown_1 <- as.data.frame(predicted_ms4.1t_brown)
pop_change = ret.lev.ms4.1t_brown - predicted_ms4.1t_brown_1$Q30
pop_percent_change = ((ret.lev.ms4.1t_brown - predicted_ms4.1t_brown_1$Q30)/predicted_ms4.1t_brown_1$Q30)*100
summary(pop_change) #0.1 increase in return levels for a 10% increase in population
summary(pop_percent_change) #abs value mean of 0.3% for a 10% increase in population
### Plot per unit change effect of population on ret levels ###
#Change in return levels (per unit change in spatial cov: population of BEST MODEL
ret.lev.ms4.1t_brown = as.data.frame(ret.lev.ms4.1t_brown)
predicted_ms4.1t_brown_df_rl_totalpop <- cbind(df_xy, ret.lev.ms4.1t_brown)
predicted_ms4.1t_brown_df_rl_totalpop["V1"][predicted_ms4.1t_brown_df_rl_totalpop["V1"]<0] <- 0 #force 0 for neg values
predicted_ms4.1t_brown_df_rl_totalpop$V1<- round(predicted_ms4.1t_brown_df_rl_totalpop$V1,
0)
colnames(predicted_ms4.1t_brown_df_rl_totalpop)[4] = "Return Levels"
ms4.1t_brown_rl_totalpop = merge(spat, predicted_ms4.1t_brown_df_rl_totalpop, by.x="cell_id", by.y="cell_id")
#1 unit change
ms4.1t_brown_retlev_totalpop_1 = tm_shape(ms4.1t_brown_rl_totalpop) +
tm_fill("Return Levels", palette = "RdPu", style="fixed",title="", colorNA=grey(0.9),
breaks = c(13, 35, 45, 55, 65, 92),
textNA = "None recorded")+
tm_layout(legend.text.size = 1, legend.title.size = 1.5,
title="J",title.size = 2, frame=F, legend.show=F)
ms4.1t_brown_retlev_totalpop_1
#5 unit change
ms4.1t_brown_retlev_totalpop_5 = tm_shape(ms4.1t_brown_rl_totalpop) +
tm_fill("Return Levels", palette = "RdPu", style="fixed",title="", colorNA=grey(0.9),
breaks = c(13, 35, 45, 55, 65, 92),
textNA = "None recorded")+
tm_layout(legend.text.size = 1, legend.title.size = 1.5,
title="K",title.size = 2, frame=F, legend.show=F)
ms4.1t_brown_retlev_totalpop_5
#10 unit change
ms4.1t_brown_retlev_totalpop_10 = tm_shape(ms4.1t_brown_rl_totalpop) +
tm_fill("Return Levels", palette = "RdPu", style="fixed",title="", colorNA=grey(0.9),
breaks = c(13, 35, 45, 55, 65, 92),
textNA = "None recorded"
)+
tm_layout(legend.text.size = 1, legend.title.size = 1.5,
title="L",title.size = 2, frame=F, legend.show=T)
ms4.1t_brown_retlev_totalpop_10
#######################################
### MEDIAN AGE OF PUBLIC APARTMENTS ###
#######################################
####to get loc.pred ####
unit_change = 1 #Impose an increase of 1 year
unit_change = 2 #Impose an increase of 2 years
unit_change = 3 #Impose an increase of 3 years
View(loc.dsgnmat.ms4.1t_brown)
loc.design.ms4.1t_brown = cbind(loc.dsgnmat.ms4.1t_brown[,c(1:10)],
loc.dsgnmat.ms4.1t_brown[, 11] + unit_change)
colnames(loc.design.ms4.1t_brown)[11] = "hdbmedage"
View(loc.design.ms4.1t_brown)
loc_coef.ms4.1t_brown = t(ms4.1t_brown$param) #9 coefficients
View(loc_coef.ms4.1t_brown)
loc_coef.ms4.1t_brown = loc_coef.ms4.1t_brown[,c(3:13)] #extract columns 3 to 11
loc_coef_pred.ms4.1t_brown = as.matrix(loc.design.ms4.1t_brown) %*% loc_coef.ms4.1t_brown #matrix multiply (1618x1)
####to get scale.pred ####
View(scale.dsgnmat.ms4.1t_brown)
scale.design.ms4.1t_brown = cbind(scale.dsgnmat.ms4.1t_brown[,c(1:11)],
scale.dsgnmat.ms4.1t_brown[, 12]
+ unit_change)
colnames(scale.design.ms4.1t_brown)[12] = "hdbmedage"
View(scale.design.ms4.1t_brown)
scale_coef.ms4.1t_brown = t(ms4.1t_brown$param) #11 coefficients
View(scale_coef.ms4.1t_brown)
scale_coef.ms4.1t_brown = scale_coef.ms4.1t_brown[,c(14:25)]
scale_coef_pred.ms4.1t_brown = as.matrix(scale.design.ms4.1t_brown) %*% scale_coef.ms4.1t_brown
####to get shape.pred #### constant,no need
shape.value.ms4.1t_brown = ms4.1t_brown$param["shapeCoeff1"]
#Invert qgev distribution to produce new 30-year return levels, requires location, scale and shape
ret.per=30
for (T in ret.per) ret.lev.ms4.1t_brown <- qgev(1 - 1/T,
loc_coef_pred.ms4.1t_brown,
scale_coef_pred.ms4.1t_brown,
shape.value.ms4.1t_brown)
predicted_ms4.1t_brown_1 <- as.data.frame(predicted_ms4.1t_brown)
hdbage_change = ret.lev.ms4.1t_brown - predicted_ms4.1t_brown_1$Q30
hdbage_percent_change = ((ret.lev.ms4.1t_brown - predicted_ms4.1t_brown_1$Q30)/predicted_ms4.1t_brown_1$Q30)*100
summary(hdbage_change) #1.8 increase in return levels for a 3 year increase in building age
summary(hdbage_percent_change) #3.8% mean for a 3 year increase in building age
### Plot per unit change effect of age of public apartment on ret levels ###
#Change in return levels (per unit change in spatial cov: building age of BEST MODEL
ret.lev.ms4.1t_brown = as.data.frame(ret.lev.ms4.1t_brown)
predicted_ms4.1t_brown_df_rl_hdbmedage <- cbind(df_xy, ret.lev.ms4.1t_brown)
predicted_ms4.1t_brown_df_rl_hdbmedage["V1"][predicted_ms4.1t_brown_df_rl_hdbmedage["V1"]<0] <- 0 #force 0 for neg values
predicted_ms4.1t_brown_df_rl_hdbmedage$V1<- round(predicted_ms4.1t_brown_df_rl_hdbmedage$V1,
0)
colnames(predicted_ms4.1t_brown_df_rl_hdbmedage)[4] = "Return Levels"
ms4.1t_brown_rl_hdbmedage = merge(spat, predicted_ms4.1t_brown_df_rl_hdbmedage, by.x="cell_id", by.y="cell_id")
#1 year increase
ms4.1t_brown_retlev_hdbmedage_1 = tm_shape(ms4.1t_brown_rl_hdbmedage) +
tm_fill("Return Levels", palette = "Oranges", style="fixed",title="", colorNA=grey(0.9),
breaks = c(14, 36, 46, 56, 66, 94),
textNA = "None recorded")+
tm_layout(legend.text.size = 1, legend.title.size = 1.5,
title="M",title.size = 2, frame=F, legend.show=F)
ms4.1t_brown_retlev_hdbmedage_1
#2 year increase
ms4.1t_brown_retlev_hdbmedage_2 = tm_shape(ms4.1t_brown_rl_hdbmedage) +
tm_fill("Return Levels", palette = "Oranges", style="fixed",title="", colorNA=grey(0.9),
breaks = c(14, 36, 46, 56, 66, 94),
textNA = "None recorded")+
tm_layout(legend.text.size = 1, legend.title.size = 1.5,
title="N",title.size = 2, frame=F, legend.show=F)
ms4.1t_brown_retlev_hdbmedage_2
#3 year increase
ms4.1t_brown_retlev_hdbmedage_3 = tm_shape(ms4.1t_brown_rl_hdbmedage) +
tm_fill("Return Levels", palette = "Oranges", style="fixed",title="", colorNA=grey(0.9),
breaks = c(14, 36, 46, 56, 66, 94),
textNA = "None recorded"
)+
tm_layout(legend.text.size = 1, legend.title.size = 1.5,
title="O",title.size = 2, frame=F, legend.show=T)
ms4.1t_brown_retlev_hdbmedage_3
#For MANUSCRIPT overleaf; combined into 1 plot
png("Per_unitchange_spatcov NEW.png", units="cm", width=40, height=50, res=300)
tmap_arrange(ms4.1t_brown_retlev_fw_1, ms4.1t_brown_retlev_fw_5,
ms4.1t_brown_retlev_fw_10,
ms4.1t_brown_retlev_imp_1, ms4.1t_brown_retlev_imp_5,
ms4.1t_brown_retlev_imp_10,
ms4.1t_brown_retlev_veghmtc_1, ms4.1t_brown_retlev_veghmtc_5,
ms4.1t_brown_retlev_veghmtc_10,
ms4.1t_brown_retlev_totalpop_1, ms4.1t_brown_retlev_totalpop_5,
ms4.1t_brown_retlev_totalpop_10,
ms4.1t_brown_retlev_hdbmedage_1, ms4.1t_brown_retlev_hdbmedage_2,
ms4.1t_brown_retlev_hdbmedage_3,
ncol=3)
dev.off()