-
Notifications
You must be signed in to change notification settings - Fork 0
/
02_Deception_comparisons.Rmd
292 lines (191 loc) · 9.56 KB
/
02_Deception_comparisons.Rmd
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
---
title: "Deception Study are DEMs"
author: "G. Perkins"
date: "22/02/2021"
output: html_document
---
```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE,
warning = FALSE,
message = FALSE)
```
Within the Deception study site, we can compare the TRIM, eDTM, isat points along with lidar data.
```{r load data, echo = FALSE }
# Read in geometic data set (independent of data assembly process) and compare to
# ensemble dtm, lidar, trim.
library(raster)
library(sf)
library(bcdata)
library(bcmaps)
library(dplyr)
library(rvest)
library(ggplot2)
library(tidyr)
## Load in eDTM layer and trim (from sync location)
data.dir <- "D:/Hengle_BCDEM30"
sync.dir <- "C:/Sync/BC_DEM"
temp.dir <- file.path(data.dir,"temp")
liDAR.dir <-
"D:\\PEM_DATA\\BEC_DevExchange_Work\\Deception_AOI\\1_map_inputs\\covariates"
aoi <- st_read(file.path(
"D:\\PEM_DATA\\BEC_DevExchange_Work\\Deception_AOI\\0_raw_inputs\\base_layers\\aoi.gpkg"), quiet = TRUE)
#BC Albers proj4 string
bcalb = '+proj=aea +lat_1=50 +lat_2=58.5 +lat_0=45 +lon_0=-126 +x_0=1000000 +y_0=0 +ellps=GRS80 +towgs84=0,0,0,0,0,0,0 +units=m +no_defs +datum=NAD83'
if(file.exists(file.path(temp.dir, "deception_dem_pt.gpkg"))){
#print("loading file")
geopt <- st_read(file.path(temp.dir, "deception_dem_pt.gpkg"), quiet = TRUE)
} else {
# load in provinical DEMs
# 1) edtm:
edtm <- raster(file.path(data.dir, "BC30m","dtm_elev.lowestmode_gedi.eml_m_30m_0..0cm_2000..2018_bc.epsg3005_v0.1.tif"))
#2 - TRIM:
trim <- raster(file.path(data.dir, "trim", "TRIM","bc_elevation_25m_bcalb.tif"))
#3 - LiDAR
l2_5 <- raster(file.path(liDAR.dir, "2.5m", "dem.tif"))
l5 <- raster(file.path(liDAR.dir, "5m", "dem.tif"))
l10 <- raster(file.path(liDAR.dir, "10m", "dem.tif"))
l25 <- raster(file.path(liDAR.dir, "25m", "dem.tif"))
# load in the isat2 points
isat <- st_read(file.path(temp.dir, "isat2_deception.gpkg"), quiet = TRUE)
isat1 <- isat[1:112,] # south to north - westly transect
isat2 <- isat[113:211,] # south to north - easterly transect
isat3 <- isat[212:349,] # north to south - middle transect
geopt <- isat %>%
mutate(edtm = raster::extract(edtm, isat)) %>%
mutate(trim = raster::extract(trim, isat)) %>%
mutate(l2_5 = raster::extract(l2_5, isat)) %>%
mutate(l5 = raster::extract(l5, isat)) %>%
mutate(l10 = raster::extract(l10, isat)) %>%
mutate(l25 = raster::extract(l25, isat))
geopt <- geopt %>%
mutate(rn = row_number()) %>%
mutate(trans = case_when(
rn < 113 ~ "1",
rn >= 113 & rn <212 ~ "2",
rn > 211 ~ "3"
))
st_write(geopt, file.path(temp.dir, "deception_dem_pt.gpkg"))
}
## add tiled data - found no difference between pts
# edtm_tile <- raster(file.path(data.dir, "BC30m", "tt30_tiles","T1265", "T1265", "dtm_M_T1265.tif"))
#
# geopt <- geopt %>%
# mutate(edtm_tile = raster::extract(edtm_tile, geopt))
#
# unique(geopt$edtm_tile -geopt$edtm )
```
## Cross Section comparisons
The isat point data crossed within the deception boundary for 3 passes. Split into transects 1, 2, 3. Transect 2 and 3 represented evenly spaced points, while isat transect 1 had gaps.
{width=50%}
```{r, echo = FALSE}
# # transect 2
# ggplot(geopt) +
# geom_point(aes(x = rn, y = edtm, colour = "red")) +
# geom_point(aes(x = rn, y = h_te_mean, colour = "green")) +
# geom_point(aes(x = rn, y = trim, colour = "blue")) +
# geom_line(aes(x = rn, y = edtm, colour = "red")) +
# geom_line(aes(x = rn, y = trim, colour = "blue")) +
# facet_wrap(~trans, scales = "free_x")
#
```
#### Compare between the data sources (trim, eDTM, isat, lidar (2.5 - 25m))
Firstly compare the eDTM, trim and isat points for the two consistent transects (2 - easterly edge) and 3 (middle transect).
```{r , echo = FALSE}
geopt <- geopt %>% filter(trans %in% c("2","3"))
geolong <- geopt %>%
pivot_longer(cols = c(edtm, trim,l2_5 ,l5,l10,l25,h_te_mean), names_to = "dem_type", values_to = "value") %>%
mutate(dem_type = ifelse(dem_type == "h_te_mean", "isat", dem_type))
geolong2 <- geolong %>%
filter(dem_type %in% c("edtm", "trim", "isat"))
geolong22 <- geolong2 %>% filter(trans =="2")
geolong23 <- geolong2 %>% filter(trans =="3")
ggplot(geolong22, aes(x = rn, y = value)) +
geom_point(aes(colour = dem_type)) +
geom_line(aes(colour = dem_type)) +
facet_grid(~trans, scales = "free_x" )
ggplot(geolong23, aes(x = rn, y = value)) +
geom_point(aes(colour = dem_type)) +
geom_line(aes(colour = dem_type)) +
facet_grid(~trans, scales = "free_x" )
```
- Trim is consistently higher elevation than eDTM or isat points.
- complex relief (ie transect 3 - 275m to 300m has more variation in patterns)
### Compare lidar to eDTM and TRIM datasets.
Compare the 25m lidar upscaled values to eDTM and TRIM. Tested variation in lidar reolution (2.5, 5, 10,25 and showed little difference in values )
```{r, echo = FALSE}
geolong3 <- geolong %>%
filter(dem_type %in% c("edtm", "trim", "l25"))
geolong22 <- geolong3 %>% filter(trans =="2")
geolong23 <- geolong3 %>% filter(trans =="3")
ggplot(geolong22, aes(x = rn, y = value)) +
geom_point(aes(colour = dem_type)) +
geom_line(aes(colour = dem_type))
ggplot(geolong23, aes(x = rn, y = value)) +
geom_point(aes(colour = dem_type)) +
geom_line(aes(colour = dem_type))
```
# Deception AOI extent.
```{r, eval = FALSE, echo = FALSE}
#BC Albers proj4 string
bcalb = '+proj=aea +lat_1=50 +lat_2=58.5 +lat_0=45 +lon_0=-126 +x_0=1000000 +y_0=0 +ellps=GRS80 +towgs84=0,0,0,0,0,0,0 +units=m +no_defs +datum=NAD83'
# read in the datasets
l25 <- raster(file.path(liDAR.dir, "25m", "dem.tif"))
#2 - TRIM:
trim <- raster(file.path(data.dir, "trim", "TRIM","bc_elevation_25m_bcalb.tif"))
# mosic the tiles for Deception together
t1 <- raster(file.path(data.dir, "BC30m", "tt30_tiles","T1265", "T1265", "dtm_M_T1265.tif"))
t2 <- raster(file.path(data.dir, "BC30m", "tt30_tiles","T1264", "T1264", "dtm_M_T1264.tif"))
t3 <- raster(file.path(data.dir, "BC30m", "tt30_tiles","T1318", "T1318", "dtm_M_T1318.tif"))
t4 <- raster(file.path(data.dir, "BC30m", "tt30_tiles","T1319", "T1319", "dtm_M_T1319.tif"))
edtm30 <- mosaic(t1, t2, t3, t4, fun = mean)
##crop the trim to AOI (lidar extent)
l25_dec <- crop(l25, aoi)
trim_dec <- crop(trim, aoi)
edtm30_dec <- crop(edtm30, aoi)
extent(edtm30_dec)
extent(l25_dec)
extent(trim_dec)
edtm30_dec_rs <- resample(edtm30_dec, l25_dec, "bilinear")
trim_dec_rs <- resample(trim_dec, l25_dec, "bilinear" )
edtm30_dec_rs <- crop(edtm30_dec_rs, l25_dec)
trim_dec_rs <- crop(trim_dec_rs, l25_dec)
dems <- stack(edtm30_dec_rs, trim_dec_rs, l25_dec)
# writeRaster(dems, temp.dir, "dems.tif")
#
# writeRaster(l25_dec , file.path(temp.dir,'lidar_dec_stack.tif'), overwrite = TRUE)
# writeRaster(edtm30_dec_rs , file.path(temp.dir,'edtm_dec_stack.tif'), overwrite = TRUE)
# writeRaster( trim_dec_rs, file.path(temp.dir,'trim_dec_stack.tif'), overwrite = TRUE)
delta_lidar <- dems$dem - dems$layer
delta_trim <- dems$dem - dems$bc_elevation_25m_bcalb
writeRaster(delta_lidar , file.path(temp.dir,'lidar_edtm_diff.tif'), overwrite = TRUE)
writeRaster(delta_trim , file.path(temp.dir,'lidar_trim_diff.tif'), overwrite = TRUE)
#
# sf::gdal_utils(
# util = "warp",
# source = file.path(temp.dir,'edtm_dec.tif'),
# destination = file.path(temp.dir,"edtm_dec_3005.tif"),
# options = c("-t_srs", "EPSG:3005",
# "-r", "cubic"))
```
I calculated the difference from lidar to the eDTM and TRIM data sets with Deception study area. Note for this to occur some reprojection was necesary to line up lidar, eDTM and TRIM datasets.
The eDTM delta values ranged from (-9 to 47.4 ) indicating the underprediction of eDTM in general. The delta or lidar - trim = (-24 to 34) indicating more spread in uncertainty by underestimate and over estimate of values.
{width=45%}
{width=45%}
### Cutblocks
Zooming in on an areas of interest to assess the impact of cutblocks
{width=50%}
{width=50%}
{width=50%}
## Areas of steep terrain.
{width=50%}
{width=50%}
{width=50%}
Interestingly in some areas where TRIM showed larger error (ie. - 25m) the eDTM was more accuracy.
{width=50%}
{width=50%}
### Overall thoughts
- TRIM estimates showed larger proportion of data with -5m to +5m than eDTM.
- eDTM showed bias to under estimate of elevation.
- eDTM sensitive to cutblocks
- both TRIM and EDTM showed higher error in areas with high hillshade.
- eDTM outperformed TRIM in some areas.