lab 4.qmd

---
title: "lab 4"
format: html
editor: visual
---
#Name: Liying Deng

```{r}
library(tidyverse)
library(data.table)
library(leaflet)
library(dplyr)
```
#Question 1
```{r}
if (!file.exists("met_all.gz"))
  download.file(
    url = "https://raw.githubusercontent.com/USCbiostats/data-science-data/master/02_met/met_all.gz",
    destfile = "met_all.gz",
    method   = "libcurl",
    timeout  = 60
    )
met <- data.table::fread("met_all.gz")
```

#Question 2
```{r}
met <- met[met$temp > -17, ]
met[met$elev==9999.0, ] <- NA
library(lubridate)
met[,date:= as.Date(paste(year, month, day, sep= "-"))]
met_avg <- met[,.(
  temp     = mean(temp,na.rm=TRUE),
  rh       = mean(rh,na.rm=TRUE),
  wind.sp  = mean(wind.sp,na.rm=TRUE),
  vis.dist = mean(vis.dist,na.rm=TRUE),
  dew.point = mean(dew.point, na.rm = T),
  lat      = mean(lat),
  lon      = mean(lon), 
  elev     = mean(elev,na.rm=TRUE)
), by=c("USAFID")]
met_avg[, region := fifelse(lon < -98 & lat >= 39.71, "NW",   
               fifelse(lon < -98 & lat < 39.71, "SW",       
               fifelse(lon >= -98 & lat >= 39.71, "NE",     
                       "SE")))]  
met_avg[, elev_cat  := cut(
  x      = elev,
  breaks = c(-Inf, 0, 1500, 3000, Inf),
  labels = c("low", "med", "high", "very high"),
  right  = FALSE
)]
```

#Question 3
```{r}
met_avg <- met_avg [!is.na(wind.sp)&!is.na(dew.point)]
ggplot(met_avg)+
  geom_violin(aes(x=1, y=wind.sp, fill="wind.sp"))+
  geom_violin(aes(x=1, y=dew.point, fill="dew.point"))+
  facet_grid(~region)
```
# Each region exhibits distinct characteristics, leading to variations in wind speed. The Southwest region, in particular, has significantly higher dew point levels compared to other regions, with a relatively large range of variation.

#Question 4
```{r}
met_avg[!is.na(wind.sp)&!is.na(dew.point)] |>
  ggplot(mapping = aes(x = dew.point, y = wind.sp, color = region, linetype = region)) +
  geom_point() + 
  geom_smooth(method = lm, se = FALSE, col = "Blue")
```
#The dew point increases, the wind speed tends to remain relatively constant or decrease slightly

#Question 5
```{r}
met_avg[!is.na(elev_cat) & elev_cat != "clear"] |>
  ggplot() + 
  geom_bar(mapping = aes(x = elev_cat, fill = region), position = "dodge")+
  scale_fill_viridis_d()+
  labs(title="the weather stations by elevation category",
        x='elev_cat',
        y="number of weather stations",
       fill="Region")+
  theme_minimal()
```
#The graph reveals a clear preference for placing weather stations at medium elevations across all regions, particularly in the NW and SE regions.

#Question 6
```{r}
met_avg[!is.na(elev_cat) & elev_cat != "clear"] |>
  ggplot() + 
    stat_summary(mapping = aes(x = region, y = dew.point,fill="dew.point"),
      fun.data = "mean_sdl",
      geom ="bar")+
    stat_summary(mapping = aes(x = region, y = dew.point,fill="dew.point"),
      fun.data = "mean_sdl",
      geom ="errorbar")+
    stat_summary(mapping = aes(x = region, y = wind.sp,fill="wind.sp"),
      fun.data = "mean_sdl",
      geom ="bar")+
    stat_summary(mapping = aes(x = region, y = wind.sp,fill="wind.sp"),
      fun.data = "mean_sdl",
      geom ="errorbar")+
    labs(title="Mean Dew point and wind speed by region",
      x="Region",
      y="mean value")+
    theme_minimal()
```
#The SE region exhibits the highest humidity levels, while the NW region is notably drier and calmer. 

#Question 7
```{r}
pal <- colorNumeric(
  palette = "YlGnBu",
  domain = met_avg$rh,
)
rhmap <- leaflet(met_avg) |> 
  addProviderTiles('CartoDB.Positron') |> 
  addCircles(
    lat = ~lat, lng=~lon,
    label = ~paste0(round(rh,2), ' C'), color = ~pal(rh),
    opacity = 1, fillOpacity =1, radius = 5
    ) |>
  addLegend('bottomleft', pal=pal, values=met_avg$rh,
          title='Relative Humidity', opacity=1)
top_10 <- met_avg %>%
    arrange(desc(rh)) %>%
    slice(1:10)
rhmap <- rhmap %>%
    addMarkers(
        data = top_10,
        ~lon, ~lat,
        popup = ~paste(USAFID, "<br>RH:", rh, "%"),
        label = ~USAFID
    )
rhmap
```
# Moderate to High RH in the Central U.S, Lower RH in the Western U.S and Higher RH in the Eastern and Southeastern U.S.

#Question 8
```{r}
ggplot(met_avg, aes(x = wind.sp, y = rh, color = region)) + 
  geom_point(alpha = 0.6, size = 2) +
  labs(
    title = "Scatter Plot of Wind Speed vs. Relative Humidity by Region",
    x = "Wind Speed ",
    y = "Relative Humidity "
  ) +
  theme_minimal() +  
  theme(legend.position = "bottom") 
ggplot(met_avg, aes(x = wind.sp, color = region, fill = region)) + 
  geom_density(alpha = 0.5) +  # Add density with transparency
  labs(
    title = "Density Plot of Wind Speed by Region",
    x = "Wind Speed",
    y = "Density"
  ) +
  scale_y_continuous(expand = expansion(mult = c(0, 0.05)))
  theme_minimal() +
  theme(legend.position = "bottom")
```