---
title: "ANLY512 lab2"
author: Tian Zhu, Yumeng Du 
output: 
  flexdashboard::flex_dashboard:
    orientation: columns
    source_code: embed
    vertical_layout: fill
---

```{r setup, include=FALSE}
library(flexdashboard)
library(knitr)
library(ggplot2)
library(dplyr)
library(plotly)
library(rWBclimate)
library(ggplot2)
library(devtools)
library(shiny)
library(shinydashboard)
```



###Climate Change


Introduction
===================================== 

Climate change has been a huge environmental threat and likely one of the most important challenges of our time. Climate change may refer to a change in average weather conditions, or in the time variation of weather within the context of longer-term average conditions. To understand climate change, we used “rWBclimate” package to extract climate data from World Bank database to showcase the trend of global warming in Europe. 

Dashboard
===================================== 
-----------------------------------------------------------------------

### Annual Average Temperature in 20-year intervals

```{r}
gbr.dat.t <- get_ensemble_temp("GBR", "annualavg", 1800, 2200)
## Loading required package: rjson
### Subset to just the median percentile
gbr.dat.t <- subset(gbr.dat.t, gbr.dat.t$percentile == 50)
gbr.dat.t$fromYear=as.numeric(gbr.dat.t$fromYear)
gbr.dat.t$toYear=as.numeric(gbr.dat.t$toYear)
gbr.dat.t$data=as.numeric(gbr.dat.t$data)
## Plot and note the past is the same for each scenario
ggplot(gbr.dat.t, aes(x=fromYear,y=data,group=scenario,colour=scenario)) +
 geom_point() +
 geom_path() +
 theme_bw() +
 xlab("Year") +
 ylab("Annual Average Temperature in 20 year intervals")
```

Column {data-width=350}
-----------------------------------------------------------------------

### Annual Average precipitation in mm

```{r}
gbr.dat.p <- get_ensemble_precip("GBR", "annualavg", 1800, 2200)
gbr.dat.p <- subset(gbr.dat.p, gbr.dat.p$percentile == 50)
gbr.dat.p$fromYear=as.numeric(gbr.dat.p$fromYear)
gbr.dat.p$toYear=as.numeric(gbr.dat.p$toYear)
gbr.dat.p$data=as.numeric(gbr.dat.p$data)
ggplot(gbr.dat.p, aes(x = fromYear, y = data, group = scenario, colour = scenario)) +
    geom_point() + geom_path() + theme_bw() + xlab("Year") + ylab("Annual Average precipitation in mm")
```

### historical plot

```{r}

gbr.modelpast <- subset(gbr.dat.t, gbr.dat.t$scenario == "past")
gbr.historical <- get_historical_temp("GBR", "year")
### Plot create historical plot
hist.plot <- ggplot(gbr.historical, aes(x = year, y = data)) + geom_point() +
    geom_path()

### Create a centroid for the past
gbr.modelpast$centroid <- round((gbr.modelpast$fromYear + gbr.modelpast$toYear)/2)

### Create averages based the same windows used in the model output for
### comparison
win_avg <- function(from, to, df) {
    win <- subset(df, df$year >= from & df$year <= to)

    return(c(mean(win$data), round(mean(c(from, to)))))
}
hist.avg <- matrix(0, ncol = 2, nrow = 0)
for (i in 1:dim(gbr.modelpast)[1]) {
    hist.avg <- rbind(hist.avg, win_avg(gbr.modelpast$fromYear[i], gbr.modelpast$toYear[i],
        gbr.historical))
}
colnames(hist.avg) <- c("data", "centroid")

### Create new dataframe of historical averages and model averages
hist.comp <- rbind(hist.avg, cbind(gbr.modelpast$data, gbr.modelpast$centroid))
hist.comp <- as.data.frame(hist.comp)
hist.comp$Output <- c(rep("Historical", 4), rep("Model", 4))

### overlay the averages with the original raw data plot
hist.plot <- hist.plot + geom_point(data = hist.comp, aes(x = centroid, y = data,
    colour = Output, group = Output, size = 3)) + geom_path(data = hist.comp,
    aes(x = centroid, y = data, colour = Output, group = Output)) + guides(size = FALSE)

hist.plot + xlab("Year") + ylab("Annual average temperature in deg C") + theme_bw()
```

Questions about Climate Change
====================================

1.How accurate is the model in predicting average temperature in the future?

2.Is there a significant relationship between average temperature and average precipitation?

3.What is the projected average temperature in 200 years? Is the result reliable?