chl-variability-machine-learning/2_plot_simple_TS_Chl_ice.R at main · vlampe/chl-variability-machine-learning · GitHub

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
## plot time series
## author: Vanessa Lampe
## date: 2023-10-23
## version: 0
## git:

library(tidyverse)
library(readxl)
library(writexl)
library(ggpubr)
source("weighted_moving_average.R")
wmv <- weighted_moving_average   # reassign function bc laziness


## read output from extract_warm_and_cold_clusters.R
ts_dat <- read_csv2("output/Output_3k_cluster_analysis_thesis.csv", na = c("", "Inf", "-Inf","NA"))


#_______________________________________________________________________________
# Simple plots

# TS of mean total Chl in FS
ggplot(data = ts_dat, aes(x = date, y = `mean chl total [mg / m^3]`)) +
  geom_line()

# compare SST in clusters
ggplot(data = ts_dat, aes(x = date)) +
  geom_line(aes(y = `mean temp warm [ºC]`, col = "warm")) +
  geom_ribbon(aes(ymin = `mean temp warm [ºC]` - `sd temp warm [ºC]`,
                  ymax = `mean temp warm [ºC]` + `sd temp warm [ºC]`,
                  fill = "warm"), alpha = .3) +
  geom_line(aes(y = `mean temp cold [ºC]`, col = "cold")) +
  geom_ribbon(aes(ymin = `mean temp cold [ºC]` - `sd temp cold [ºC]`,
                  ymax = `mean temp cold [ºC]` + `sd temp cold [ºC]`,
                  fill = "cold"), alpha = .3) +
  geom_line(aes(y = `mean temp front [ºC]`, col = "front")) +
  geom_ribbon(aes(ymin = `mean temp front [ºC]` - `sd temp front [ºC]`,
                  ymax = `mean temp front [ºC]` + `sd temp front [ºC]`,
                  fill = "front"), alpha = .3) +
  labs(y = "STT [ºC]", colour = "cluster") +
  scale_x_date(name = "",
               date_breaks = "1 month",
               date_labels = "%b") +
#  scale_y_continuous(limits = c(0, NA), minor_breaks = F, name = "Chl a conc. [µg / L]") +
  scale_color_manual(aesthetics =c("colour", "fill"), values = c("warm" = "red", "cold" = "blue","front" = "grey"),
                     breaks = c("cold", "front", "warm")) +
  guides(fill = "none") +
  theme_bw()


# compare raw Chl in clusters
ggplot(data = ts_dat, aes(x = date)) +
  geom_line(aes(y = `mean chl warm [mg / m^3]`, col = "warm")) +
  geom_line(aes(y = `mean chl cold [mg / m^3]`, col = "cold")) +
  geom_line(aes(y = `mean chl front [mg / m^3]`, col = "front")) +
  labs(y = expression(paste("Chl ", italic("a"), " mg ", m^{-3})), colour = "cluster") +
 # labs(y = "Chl *a* [mg m^(-3)] ", colour = "cluster") +
  scale_x_date(name = "",
               date_breaks = "1 month",
               date_labels = "%b") +
  #  scale_y_continuous(limits = c(0, NA), minor_breaks = F, name = "Chl a conc. [µg / L]") +
  scale_color_manual(values = c("warm" = "red", "cold" = "blue","front" = "grey"),
                     breaks = c("cold", "front", "warm"))
  #theme(axis.title.y = ggtext::element_markdown())


# compare chl in clusters, weighed moving average
wmeans <- ts_dat
wmeans$w_w_chl <- wmv(values = wmeans$`mean chl warm [mg / m^3]`,
                      weights = wmeans$`n.obs. chl warm [pixels]`,
                      width = 15)
wmeans$w_c_chl <- wmv(values = wmeans$`mean chl cold [mg / m^3]`,
                      weights = wmeans$`n.obs chl cold [pixels]`,
                      width = 15)
wmeans$w_f_chl <- wmv(values = wmeans$`mean chl front [mg / m^3]`,
                      weights = wmeans$`n.obs. chl front [pixels]`,
                      width = 15)


ggplot(data = wmeans, aes(x = date)) +
  geom_line(aes(y = (`w_w_chl`), col = "warm")) +
  geom_line(aes(y = `w_c_chl`, col = "cold")) +
  geom_line(aes(y = `w_f_chl`, col = "front")) +
  scale_color_manual(values = c("warm" = "red", "cold" = "blue","front" = "grey"),
                     breaks = c("cold", "front", "warm")) +
  labs(y = expression(paste("Chl ", italic("a"), " mg ", m^{-3})), colour = "cluster")
# total dazu in schwarz


# compare PFT compositon in clusters

# rearrange data a bit

PFT_dat <- ts_dat %>%
  select(date, `mean PICO warm [mg / m^3]`:`mean MICRO front [mg / m^3]`) %>%
  pivot_longer(-date, names_to = c("type", "regime"), names_prefix = "mean ", names_sep = " ", values_to = "mean_vals")

chl_dat <- ts_dat %>%
  select(date, `mean chl warm [mg / m^3]`, `mean chl cold [mg / m^3]`, `mean chl front [mg / m^3]`) %>%
  pivot_longer(-date, names_to = c("type", "regime"), names_prefix = "mean ", names_sep = " ", values_to = "mean_vals")

# line plots
ggplot(data = PFT_dat, aes(x = date, y = mean_vals, col = type)) +
  geom_line() +
  facet_grid(.~regime) +
  scale_x_date(name = "",
               date_breaks = "1 month",
               date_labels = "%b",
               limits = as.Date(c("2018-04-01", "2018-09-15")))


# stacked area plot
ggplot(data = PFT_dat, aes(x = date, y = mean_vals, group = type)) +
  geom_area(aes(fill = type)) +
  facet_grid(.~regime) +
  scale_x_date(name = "",
               date_breaks = "1 month",
               date_labels = "%b",
               limits = as.Date(c("2018-04-01", "2018-09-15"))) +
geom_line(data = chl_dat)
# -> Markus sagt: pft algos sind noch nicht komplett ausgereift. Daher kein match total chl zu sum(pfts)
# ich konzentriere mich weiter auf total chl


# compare Ice in clusters
ggplot(data = ts_dat, aes(x = date)) +
  geom_line(aes(y = `ice cover warm region [%]`*100, col = "warm")) +
  geom_ribbon(aes(ymin = (`ice cover warm region [%]` - `ice cover warm region sd [%]`)*100,
                  ymax = (`ice cover warm region [%]` + `ice cover warm region sd [%]`)*100,
                  fill = "warm"), alpha = .3) +
  geom_line(aes(y = `ice cover cold region [%]`*100, col = "cold")) +
  geom_ribbon(aes(ymin = (`ice cover cold region [%]` - `ice cover cold region  sd [%]`)*100,
                  ymax = (`ice cover cold region [%]` + `ice cover cold region  sd [%]`)*100,
                  fill = "cold"), alpha = .3) +
  geom_line(aes(y = `ice cover front region [%]`*100, col = "front")) +
  geom_ribbon(aes(ymin = (`ice cover front region [%]` - `ice cover front region sd [%]`)*100,
                  ymax = (`ice cover front region [%]` + `ice cover front region sd [%]`)*100,
                  fill = "front"), alpha = .3) +
  geom_abline(intercept = 15, slope = 0, linetype = "dotted") +
  labs(y = "sea ice cover [%]", colour = "cluster") +
  scale_x_date(name = "",
               date_breaks = "1 month",
               date_labels = "%b") +
  scale_y_continuous(breaks = c(0, 15, 25, 50, 75, 100)) +
  scale_color_manual(aesthetics =c("colour", "fill"), values = c("warm" = "red", "cold" = "blue","front" = "grey"),
                     breaks = c("cold", "front", "warm")) +
  guides(fill = "none") +
  coord_cartesian(ylim=c(-0.7, 100), expand = F) +
  theme_bw()