The Limerick Navigation (upper end) in flood November 2009

The Dublin Penny Journal (conducted by P Dixon Hardy, MRIA) of June 11, 1836 (Vol IV No 206) carried an extensive account of the first report of the commissioners appointed by the House of Commons to examine the River Shannon and its tributaries. The Journal reported, inter alia, that:

The river from Limerick to Killaloe, the second portion of the commissioners, is the most impracticable of the entire navigation. The fall of the Shannon, in its whole course, is 147 feet, of which 97 feet occur in this short space of fifteen miles.

Three watercourses

The natural course of the River Shannon is from Killaloe in County Clare (and its sister town, across the river, Ballina in County Tipperary) down past Parteen Villa to O’Briensbridge in County Clare (across from Montpelier in County Limerick) to Castleconnell and the Falls of Doonass. Then past the confluence with the Mulkear near Annacotty, on to Plassey, where the University of Limerick is, and then north-west in a large loop to Athlunkard and Corbally befoere coming south again into Limerick. It branches around an island; the eastern branch is called the Abbey River.

Sketch-map of the Limerick Navigations

In the eighteenth century, the Limerick Navigation was built to bypass the rapids at Killaloe, Castleconnell, Doonass, Plassey and Corbally. The navigation had five sections. Starting at Limerick, they are:

  • a one-mile canal from a lock at the tidal Abbey River, cutting off the Corbally loop and, after passing through a second lock, rejoining the Shannon at Arthur’s Ferry
  • a one-mile stretch of river up to Plassey, with the towing-path on the east side
  • five miles of canal from the west bank of the river at Plassey up to Errina, where it rejoins the river
  • a five-mile river stretch, with a towing-path on the west bank, up through O’Briensbridge
  • a two-mile canal, with three locks, through Killaloe.

The navigation was poorly designed and, despite many improvements over the years, continued to cause problems until it was bypassed, in the 1920s, by the new headrace and tailrace canals serving the hydroelectric power station at Ardnacrusha. That navigation too has its difficulties, and after it was opened the Grand Canal Company boycotted it for some time until certain improvements were made. The full story of both navigations requires extensive treatment and will have to wait until I have several days to spare. This page shows some of the effects of the floods of November 2009 on the upper reaches of the three watercourses: the river, the old Limerick Navigation and the headrace. The lower reaches will be covered elsewhere.

The flow of water

The Parliamentary Gazetteer of Ireland, adapted to the new poor-law, franchise, municipal and ecclesiastical arrangements, and compiled with a special reference to the lines of railroad and canal communication, as existing in 1844-45; illustrated by a series of maps, and other plates; and presenting the results, in detail, of the census of 1841, compared with that of 1831 (A Fullarton & Co, Dublin, London, and Edinburgh, 1846) quoted Robert Kane MD’s The Industrial Resources of Ireland (Hodges and Smith, Dublin, 1844):

The average difference between summer and winter level of the Shannon at Killaloe, where, narrowing from Lough Derg, it reassumes the river form is about six feet, but the total of the rises of the water during the year, are found from a discussion of the observations of three years, to be eleven feet.

The rising of the waters occupied in average 77 days: in falling to the summer level they occupied 107 days. The quantity of water thus accumulated in the great natural reservoir of the Lough was 532,554,096 cubic yards, or 403,416,600 tons which is discharged in 107 days at the rate of 155,926 tons per hour. By this, a force continuous day and night of 177 horse-power per foot of fall, may be produced. An equal force is of course available whilst the river is rising, and thus through 184 days or six months of the year, this enormous power is in action, independent of the ordinary discharge which goes on when the waters are at the lowest. When the river is high, the motive force available is far greater than that just now mentioned.

An example furnished to me by Mr. Mulvany will show this sufficiently. “On the 2d of December, 1836, when the water was 13 feet on the upper sill of Killaloe lock, the observed discharge was 882,450 cubic feet per minute, and on the 10th of that month, the height was 14 feet 1 inch, on the 18th, 14 feet 4 inches, at which height it continued until the 25th, with of course a greatly increased discharge, on the latter day it began to descend gradually. During the period mentioned, the whole lake rose four inches between the 3d and 4th, and five inches in two days, between the 5th and 7th, and two inches in other days. These grand rises, at that height of water extended over the flooded lands as well as the lake, that is, over a surface of from 36,000 to 38,000 statute acres.”

Now the discharge for the month of December 1836, may certainly, from the description above given, be taken at one million cubic feet per minute, that is, one and two-third million of tons of water per hour, capable of producing 1,885 horse-power per foot of fall. The minimum discharge of the Shannon at Killaloe has been estimated by Mr. Mulvany, in the driest summer, so low as 100,000 cubic feet of water per minute. This is equal to a force of 188 horse-power per foot of fall. At this minimum, however, the flow is kept but for a very short time, certainly not more than a month in the year, which is also the duration that may be allotted to the maximum elevation of the waters.

From those figures, Kane deduced that the 97 feet of fall from Killaloe to Limerick provided “a total of 33,950 horse-power in continuous action, day and night, throughout the year.” And Thomas McLaughlin drew on the work of Kane and others in persuading the government of the new Irish Free State to build, at Ardnacrusha, a power station that, in 1931, was capable of generating 96% the total amount of electricity used in the state.

Ardnacrusha power station from the headrace

The penstocks that feed the turbines at Ardnacrusha

Boats at the bottom of the lower chamber in Ardnacrusha lock, seen from the top of the upper chamber

But it is the volumes of water that are of most interest here. Ardnacrusha is capable of taking 400 cubic metres of water per second, and a cubic metre weighs about a tonne so that’s 1,440.000 tonnes an hour. The weir at Parteen Villa controls the amounts of water entering the headrace and the old course of the river: in normal conditions, the first 10 tons are sent down the river, through O’Briensbridge and Castleconnell, and the rest goes to Ardnacrusha. In what might be called normal floods, any excess is sent down the river. Bob Cullen, formerly of ESB Ardnacrusha, said that the average flow at Killaloe was 180 tons per second but that it varied from 10–15 tons to over 700 tons (Bob Cullen “Some Notable Features of the Design and Operational History of Ardnacrusha since 1929” in Bielenberg, Andy, ed The Shannon Scheme and the electrification of the Irish Free State: an inspirational milestone The Lilliput Press, Dublin, 2002).

The section of Parteen Villa weir that controls water going down the river

Parteen Villa weir blocks access by boats from upstream to the old navigation; the abandonment of the Plassey–Errina Canal blocks it from downstream. However, the stretch of river from the upstream end of Castleconnell, through O’Briensbridge and Montpelier, to below the weir still has a navigable depth of water: the minimum level is set by the “regulating weir” at World’s End, Castleconnell. As the upper end of the Plassey–Errina Canal joins the river in that section, the depth of the canal as far as the first (Errina) lock is also dependent on that weir, which was installed by contractors working for the Commissioners for the Improvement of the Navigation of the River Shannon in the 1840s.

The diversion of most of the water to the power station did not just result in closure of the old navigation: it had deleterious effects of the fisheries, especially the salmon fishery, and it removed much of the scenic effect of the Falls of Doonass.

November 2009

The severity of the floods of November 2009 is such that very large amounts of water have been sent down the original course of the river. On Friday 20 November  I rang the ESB, which controls the water levels on the Shannon, using Parteen Villa weir. I was told that they were sending 700 tons a second down the river, so much water that the level in the headrace was below normal.

The headrace from the bridge at Clonlara 20 November 2009
The headrace from the bridge at O’Briensbridge 22 November 2009

On 23 November, we were told that the ESB would be letting down more water; warnings were issued to people in O’Briensbridge, Montpelier, Castleconnell and areas of Limerick. The volume of water was such that the levels were at, or perhaps even above, those that would have been seen as normal before Ardnacrusha was built.


25 November 2009:

The lock 1

The lock 3

The lock 2


Here is the quay at World’s End at normal levels.

The quay at World’s End (March 2009)

And now.

The quay at World’s End (23 November 2009)

The weir in normal conditions.

Castleconnell regulating weir (March 2009)

And now.

Castleconnell regulating weir (23 November 2009)

A huge area was flooded, including Macnab’s Bog.

Floods at World’s End (23 November 2009)

The Plassey–Errina Canal

The regulating weir at the World’s End at Castleconnell maintains the minimum level in the upper stretch of the Plassey–Errina Canal, from the point at which it leaves the Shannon (just below O’Briensbridge) to the top lock at Errina. Just upstream of that lock is the bridge at Errina. Here is what I think is a slot for stop-planks in the stonework for the bridge.

The slot (September 2007)

The slot (23 November 2009)


These photos give an idea of the sorts of levels that the waterways infrastructure at O’Briensbridge was designed to cope with.

The twelve arches of O’Briensbridge (November 2006)

The twelve arches of O’Briensbridge (23 November 2009)

The towing-path arch (November 2006)

The towing-path arch (23 November 2009)

The navigation arch from upstream (26 November 2009)

The quays below the bridge (March 2007)

The quays below the bridge (23 November 2009)

The towing-path and quay from the bridge (26 November 2009)

The arch (February 2008)

The arch (23 November 2009)

The arch almost covered (26 November 2009)

The steps from below (February 2008)

The steps from above (23 November 2009)

The top step covered (26 November 2009)

The boatshed (February 2008)

The boatshed (23 November 2009)

Here is an article from 2012 about variations in winter conditions and here is another (h/t FT Alphaville).



15 responses to “The Limerick Navigation (upper end) in flood November 2009

  1. unbelievable

  2. Once they CAN control the release of water we might be ok but if the vol of water gets any greater we could be in very serious trouble. It’s raining again. I hope they can let down the water in the canal again tonight on low tide.

  3. Well done Brian – nice work.

  4. Thanks, Adrian. bjg

  5. Great site lads, keep up the picture taking.
    I’m gone from that area about 20years.
    I’m very sad to see the damage, RTE are a sham,
    they are not covering this at all

  6. There are good aerial photos on the Limerick Leader site . Only one lad involved here, though. bjg

  7. A very informative website

  8. Thanks, Brendan. bjg

  9. Very interesting. If the ESB were releasing 700 tons/sec down the river, then Parteen weir must have been at maximum capacity! One wonders what would have occurred if levels upstream had continued to rise??

  10. Alan: I confess that I don’t fully understand how the system works (I hope to ask the ESB for some information after Christmas). I don’t think Parteen Villa Weir itself would have been in any danger, as it is a weir rather than a dam and could simply have opened its sluice-gates somewhat more. But the areas downstream would have suffered rather more than they did, with more widespread flooding. I am conscious of the benefits of living near the top of a hill, but folk nearby were flooded out of their homes. bjg

  11. bjg:

    as far as I’m aware, the total capacity of all gates at Parteen is 700 m3 per sec to river channel, and 400 to the headrace. This is supposedly more than can flow through Killaloe, hence the level in the flooded section (and the headrace canal ) would have started to drop as the release increased toward maximum. I read this once upon a time in some ESB literature; can’t find it now. But based on reported observations of water overflowing the embankments near Parteen weir, it sounds as if something went wrong. I’ve no doubt the weir could take any level, but I’m assuming the water would start overflowing all the way up the embankments parallel to the Kilmastulla river, which would’ve been quite a sight!!

  12. hi …livin in london …from montpelier,(best left peg bridge celtic ever had)…love all flood pics ………..well done….

  13. Thanks, Sean. bjg

  14. Pingback: Draining Lough Derg | Irish waterways history

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.