-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy patheider_presentation.qmd
227 lines (156 loc) · 6.77 KB
/
eider_presentation.qmd
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
---
title: "STEI Estimates from the ACP and Triangle Surveys"
author: "Erik Osnas"
date: "`r Sys.Date()`"
format:
revealjs:
embed-resources: true
---
## Outline
- Study Areas
- General Methods
- Results
## Study Areas
```{r}
#| warning: false
#| message: false
library(tidyverse)
library(sf)
path = "../ACP-Mapping/Data/ACP_2023/analysis_output/ACP_DesignStrata_QC.gpkg"
acp <- st_read(dsn = path, quiet = TRUE) #|> st_union()
path = "../ACP-Mapping/Data/ACP_2023/analysis_output/ACP_DesignTrans_QC.gpkg"
trans.acp <- st_read(dsn = path, quiet = TRUE) |>
filter(Year == 2007)
trans <- st_read(dsn = "data/Barrow_STEI_standardized_transects_Aug2024.gpx",
layer = "routes", quiet = TRUE) |>
select(Transect = name) |>
filter(str_sub(Transect, 1, 1) == "A", row_number() > 10)
triangle <- st_read(dsn = "data/Barrow_Triangle_STEI_Aerial_SA", quiet = TRUE)
ggplot(data = acp) + geom_sf(lwd = 1) + geom_sf(data = trans[c(TRUE, FALSE),], col = "red") + geom_sf(data = trans.acp)
```
## Observed STEI locations
```{r}
#| warning: false
#| message: false
#need to plot bird locations
path <- "../ACP-Mapping/Data/ACP_2023/analysis_output/Bird-QC-Obs-2024-03-21.csv"
acp.obs <- birds <- read_csv(file = path, ) %>%
filter(Species == "STEI") |>
st_as_sf(coords = c("Lon", "Lat"), crs = 4326) |>
mutate(Survey = "ACP") |>
select(Survey)
abr.obs <- readxl::read_xlsx(path = "data/STEI_obs_1999-2023.xlsx",
sheet = "STEI Obs 1999-2023") |>
rename(Transect = "Standard Transect") |>
arrange(Year, Transect) |>
st_as_sf(coords = c("LongDD83", "LatDD83"), crs = 4269) |>
st_transform(crs = 4326) |>
mutate(Survey = "Triangle") |>
select(Survey)
df <- rbind(abr.obs, acp.obs)
ggplot(data = acp) + geom_sf() + geom_sf(data = df, aes(col = Survey))
```
## Methods: Chopping up transects
## Methods: Generalized Additive Model
## Generalized Additive Model
$$Count \sim s(X, Y) + s(Year) + ... + s(Observer)$$
- ... = many other models
- $Count$ is the number of pairs and males (flying or not flying?)
- later during prediction this is transformed to "indicated birds"
- uncertainty is estimated by sampling from the model posterior distribution (detection corrections can easily be incorporated during this step)
## Results
- Triangle -- design v. model
- ACP
- Combined
## Results: Triangle
Design-based estimate
::: columns
::: {.column width="50%"}
No fliers 
:::
::: {.column width="50%"}
with fliers 
:::
:::
## Results: Triangle
Design v. model-based estimate (with fliers)
::: columns
::: {.column width="50%"}
:::
::: {.column width="50%"}
:::
:::
## Results: Triangle
## Results: Triangle
## Results: ACP
## Results: ACP
## Results: ACP
Zoomed to Triangle
## Results: Triangle v. ACP
::: columns
::: {.column width="50%"}
:::
::: {.column width="50%"}
:::
:::
## Results: ACP outside Triangle
## Results: Combining Triangle and ACP
## Results: Combining Triangle and ACP
## Results: Combining Triangle and ACP
## Results: Combining Triangle and ACP
## Incorporating detection
Double Observer-Sightability study 2017 - 2023 using decoys
- Detection depended on distance from transect and other covariates
- Decoys on tundra were rarely or never detected
- Unconditional detection estimates were \~1/2 that of conditional
## Incorporating detection
## Results: STEI in Triangle
## Results: ACP and Triangle combined
## Combined and ACP (outside)
::: columns
::: {.column width="50%"}
:::
::: {.column width="50%"}
:::
:::
## Combined (High and Tesh. only)
## Summary {.smaller}
+------------------------------------------+----------------------------------------------------+-----------------------------------------------------------------------+
| Triangle | ACP | Triangle + ACP |
+==========================================+====================================================+=======================================================================+
| -lower variability; high sampling effort | -huge variability; low/sparse sampling effort | -lower bound improved by Triangle data; high uncertain on upper bound |
+------------------------------------------+----------------------------------------------------+-----------------------------------------------------------------------+
| -small covered area | -large covered area | -large covered area |
+------------------------------------------+----------------------------------------------------+-----------------------------------------------------------------------+
| -no spatial or temporal extrapolation | -spatial extrapolation into sparsely sampled areas | -spatial and temporal extrapolation |
+------------------------------------------+----------------------------------------------------+-----------------------------------------------------------------------+
| -20 yr mean (w/ detection): | -20 yr mean (w/ detection): | -20 yr mean (w/ detection): |
| | | |
| 209.6 (170.7, 250.8) | 877 (123 - 6400) | 2134.1 (270.3, 4833) |
+------------------------------------------+----------------------------------------------------+-----------------------------------------------------------------------+
## Design Simulations
## Design Simulations: CV
