January Median Rainfall Analysis in Ireland (1850-2014)
Published: 2026-01-13
Format: html
Self-contained-math: true
Self-contained: true
Introduction
Positioned on the eastern edge of the North Atlantic, Ireland is located between latitude 51°N and 56°N, longitude 5°W and 11°W, spanning an area of 84 000 km2. In Ireland, annual rainfall figures can be observed to vary, characterized by low-intensity and long-duration events. The totals often reach over 3000 mm in hills and mountain areas in the west; in contrast, the Midlands generally receives around 800 mm and sheltered areas in the east receive over 600mm (Met Éireann, 2024a; Mateus and Coonan, 2023).
This blog shows the creation of a Rainfall map of Ireland using data from 25 weather stations across the country. The analysis will focus on the precipitation time series from 1850 to 2014 to determine the median rainfall level in January at each station.
The blog also provides an explanation of the data utilized and the code developed to perform the analysis for the creation of the map, and a brief discussion of the pattern identification.
The availability of a rainfall time series in Ireland helps us understand changes over time and identify emerging trends, while also identifying effects in sectors such as agriculture, transport, urban planning, and water resource management. (Noone et al., 2016).
Rainfall Data of Ireland
For the creation of the map, two datasets downloaded from the Moodle Maynooth University Website, named ‘Rain’ and ‘Station’, were used. These datasets are based on the homogenized precipitation network for the Island of Ireland developed by Noone et al. (2016) and cover a period of 164 years, from 1850 to 2014.
The ‘Rain dataset’ provided information on monthly total precipitation in millimetres (mm), with readings of zero precipitation detected in different months and years in stations like Markree Castle, Rathdrum, Valentia, Killarney and Waterford. As well as a maximum of 460.50 mm detected in Cork Airport Station. The calculated median was 82.20 mm, while the mean was 90.97 mm. This indicates that the precipitation distribution is slightly skewed to the right. This means the values of most of the readings are lower than the mean.
Subsequently, this blog presents the medians obtained from the analysis of data for each station, focusing only on the month of January from the entire time series.
Methodology and R Analysis
To have a better understanding of the utilized data, the first step was setting the work directory ‘setwd’ and loading the data from ‘rainfall.Rdata’. Subsequently, a data exploration was performed using the commands ‘head’and ’tail’, applying them to both datasets ‘rain’ and ‘stations’, and the ‘summary(rain$Rainfall)’ command to the ‘rain’ dataset to obtain the minimum and maximum precipitations, and the calculated median and mean.To carry out the development of the interactive map and the data processing, several libraries were used in the R environment.
## Year Month Rainfall Station
## 1 1850 Jan 169.0 Ardara
## 2 1851 Jan 236.4 Ardara
## 3 1852 Jan 249.7 Ardara
## 4 1853 Jan 209.1 Ardara
## 5 1854 Jan 188.5 Ardara
## 6 1855 Jan 32.3 Ardara## Year Month Rainfall Station
## 49495 2009 Dec 99.90000 Waterford
## 49496 2010 Dec 70.20000 Waterford
## 49497 2011 Dec 80.67308 Waterford
## 49498 2012 Dec 113.84615 Waterford
## 49499 2013 Dec 136.15385 Waterford
## 49500 2014 Dec 28.75000 Waterford## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.00 53.60 82.20 90.97 118.10 460.50## Station Elevation Easting Northing Lat Long County Abbreviation
## 1 Athboy 87 270400 261700 53.60 -6.93 Meath AB
## 2 Foulksmills 71 284100 118400 52.30 -6.77 Wexford F
## 3 Mullingar 112 241780 247765 53.47 -7.37 Westmeath M
## 4 Portlaw 8 246600 115200 52.28 -7.31 Waterford P
## 5 Rathdrum 131 319700 186000 52.91 -6.22 Wicklow RD
## 6 Strokestown 49 194500 279100 53.75 -8.10 Roscommon S
## Source
## 1 Met Eireann
## 2 Met Eireann
## 3 Met Eireann
## 4 Met Eireann
## 5 Met Eireann
## 6 Met Eireann## Station Elevation Easting Northing Lat Long County
## 20 Markree Castle 49 169794.04 320195.93 54.12 -8.46 Sligo
## 21 Phoenix Park 39 311871.41 236856.22 53.37 -6.31 Dublin
## 22 Roches Point 40 182877.68 60742.10 51.79 -8.24 Cork
## 23 Shannon Airport 6 137792.00 159203.87 52.68 -8.92 Clare
## 24 Valentia 11 46471.78 78165.16 51.93 -10.23 Kerry
## 25 Waterford 42 261478.55 116752.28 52.29 -7.09 Waterford
## Abbreviation Source
## 20 MK Tabony
## 21 PP Tabony
## 22 RP CRU Met Eireann
## 23 SA Tabony
## 24 V Tabony
## 25 W TabonyTo carry out the development of the interactive map and the data processing, several libraries were used to set the R environment.
‘Tidyverse’ was used for efficient data manipulation, ‘leaflet’ and ‘tmap’ were used for creating interactive maps, ‘sf’ allowed spatial data coordination, while ‘leaflpop’ and ‘leaflet.extras’ provided enhanced features.
The command, which is shown below, took the precipitation data from the ‘Rain’ dataset. This information was then organized chronologically by year and month, covering the period from 1850 to 2014. Subsequently, a cleaning process was carried out, extracting the months of January from the data set and saving it into a new object called ‘rain_jan’.
The ‘rbind’ command was used to combine the top (head) and bottom (tail) records of ‘rain_jan’ dataset. Then using the ’knitr’ library and ‘kable’ funtion, the data was transform from raw computer output into ‘Processed January Rainfall Data’ table 1
| Year | Month | Rainfall (mm) | Station |
|---|---|---|---|
| 1850 | Jan | 169.0 | Ardara |
| 1850 | Jan | 96.9 | Derry |
| 1850 | Jan | 107.6 | Malin Head |
| 1850 | Jan | 92.5 | Armagh |
| 1850 | Jan | 115.7 | Belfast |
| 2014 | Jan | 288.1 | Valentia |
| 2014 | Jan | 200.4 | Cork Airport |
| 2014 | Jan | 259.5 | Killarney |
| 2014 | Jan | 174.2 | Roches Point |
| 2014 | Jan | 124.4 | Waterford |
With the aim of facilitating the observation of January data from each station. A custom function was created, named ‘local_jan_plot’. This involved using ‘mutate’ to generate a list of filenames of each station and converting the data time series using ‘ts’.
Simultaneously, a discontinuous red line was incorporated to show the median and a solid green line using ‘Lm’ to observe the trend over time. Then the process was automated using ‘png()’ and ‘dev.off()’ commands to efficiently save individual graphs in the set work directory, and the same process was reproduced for the 25 stations.
The next stage of the analysis was to obtain the median data and the geographical position. This was done by first taking all the January data from 1850 to 2014 for each station using ‘group_by’ and ‘summarize’ commands to calculate the median. Subsequently, the command ’leaft_join’was using to merge the station and geographical data to facilitate map referencing.
The final data used included the station, latitude, longitude, and the calculated median. Then ‘st_as_sf’ and ‘crs 4326’ were used to define the WGS84 coordinate system to transform this table into a spatial object ready for mapping.
The final stage involved the creation of the high-quality map using ‘leaflet’ and ‘tmap’ libraries. Here, the colour symbology for every station was defined using ‘colorBin’ command with Yellow-Green-Blue (YlGnBu) scheme, based on the calculated January median. This facilitated the identification of every station by precipitation intervals from 60mm to 200mm.
The code also established that when clicking a station on the Map ‘leafpop’ library would trigger a pop-up, showing the January precipitation data over time of that specific station, along with a green line to visualize the observed trend and a red dashed line to indicate the historical median.
Additionally, the pop-up displays the median data for each station. Similarly, other features were integrated on the map using the ‘leaflet.extras’ library including ‘addScaleBar, ’addMiniMap’ ‘and addMeasure’ tool, with the aim of facilitating the calculation of distances between stations.
Results and Discussion
According to the analysis of all Januarys from 1850 to 2014 time series, and calculated medians from all stations, results consistently show that stations located in the west of Ireland and closer to the Atlantic Ocean tend to have a greater precipitation median. This is the case for Killarney (177.7 mm) and Valencia (166 mm), located in County Kerry, and Ardara (171.6 mm) in County Donegal. Therefore, geographic location, specifically proximate to the west of Ireland and the prevailing winds controlled by the North Atlantic Oscillation pressure system, significantly influences precipitation in areas where westerly winds reach the coast and western areas collide with land and mountains, causing orographic lifting (Gleeson et al., 2017; Met Éireann, 2024b). However, by the time winds reach areas like Dublin in the east, the amount of rainfall is much slower; this is due to the rain shadow effect, which is created by physiographic units. (Gleeson et al., 2017; Met Éireann, 2024b).
Table 2: January Rainfall Summary by Station (1850-2014).
| # | Station | Median Rainfall (mm) | Trend | Max Rainfall (mm) | Min Rainfall (mm) |
|---|---|---|---|---|---|
| 1 | Killarney | 177.7 | -0.0182 | 367.00 | 20.9 |
| 2 | Ardara | 171.6 | 0.1667 | 352.40 | 23.1 |
| 3 | Valentia | 166.0 | 0.0745 | 347.60 | 0.0 |
| 4 | Cappoquinn | 147.4 | -0.0817 | 342.00 | 17.3 |
| 5 | Cork Airport | 134.6 | -0.0953 | 359.50 | 9.9 |
| 6 | Rathdrum | 126.6 | 0.0797 | 328.20 | 19.6 |
| 7 | Enniscorthy | 126.4 | 0.0597 | 302.70 | 12.9 |
| 8 | University College Galway | 124.1 | 0.0829 | 232.60 | 9.2 |
| 9 | Portlaw | 123.3 | 0.0182 | 317.00 | 11.7 |
| 10 | Markree Castle | 116.0 | 0.1528 | 232.10 | 17.1 |
| 11 | Roches Point | 111.0 | -0.0083 | 255.60 | 19.0 |
| 12 | Malin Head | 106.9 | 0.1335 | 211.20 | 22.2 |
| 13 | Foulksmills | 105.7 | -0.0191 | 251.40 | 13.1 |
| 14 | Waterford | 102.6 | -0.0014 | 256.90 | 2.3 |
| 15 | Strokestown | 102.5 | 0.1247 | 214.50 | 11.5 |
| 16 | Belfast | 102.1 | 0.0510 | 229.40 | 16.1 |
| 17 | Drumsna | 99.1 | 0.1359 | 206.41 | 23.1 |
| 18 | Derry | 97.3 | 0.1088 | 192.95 | 16.3 |
| 19 | Shannon Airport | 92.9 | 0.0855 | 221.00 | 6.2 |
| 20 | Athboy | 87.1 | 0.0695 | 159.30 | 12.2 |
| 21 | Mullingar | 80.6 | 0.0545 | 147.10 | 6.3 |
| 22 | Birr | 77.5 | 0.0322 | 186.20 | 9.3 |
| 23 | Armagh | 75.0 | 0.0395 | 170.80 | 9.5 |
| 24 | Phoenix Park | 67.6 | 0.0261 | 183.30 | 8.8 |
| 25 | Dublin Airport | 63.0 | 0.0440 | 171.10 | 8.2 |
Additionally, as observed in the graphs, most stations show positive trends, such as Ardara with a marked increase. Other areas, such as Cork Airport, located further south, exhibit slight decreases. This difference is particularly interesting as Valentia, which typically has higher median precipitation, recorded zero mm in 1924. This finding is significant, raising questions about whether it is the result of potential Climate Change Impacts or attributed to technical factors such as instrument manipulation or equipment changes, which are common challenges in long-term historical records (Noone et al., 2016).
Conclusion
This analysis demonstrates that the creation of high-quality maps based on long-term time series and precise data is crucial to understanding climatic behaviour and its precipitation patterns. The ability to identify trends over time allows for the attribution of climate change or the predominance of positive or negative NAO (Gleeson et al., 2017), and also technical factors like human errors or equipment changes. This information is crucial to the development of better and more accurate risk assessment, especially in vulnerable areas where the implementation of preventive and action plans is essential to protecting communities and infrastructure (Mateus and Coonan, 2023).
Furthermore, the analysis of detected trends and their attribution to natural, human or technical factors is important to validate the accuracy of climate models and identify the impacts of climate change over time and predict the increase or decrease of rainfall. Not only for a better management of water resources, but also to define more precise construction standards based on the weather characteristics of a specific area that can face the future weather conditions and develop more effective technologies that help to reduce the carbon footprint (Mateus and Coonan, 2023; Carlier et al., 2021).
According to information obtained from the trend lines for each, a prevalence of positive trends is observed; this means that, in general, Ireland has been getting wetter over the past 164 years; however, it is important to validate this information through detection and attribution analysis of long-term time series rainfall data.
References
Gleeson, E., Gallagher, S., Clancy, C. and Dias, F. (2017) ‘NAO and extreme ocean states in the Northeast Atlantic Ocean’, Advances in Science and Research, 14, pp. 23–33. doi: 10.5194/asr-14-23-2017.
Mateus, C. and Coonan, B. (2023) Estimation of Point Rainfall Frequencies in Ireland. Technical Note No. 68. Dublin: Met Éireann. Available at: https://www.met.ie/cms/assets/uploads/2023/03/Mateus-and-Coonan_2023_Estimation-of-point-rainfall-frequencies-in-Ireland_FINAL.pdf.
Maynooth University (2024) Rainfall data: Rain and Station (1850-2014) [datasets]. Available at: https://moodle.maynoothuniversity.ie/mod/folder/view.php?id=921154 (Accessed: 7 January 2026).
Met Éireann (2024a) Rainfall. Available at: https://www.met.ie/climate/what-we-measure/rainfall (Accessed: 7 January 2026).
Met Éireann (2024b) What are teleconnections and how do they influence Ireland’s weather? Available at: https://www.met.ie/what-are-teleconnections-and-how-do-they-influence-irelands-weather (Accessed: 7 January 2026).
Noone, S., Murphy, C., Coll, J., Matthews, T., Mullan, D., Wilby, R. L. and Walsh, S. (2016) ‘Homogenization and analysis of an expanded long-term monthly rainfall network for the Island of Ireland (1850–2010)’, International Journal of Climatology, 36(8), pp. 2837–2853. doi: 10.1002/joc.4522.