Scratch.qmd

---
title: "Climate and Productivity"
format: html
---


```{r}
#| warning: false
#| echo: false
library(targets)
library(ggplot2)
library(dplyr)
library(bayesplot)
library(patchwork)
library(viridis)

```


[Website](https://ssc.ca/en/case-study/understanding-how-canadas-economy-might-be-impacted-climate-change)


# What data do we have?

We have weather stations at the following points:
```{r}
#| code-fold: true
#| code-summary: "Show the code"
map <- tar_read(plot_geoms) + 
  geom_point(data = tar_read(climate_dat), aes(x = Longitude..x., y = Latitude..y.))

map
```


```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

color_scheme_set(scheme = "teal")


dat_trunc <- tar_read(climate_dat) %>%
  # filter(Date.Time < "Jan 1999") %>%
  filter(Station.Name == "BROMPTONVILLE")

p <- ggplot(data = dat_trunc, aes(x = Month,
                                  y = Snow.Grnd.Last.Day..cm.,
                                  color = Year,
                                  group = Year)) +
  geom_point() +
  geom_line() + 
  theme_default() +
  scale_color_viridis(option = "inferno") + 
  ggtitle("Change in Mean Temp Over Time")


p
```

## Climate Data

We have the following parameters in the climate dataset:

::: {.panel-tabset}
## Longitude..x., Latitude..y.

Just lat, lon. 

## Station.Name

Gives name of weather station. See also: Date.Time.

## Climate.ID

I don't know what this is yet. Someone should find out 

- Jason

## Date.Time (Year, Month)

Not all the stations have coverage throughout the entire range. 

```{r}
#| echo: false
#| warning: false
dat_a <- tar_read(climate_dat) %>% filter(substr(Station.Name, 1, 1) == "A")
dat_b <- tar_read(climate_dat) %>% filter(substr(Station.Name, 1, 1) == "B")
dat_c <- tar_read(climate_dat) %>% filter(substr(Station.Name, 1, 1) == "C")
dat_d <- tar_read(climate_dat) %>% filter(substr(Station.Name, 1, 1) == "D")
dat_e <- tar_read(climate_dat) %>% filter(substr(Station.Name, 1, 1) == "E")
dat_f <- tar_read(climate_dat) %>% filter(substr(Station.Name, 1, 1) == "F")
dat_g <- tar_read(climate_dat) %>% filter(substr(Station.Name, 1, 1) == "G")
dat_h <- tar_read(climate_dat) %>% filter(substr(Station.Name, 1, 1) == "H")
dat_i <- tar_read(climate_dat) %>% filter(substr(Station.Name, 1, 1) == "I")
dat_j <- tar_read(climate_dat) %>% filter(substr(Station.Name, 1, 1) == "J")
dat_k <- tar_read(climate_dat) %>% filter(substr(Station.Name, 1, 1) == "K")
dat_l <- tar_read(climate_dat) %>% filter(substr(Station.Name, 1, 1) == "L")
dat_m <- tar_read(climate_dat) %>% filter(substr(Station.Name, 1, 1) == "M")
dat_n <- tar_read(climate_dat) %>% filter(substr(Station.Name, 1, 1) == "N")
dat_o <- tar_read(climate_dat) %>% filter(substr(Station.Name, 1, 1) == "O")
dat_p <- tar_read(climate_dat) %>% filter(substr(Station.Name, 1, 1) == "P")
dat_q <- tar_read(climate_dat) %>% filter(substr(Station.Name, 1, 1) == "Q")
dat_r <- tar_read(climate_dat) %>% filter(substr(Station.Name, 1, 1) == "R")
dat_s <- tar_read(climate_dat) %>% filter(substr(Station.Name, 1, 1) == "S")
dat_t <- tar_read(climate_dat) %>% filter(substr(Station.Name, 1, 1) == "T")
dat_u <- tar_read(climate_dat) %>% filter(substr(Station.Name, 1, 1) == "U")
dat_v <- tar_read(climate_dat) %>% filter(substr(Station.Name, 1, 1) == "V")
dat_w <- tar_read(climate_dat) %>% filter(substr(Station.Name, 1, 1) == "W")
dat_x <- tar_read(climate_dat) %>% filter(substr(Station.Name, 1, 1) == "X")
dat_y <- tar_read(climate_dat) %>% filter(substr(Station.Name, 1, 1) == "Y")
dat_z <- tar_read(climate_dat) %>% filter(substr(Station.Name, 1, 1) == "Z")
```

::: {.panel-tabset}

## A
```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

ggplot(dat_a, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `A` Names") +
  theme_default()

```

## B
```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

d1 <- filter(dat_b, substr(Station.Name, 2, 2) < "N")
d2 <- filter(dat_b, substr(Station.Name, 2, 2) >= "N")

p1 <- ggplot(d1, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `B` Names") +
  theme_default()

p2 <- ggplot(d2, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `B` Names") +
  theme_default()

p1

p2

```

## C
```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

c1 <- filter(dat_c, substr(Station.Name, 2, 2) < "N")
c2 <- filter(dat_c, substr(Station.Name, 2, 2) >= "N")

ggplot(c1, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `C` Names") +
  theme_default()

ggplot(c2, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `C` Names") +
  theme_default()

```



## D

```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

d1 <- filter(dat_d, substr(Station.Name, 2, 2) < "N")
d2 <- filter(dat_d, substr(Station.Name, 2, 2) >= "N")

ggplot(d1, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `D` Names") +
  theme_default()

ggplot(d2, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `D` Names") +
  theme_default()
```

## E

```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

e1 <- filter(dat_e, substr(Station.Name, 2, 2) <= "Z")

ggplot(e1, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `E` Names") +
  theme_default()

```



## F

```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

f1 <- filter(dat_f, substr(Station.Name, 2, 2) < "N")
f2 <- filter(dat_f, substr(Station.Name, 2, 2) >= "N")

ggplot(f1, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `F` Names") +
  theme_default()

ggplot(f2, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `F` Names") +
  theme_default()
```

## G
```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

g1 <- filter(dat_g, substr(Station.Name, 2, 2) < "N")
g2 <- filter(dat_g, substr(Station.Name, 2, 2) >= "N")

ggplot(g1, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `G` Names") +
  theme_default()

ggplot(g2, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `G` Names") +
  theme_default()
```

## H

```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

h1 <- filter(dat_h, substr(Station.Name, 2, 2) < "N")
h2 <- filter(dat_h, substr(Station.Name, 2, 2) >= "N")

ggplot(h1, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `H` Names") +
  theme_default()

ggplot(h2, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `H` Names") +
  theme_default()
```

## I
```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

i1 <- filter(dat_i, substr(Station.Name, 2, 2) <= "N")

ggplot(i1, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `I` Names") +
  theme_default()

```

## J
```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

j1 <- filter(dat_j, substr(Station.Name, 2, 2) <= "Z")

ggplot(j1, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `J` Names") +
  theme_default()

```

## K
```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

k1 <- filter(dat_k, substr(Station.Name, 2, 2) <= "Z")

ggplot(k1, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `K` Names") +
  theme_default()

```

## L

```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

l1 <- filter(dat_l, substr(Station.Name, 2, 2) < "N")
l2 <- filter(dat_l, substr(Station.Name, 2, 2) >= "N")

ggplot(l1, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `L` Names") +
  theme_default()

ggplot(l2, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `L` Names") +
  theme_default()
```

## M

```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

m1 <- filter(dat_m, substr(Station.Name, 2, 2) < "N")
m2 <- filter(dat_m, substr(Station.Name, 2, 2) >= "N")

ggplot(m1, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `M` Names") +
  theme_default()

ggplot(m2, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `M` Names") +
  theme_default()
```

## N


```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

n1 <- filter(dat_n, substr(Station.Name, 2, 2) < "N")
n2 <- filter(dat_n, substr(Station.Name, 2, 2) >= "N")

ggplot(n1, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `N` Names") +
  theme_default()

ggplot(n2, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `N` Names") +
  theme_default()
```


## O

```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

o1 <- filter(dat_o, substr(Station.Name, 2, 2) < "N")
o2 <- filter(dat_o, substr(Station.Name, 2, 2) >= "N")

ggplot(o1, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `O` Names") +
  theme_default()

ggplot(o2, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `O` Names") +
  theme_default()
```

## P
```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

p1 <- filter(dat_p, substr(Station.Name, 2, 2) < "N")
p2 <- filter(dat_p, substr(Station.Name, 2, 2) >= "N")

ggplot(p1, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `P` Names") +
  theme_default()

ggplot(p2, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `P` Names") +
  theme_default()
```

## Q
```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

q1 <- filter(dat_q, substr(Station.Name, 2, 2) <= "Z")

ggplot(q1, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `O` Names") +
  theme_default()
```

## R
```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

r1 <- filter(dat_r, substr(Station.Name, 2, 2) <= "Z")

ggplot(r1, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `R` Names") +
  theme_default()
```

## S
```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

s1 <- filter(dat_s, substr(Station.Name, 2, 2) <= "P")
s2 <- filter(dat_s, (substr(Station.Name, 2, 2) > "P") & (substr(Station.Name, 2, 2) <= "T"))
s3 <- filter(dat_s, substr(Station.Name, 2, 2) > "T")


ggplot(s1, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `S` Names") +
  theme_default()

ggplot(s2, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `S` Names") +
  theme_default()

ggplot(s3, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `S` Names") +
  theme_default()
```

## T
```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

t1 <- filter(dat_t, substr(Station.Name, 2, 2) <= "Z")

ggplot(t1, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `T` Names") +
  theme_default()
```


## U
```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

u1 <- filter(dat_u, substr(Station.Name, 2, 2) <= "Z")

ggplot(u1, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `U` Names") +
  theme_default()
```

## V
```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

v1 <- filter(dat_v, substr(Station.Name, 2, 2) <= "Z")

ggplot(v1, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `V` Names") +
  theme_default()
```

## W
```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

w1 <- filter(dat_w, substr(Station.Name, 2, 2) <= "Z")

ggplot(w1, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `W` Names") +
  theme_default()
```

## X

Note: No X names.
```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

x1 <- filter(dat_x, substr(Station.Name, 2, 2) <= "Z")

ggplot(x1, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `X` Names") +
  theme_default()
```

## Y
```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

y1 <- filter(dat_y, substr(Station.Name, 2, 2) <= "Z")

ggplot(y1, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `Y` Names") +
  theme_default()
```

## Z

Note: there are no Z names.
```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

z1 <- filter(dat_z, substr(Station.Name, 2, 2) <= "Z")

ggplot(z1, aes(x = Date.Time, y = Station.Name)) + 
  geom_line() +
  ggtitle("Coverage — `Z` Names") +
  theme_default()
```


:::


## Mean.Max.Temp...C.

Also don't know what this is yet, but I can guess...

```{r}
#| code-fold: true
#| code-summary: "Show the code"
#| warning: false

dat_trunc <- tar_read(climate_dat) %>%
  # filter(Date.Time < "Jan 1999") %>%
  filter(Station.Name == "MAGOG")

p <- ggplot(data = dat_trunc, aes(x = Month,
                                  y = Mean.Max.Temp...C.,
                                  color = Year,
                                  group = Year)) +
  geom_point() +
  geom_line() + 
  theme_default() +
  scale_color_viridis(option = "inferno") + 
  ggtitle("Change in Mean Max Temp Over Time — MAGOG")


p
```

## Mean.Max.Temp.Flag


:::

## Productivity Data



# Add Geom to climate data:

```{r}
tar_load(climate_dat)
tar_load(on_fort)


```