I was reading Alasdair Rae’s street types blog post yesterday and I thought about doing something similar for Ireland. I didn’t have much time today so I said I’d put something quick together for Cork City. Alasdair was lucky that he could use the amazing open data from the UK’s Ordnance Survey. For Ireland, we don’t have anything similar from our national mapping agency but fear not because it’s OpenStreetMap to the rescue. In my humble opinion individuals often spend far too much time trying to get a usable output from overpass-turbo when there’s an easier way. What I did was download the .osm.pbf file for Ireland from Geofabrik. Windows Subsystem for Linux has been a lifesaver since it came along, most of the heavy lifting I can now do through it. The steps I followed are as follows:

1. Download the .osm.pbf file using wget.

2. Install the osmium-tool

 

 

 

 

3.Filter out just the ‘highways’ from the dataset and save it to a new .osm.pbf file

 

 

 

 

4. Convert the .osm.pbf file to a new shapefile ( I don’t normally use shapefile but converting to a geopackage was causing some issues in this instance).

 

 

 

 

 

5. Reproject the shapefile to EPSG:2157 (Irish Transverse Mercator).

 

 

 

 

 

 

I could have made this more efficient again by combining some of these actions (such as save as a new shapefile and reprojecting), I did it this way to make the process easier to explain.

Next I brought the dataset into QGIS and extracted an area just for Cork City. The main issue I then has was how to ascertain what road types occur most frequently. I exported the Cork City extract as a CSV and imported it into Excel. I then used a generic formula from the folks at exceljet to extract the last work from the ‘name’ column. I then counted the frequency of the extracted work in a pivot table to get an idea of the types I’d like to use.

 

 

 

 

 

 

The last thing I did was to create a new field using an expression to find the various road types in the ‘name’ column from the OSM data. The expression I used is the below:

CASE
	WHEN "name" LIKE '%Road%' then 1
	WHEN "name" LIKE '%Park%' then 2
	WHEN "name" LIKE '%Street%' then 3
	WHEN "name" LIKE '%Avenue%' then 4
	WHEN "name" LIKE '%Court%' then 5
	WHEN "name" LIKE '%Drive%' then 6
	WHEN "name" LIKE '%Hill%' then 7
	WHEN "name" LIKE '%Lawn%' then 8
	WHEN "name" LIKE '%Estate%' then 9
	WHEN "name" LIKE '%Place%' then 10
	WHEN "name" LIKE '%Lane%' then 11
	ELSE 0
END

I then symbolised the data and exported it. The final product is below. Obviously this was a pretty quick and dirty way to do it and if I create an atlas for every major town in Ireland I will use PostGIS which will allow me to count the frequencies and create the new field with ease.

A topic that I think about frequently is the physical changes Ireland has gone through over the last 10 or 20 years. I’ve covered it before in this blog but I’m going to look at it again today. I think you could ask anybody around the country where the development in their town has been during the Celtic Tiger and this current boom period (well, up until the Covid-19 outbreak anyway) and they would easily be able to tell tell you.

It’s true that we have national indicators from the Central Statistics Office identifying where the most planning permissions have been granted as well as data from the Department of Housing, Planning and Local Government on new ESB connections to show where new homes have been constructed. I was looking for something that would give an idea of the physical development that has occurred in the last 20 years.

Enter the CORINE Land Cover Inventory. It started in 1985 as a way for the European Environment Agency (based in Copenhagen) to monitor the main land uses in the Union. As a little factoid, it stands for Coordination of Information on the Environment. It contains 44 land use classes and the cell size is 100 metres. It already contains layers that show the change between different years but it didn’t contain any change layer from 2000 to 2018. I wasn’t interested in the 44 classes, in fact I was only interested in the categories shown below. I decided to omit construction sites, dump sites and a few other inconsequential categories. These are the main categories that will show development in Ireland between those years.

111 – Continuous urban fabric
112 – Discontinuous urban fabric
121 – Industrial or commercial units
122 – Road and rail networks and associated land
123 – Port areas
124 – Airports

I used the the wonderful r.reclass GRASS tool in QGIS in order to reclassify the rasters for both the year 2000 and the year 2018 to combine these categories. I then used raster algebra in order to subtract the year 2000 from the year 2018 raster in order to show the actual change. Below is a GIF showing the process for an area of North Kerry and Cork.

I create the below map in A1 so you can zoom in to get a good sense of what 18 years of development looks like in a particular county.

Here is a link to the PDF version if you would rather.

*Update*

I redid the map showing the development up to the year 2000 as a separate colour. I’ve also added a legend.

Here is a link to the PDF of this version.

Finally, after a few unsuccessful attempts at converting the final raster to vector I used the gdal_polygonize tool which did it seamlessly without any loss of the smaller cells. Below is the final layer as a web map that you can make full screen.

There are a few interesting area that were developed in those 18 years such as Aughinish Alumina in Co. Limerick:

There are also a few windfarms such as this one in Tyrone:

Covid-19 Cases Ireland – 26 March 2020.

Updated map for the 25th of March 2020.

I’ve been creating daily maps of Covid-19 cases in Ireland using the official Government source. I think proportional symbols and proper labelling lead to the best outcome.