library(plyr)
library(dplyr)
library(tidyr)
library(ggplot2)
library(knitr)

Reading in Data

Mis_Data<-read.csv("C:/Users/vivek/Documents/GlobalLandTemperatures/GlobalLandTemperaturesByState.csv", header = T)
Mis_Data %>%
  separate(col = dt, into = c("Year", "Month", "Day"), convert = TRUE) ->Mis_Data
Mis_Data<-na.omit(Mis_Data)


clim_data<-Mis_Data
kable(clim_data[1:10,1:7])
Year Month Day AverageTemperature AverageTemperatureUncertainty State Country
1855 5 1 25.544 1.171 Acre Brazil
1855 6 1 24.228 1.103 Acre Brazil
1855 7 1 24.371 1.044 Acre Brazil
1855 8 1 25.427 1.073 Acre Brazil
1855 9 1 25.675 1.014 Acre Brazil
1855 10 1 25.442 1.179 Acre Brazil
1855 11 1 25.400 1.064 Acre Brazil
1855 12 1 24.100 1.718 Acre Brazil
1856 1 1 25.814 1.159 Acre Brazil
1856 2 1 24.658 1.147 Acre Brazil 
#extracting Data for state of Missouri
clim_data %>%
  filter(Country=="United States")  %>%
  filter(State=="Missouri")->clim_datac

#omitting na values
  clim_datac<-na.omit(clim_datac)
kable(clim_datac[1:5,1:7])
Year Month Day AverageTemperature AverageTemperatureUncertainty State Country
1743 11 1 5.310 3.440 Missouri United States
1744 4 1 14.466 3.482 Missouri United States
1744 5 1 17.634 3.402 Missouri United States
1744 6 1 23.033 3.427 Missouri United States
1744 7 1 25.052 3.394 Missouri United States 
# Finding Average Temperatures
clim_datac %>%
  filter(Year>1742) %>%
  group_by(Year) %>%
  summarise(Temperature = mean(AverageTemperature)) ->clim_data2

Plotting the average temperature trend

qplot(Year, Temperature, data=clim_data2, main="Missouri Average Temperature \n 1744-Present",geom=c("point","smooth"))+ aes(colour = Temperature) + scale_color_gradient(low="blue", high="red")
## `geom_smooth()` using method = 'loess'

The above graph doesn’t capture the gravity of the situation. Let us dig deeper into the time lines.To see from which point in time the temperature rise started happening.

# Finding Average Temperatures years 1900- Present
clim_datac %>%
  filter(Year>1900) %>%
  group_by(Year) %>%
  summarise(Temperature = mean(AverageTemperature)) ->clim_data2
qplot(Year, Temperature, data=clim_data2, main="Missouri Average Temperature \n 1900-Present",geom=c("point","smooth"))+ aes(colour = Temperature) + scale_color_gradient(low="blue", high="red")
## `geom_smooth()` using method = 'loess'

This graph seems much better! We clearly see the trend in rising temperatures.

Finding Average Temperatures years 1950- Present

clim_datac %>%
  filter(Year>1950) %>%
  group_by(Year) %>%
  summarise(Temperature = mean(AverageTemperature)) ->clim_data2

This is more clear shows and shows a gradual rise in temperatures.

qplot(Year, Temperature, data=clim_data2, main="Missouri Average Temperature \n 1950-Present",geom=c("point","smooth"))+ aes(colour = Temperature) + scale_color_gradient(low="blue", high="red")
## `geom_smooth()` using method = 'loess'

Finding Average Temperatures years 2000- Present

clim_datac %>%
  filter(Year>2000) %>%
  group_by(Year) %>%
  summarise(Temperature = mean(AverageTemperature)) ->clim_data2
qplot(Year, Temperature, data=clim_data2, main="Missouri Average Temperature \n 2000-Present",geom=c("point","smooth"))+ aes(colour = Temperature) + scale_color_gradient(low="blue", high="red")
## `geom_smooth()` using method = 'loess'

This is the most revealing of all plots, highlighting how from year 2000 onwards, the temperatures have begun rising dangerously.

Conclusion: The data never lies! How can anybody deny global warming??