This project is an investigation into the monthly changes in precpitation for Belfast, Dublin Airport, University College Galway and Cork Airport. Precipitation was chosen due to its variability across the country. For this reason the sets of stations were divided between west and east coast stations.

library(dplyr)
## 
## Attaching package: 'dplyr'
## 
## The following objects are masked from 'package:stats':
## 
##     filter, lag
## 
## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union
library(gdata)
## gdata: read.xls support for 'XLS' (Excel 97-2004) files ENABLED.
## 
## gdata: read.xls support for 'XLSX' (Excel 2007+) files ENABLED.
## 
## Attaching package: 'gdata'
## 
## The following objects are masked from 'package:dplyr':
## 
##     combine, first, last
## 
## The following object is masked from 'package:stats':
## 
##     nobs
## 
## The following object is masked from 'package:utils':
## 
##     object.size
setwd("c://Users//Adam/Desktop")
setwd("GYC602A")
load("Rainfall.Rdata")
library(dygraphs)

Gathering the data for stations

For this assessment the data was extracted for Belfast, Dublin Airport, University College Galway and Cork airport. These stations offered various locations in the Irish context. The Galway and Cork station would offer a good index for the western side of Irelands precipitation levels while Belfast and Dublin would offer the eastern index. It is well known that the west of Ireland receives higher levels of rainfall than the east therefore there should be a good contrast between the stations.

rain %>%  group_by(Year,Month) %>% 
  summarise(Rainfall=sum(Rainfall)) %>% ungroup %>% transmute(Rainfall) %>% 
  ts(start=c(1850,1),freq=12) -> rain_ts


rain %>%  group_by(Year,Month) %>% filter(Station=="Dublin Airport") %>%
    summarise(Rainfall=sum(Rainfall)) %>% ungroup %>% transmute(Rainfall) %>%
    ts(start=c(1850,1),freq=12) ->  dub_ts
rain %>%  group_by(Year,Month) %>% filter(Station=="University College Galway") %>%
    summarise(Rainfall=sum(Rainfall)) %>% ungroup %>% transmute(Rainfall) %>%
    ts(start=c(1850,1),freq=12) ->  UCG_ts


rain %>%  group_by(Year,Month) %>% filter(Station=="Belfast") %>%
    summarise(Rainfall=sum(Rainfall)) %>% ungroup %>% transmute(Rainfall) %>%
    ts(start=c(1850,1),freq=12) ->  bel_ts

rain %>%  group_by(Year,Month) %>% filter(Station=="Cork Airport") %>%
    summarise(Rainfall=sum(Rainfall)) %>% ungroup %>% transmute(Rainfall) %>%
    ts(start=c(1850,1),freq=12) ->  cor_ts

These codes assembled the station data for this study. It filtered the stations of interest but also conducted a conversion to the rainfall data. For this study the sum of rainfall was chosen because it was decided to provide a better indication of rainfall levels rather than average. The data was also organised into a monthly basis in order to allow an investigation into monthly changes in precipitation.

Assembling the dygraphs

To begin a month plot for each station was established to observe the monthly variability of rainfall over the time period in question

monthplot(dub_ts,col='dodgerblue',col.base='indianred',lwd.base=3, xlab="Month", ylab="Precipitation (mm)", main = "Monthly Precipitation data  For Dublin Airport Station ")

monthplot(bel_ts,col='dodgerblue',col.base='indianred',lwd.base=3, xlab="Month", ylab=" Precipitation (mm)", main = "Monthly Precipitation for Belfast Station ")

monthplot(UCG_ts,col='dodgerblue',col.base='indianred',lwd.base=3, xlab="Month", ylab=" Precipitation (mm)", main = "Monthly Precipitation for Galway station ")

 monthplot(cor_ts,col='dodgerblue',col.base='indianred',lwd.base=3,xlab="Month", ylab=" Precipitation (mm)", main = "Monthly Precipitation for Cork Station ")

Due to the geographical contrasts between the four stations there will once again be an exploration of the monthly precipitation data in terms of west and east. The east of Ireland shows that the wettest season is in Autumn (SON). However, the wetterst month seems to show that December has the greatest amount of rainfall for both Dublin and Belfast station. The season which shows the least amount of rain seems to be spring (MAM). This is once again evident in both eastern stations. The western stations however show a slight difference. There are signals that Winter is the wettest(DJF) season for both Galway and Cork station. However, similar to the eastern stations December shows the highest levels of rainfall. The similarities between the eastern and western stations are once again evident when we discuss the season with the least amount of rainfall. Similarly to Belfast and Dublin, the driest season seems to be spring once again. The similarties between the two sets of stations shows that even though they are geographically, very different, the levels of rainfall that effect the west are also indicated by similar results on the east.

In order to construct an interactive graph for all our stations annual precipitation data will be used to investigate the differences between both sets of data.

dub_ts%>% dygraph(width=800,height=300,group="dub_bel_UCG_cor",main="Dublin Airport", xlab="Year", ylab="precipation mm")

 bel_ts  %>% dygraph(width=800,height=300,group="dub_bel_UCG_cor",main="Belfast",xlab="Year", ylab="precipation mm")

UCG_ts%>% dygraph(width=800,height=300,group="dub_bel_UCG_cor",main="University College Galway",xlab="Year", ylab="precipation mm")

cor_ts %>% dygraph(width=800,height=300,group="dub_bel_UCG_cor",main="Cork Airport",xlab="Year", ylab="precipation mm") %>% dyRangeSelector

Numerous ovbservations can be seen from this data set. In terms of the West east divide in precipitation there is a noticable contrast between the four stations. As seen in the above graphs there is a clear indication that the West of Ireland receives more rainfall than the east. The highest record of rainfall over all stations was recorded at Cork in December 1899. It is also clear from the above graphs that Dublin receives the lowest magnitude of precipitation with very few instances the location receiving over 200mm of rainfall. Deceber 1978 for both Dublin Airport and Belfast station was the highest record of precipitation. this suports the methodological approach of seperating the four stations into the respective geographical location (east/west).While the trends between Dublin and Belfast are quite similar there is not such an obvious trend between Cork and Galway. The variability of both stations differ greately when it comes to precipitation. the most obvious example of this was for the amount of precipitation received in December 1899. While cork received almost 500mm of precipitation this month Galway received almost half that amount. This is a clear indication that the pattern of distribution for the west of Ireland is more complex than the East. This could be due to topography features or the location of the station.

However, a major issue with time series data is the random noise component. This is viewd as anomalies of rainfall. This may in turn effect the sum of the months. By eliminating the random noise component there can be an assessment into the trend of the precipitation data for each respective station.

Dublin Airport

dub_ts <- ts(data = dub_ts, start = c(1850,1), end = c(2010,12), frequency = 12)

dub_ts  %>% window(c(1850,1),c(2010,12)) %>% 
  stl(s.window='periodic') %>% plot

dub_ts  %>% stl(s.window='periodic',t.window=361,) %>% plot

Removing the noise component there can be an argument made that precipitation has been increasing at various levels over the time series. Various peaks can be recorded along the series. However, from aproxamitely 1960 onwards there is an identifiable increase in precipitation.

Belfast

bel_ts <- ts(data = bel_ts, start = c(1850,1), end = c(2010,12), frequency = 12)

bel_ts  %>% window(c(1850,1),c(2010,12)) %>% 
  stl(s.window='periodic') %>% plot

bel_ts  %>% stl(s.window='periodic',t.window=361,) %>% plot

A similar trend can be seen for Belfast station. While the start of the 1800’s shows a similar peak to that of Dublin, as the trend reaches the 21st century the differences are evident. There could be a number of reasons for this differences.For one Belfast may be heavily influenced by the North Atlantic Oscillation. This in turn may result in slight differences in precipitation trends which can be observed in both trend diagrams.

Cork

cor_ts <- ts(data = cor_ts, start = c(1850,1), end = c(2010,12), frequency = 12)

cor_ts  %>% window(c(1850,1),c(2010,12)) %>% 
  stl(s.window='periodic') %>% plot

cor_ts  %>% stl(s.window='periodic',t.window=361,) %>% plot

Cork Airport station shows some similarities in precipitation trends. The early peak of the 1800s, which is evident in both Dublin and Belfast station is once again evident here. However, as time progresses, the trend differs. With precipitation trends seeming to show lower peaks than the eastern counterpart.

Galway

UCG_ts <- ts(data = UCG_ts, start = c(1850,1), end = c(2010,12), frequency = 12)

UCG_ts  %>% window(c(1850,1),c(2010,12)) %>% 
  stl(s.window='periodic') %>% plot

UCG_ts  %>% stl(s.window='periodic',t.window=361,) %>% plot

Galway is the final station that underwent trend analysis. Unlike the other three stations the peak of rainfall in the 1800s is not as intense at this station. There could be a number of reasons for this, such as the direction the air parcel came from which has an influence on precipitation intensity. The trend for Galway show various fluctuations but the general trend shows an increase in precipitation up until the twenty first century.

Conclusion:

This project investigated the precipitation alterations between four different weather stations located across Ireland. Due to geographical differences, there was a division of all stations into a respective east-west divide. For this investigation the sum of rainfall was used due to the belief that total rainfall shows a more intriguing picture than the mean values. This is due to the mean usually being biased to random noise. The monthly trends were extrated to investigate seasonal patterns of rainfall. An interactive time series was then constructed to see annual trends.However, the random noise component hide some trends that may have been evident. To combat this a number of graphs were included to remove the random noise that was evident. The trends indicate an increase in precipitation to be evident from the 1950s with all stations recording a large increase precipitation in the early 1850’s and then recording individual peaks and troughs.