I am coming to that. The Leas-Chathaoirleach is getting ahead of me. For the summer of 2024, there were 21 days with about 200,000 minutes of delay. In July, there were three days where the delays went to more than 350,000 minutes. Members can see that it has been a challenging summer. Anybody who has spent time in airports in Europe can identify with that.
Moving to the next slide, the main figure relates to averages. From June to August, there was an increase of about 4.8% on 2023. That is about 2.5% on 2019, which is a figure that the airlines refer to. Looking at the map, the problem is that the way in which the traffic is presenting is very different, partly because of the war and partly because of the ban. The darker area in red is where the congestion has started to arise. States like Austria, Bulgaria, Croatia, Georgia, Hungary, Serbia and Montenegro are all up by 15% on 2019. Italy and Turkey are up by about 10%. The profile of the traffic is changing. June to August en route delays were 4.6 minutes on average, which is up 52% on 2023 and 59% on 2019.
The arrival punctuality has dropped drastically to 65%. It was 66% last year, but in 2019 it was 73%. Typically, therefore, we would be looking at a maximum number of about 36,000 flights daily, which really indicates this aspect is part of the context.
The other complication starting to arise is the impact of climate change. We are now starting to see a much bigger impact from weather. Especially in Austria, Croatia and all around that area, thunderstorm activity is also starting to impact the capacity of the sector. This is where we are really starting to see a mixture of things beginning to affect the network in a strong way. This gives the committee a flavour of what is happening.
The other thing we see in the Balkans is that the area is getting much more traffic, especially when overflights are going to the Far East. They are avoiding Ukraine by coming down through the Balkans, into Turkey and out towards the Far East. These flights are another piece of the profile because they cannot go through Russia any more and this has impacted the situation. To be fair, the airlines are hurting because of it. It is directly impacting their operations and passengers are increasingly frustrated as a result.
The information on the next slide will address the question asked about the average. I have taken the numbers from the late 1990s up to today. The members can see that previously the worst year we had was 1999, and the average delay then was 5.9 minutes. It will also be seen that the dark blue area on the image represents air traffic control capacity, while the tiny little green bit denotes weather. Moving forward to 2024, we can see the average delay is now 4.6 minutes, which is the worst we have had for several years. If we look at the split between the two instances, however, we can see that weather has had a much bigger impact in this regard too. In 1999, we took some radical steps with the creation of the central flow management unit and the reduction of vertical separation to try to increase capacity. In that year, there were 20,000 flights daily. We do, therefore, have more efficiency in the system now, but it is not enough. Later, I will talk about what we need to do to try to remedy this problem.
Turning to the next slide, this shows the evolution in the context of 2023. There has been much discussion in the press about this aspect. If we look at the 2023 information, the red bit in March and April of the previous year represents French air traffic control strikes. What has happened since then is indicated by the green piece and this represents weather having started to impact in 2023. By looking at the chart depicting the situation in June, July and August, we can see how much the impact has increased on an average basis. It does not paint the whole picture, but what we are starting to see is this interplay between weather and ATC capacity.