WEBVTT Kind: captions Language: en 1 00:00:00.230 --> 00:00:02.760 - If the Thames Barrier were to fail when it was needed, 2 00:00:02.760 --> 00:00:05.350 the damage would be almost impossible to calculate. 3 00:00:05.350 --> 00:00:08.520 The best estimates I've found say that the financial cost of London flooding 4 00:00:08.520 --> 00:00:12.730 would be somewhere between £20 billion and £100 billion. 5 00:00:12.730 --> 00:00:13.840 The Houses of Parliament, 6 00:00:13.840 --> 00:00:16.000 every skyscraper in the city of London, 7 00:00:16.000 --> 00:00:19.040 every building near the river would be unusable for weeks. 8 00:00:19.040 --> 00:00:20.660 The central parts of the London Underground 9 00:00:20.660 --> 00:00:23.800 would be out of action for months, possibly more. 10 00:00:23.800 --> 00:00:26.010 That's before you get to the human costs. 11 00:00:26.010 --> 00:00:29.350 Worst case, perhaps hundreds of thousands of people 12 00:00:29.350 --> 00:00:31.430 who'd be forced out of their homes. 13 00:00:31.430 --> 00:00:35.120 Which is why this barrier is designed to not fail. 14 00:00:35.120 --> 00:00:37.660 To be clear, that still means it can have occasional glitches 15 00:00:37.660 --> 00:00:40.960 or scheduled maintenance, or even have parts of it break entirely. 16 00:00:40.960 --> 00:00:45.650 But if any of those things happen, it must still protect London. 17 00:00:45.650 --> 00:00:48.350 And there's a whole team who work here full time 18 00:00:48.350 --> 00:00:50.240 to make sure that it does. 19 00:00:50.240 --> 00:00:54.270 - We aim to move each of the ten individual gates 20 00:00:54.270 --> 00:00:57.140 at least twice a month to exercise them. 21 00:00:57.140 --> 00:00:59.940 Today is one of our monthly test closures. 22 00:00:59.940 --> 00:01:04.210 For test closures, we close just as the tide is finishing going out. 23 00:01:04.210 --> 00:01:07.410 And that way, we can wait until the tide turns 24 00:01:07.410 --> 00:01:10.300 and then we get equilibrium on both sides of the Barrier, 25 00:01:10.300 --> 00:01:11.990 and therefore, we can reopen. 26 00:01:11.990 --> 00:01:15.600 That avoids us having to be closed for the full tidal cycle. 27 00:01:15.600 --> 00:01:16.510 - A couple of months ago, 28 00:01:16.510 --> 00:01:19.160 the Barrier had a full high-tide test closure 29 00:01:19.160 --> 00:01:20.260 and it is dramatic. 30 00:01:20.260 --> 00:01:21.735 Normally, the gates just sit out of the way 31 00:01:21.735 --> 00:01:22.568 on the bottom of the river, 32 00:01:22.568 --> 00:01:25.780 but when they're needed, they're rotated up into place. 33 00:01:25.780 --> 00:01:27.520 - Well, for the annual test closure, 34 00:01:27.520 --> 00:01:29.900 because we're trying to maximise the differential, 35 00:01:29.900 --> 00:01:32.510 the water levels on each side of the barrier, 36 00:01:32.510 --> 00:01:35.100 we like to close early in the tidal cycle 37 00:01:35.100 --> 00:01:38.060 to allow that differential level to build up 38 00:01:38.060 --> 00:01:40.940 to a maximum of 4½ metres. 39 00:01:40.940 --> 00:01:42.340 On a flood defence closure, 40 00:01:42.340 --> 00:01:45.070 we close much later in the tidal cycle, 41 00:01:45.070 --> 00:01:48.074 because we just want to stop the danger element of the tide 42 00:01:48.074 --> 00:01:49.990 going up into Central London. 43 00:01:49.990 --> 00:01:51.980 We try and be gentle with the river. 44 00:01:51.980 --> 00:01:55.610 So, we start by closing the outer smaller gates. 45 00:01:55.610 --> 00:01:58.270 And then, we slowly work across into the middle. 46 00:01:58.270 --> 00:02:01.300 And sometimes, we'll bring the two 61-metre gates 47 00:02:01.300 --> 00:02:05.020 in the centre of the river up a couple of metres below the surface, 48 00:02:05.020 --> 00:02:07.800 before we finally bring them into the closed position. 49 00:02:07.800 --> 00:02:10.780 And that avoids a reflective wave of the water 50 00:02:10.780 --> 00:02:13.234 that's coming in and being stopped, 51 00:02:13.234 --> 00:02:15.460 being pushed back downstream. 52 00:02:15.460 --> 00:02:17.200 So, during a flood defence closure, 53 00:02:17.200 --> 00:02:20.530 we don't have to stop the whole tide going into Central London. 54 00:02:20.530 --> 00:02:22.930 We just need to stop the part of that tide 55 00:02:22.930 --> 00:02:25.580 that would overtop the linear defences. 56 00:02:25.580 --> 00:02:27.870 Normal spring tides that go into London each day 57 00:02:27.870 --> 00:02:31.210 do not overtop the embankment walls, for example. 58 00:02:31.210 --> 00:02:33.800 But if you've got a surge element on top of that, 59 00:02:33.800 --> 00:02:36.790 that is what would go over the top of the walls in London, 60 00:02:36.790 --> 00:02:39.650 and that's the element that we need to stop with the Barrier. 61 00:02:39.650 --> 00:02:42.280 Once we've stopped the danger element of the tide 62 00:02:42.280 --> 00:02:43.580 and got a differential, 63 00:02:43.580 --> 00:02:47.590 we've effectively created a reservoir upstream of the Barrier. 64 00:02:47.590 --> 00:02:50.440 We can then over-rotate the large gates 65 00:02:50.440 --> 00:02:53.220 and allow a controlled flow of water upstream 66 00:02:53.220 --> 00:02:56.170 into the reservoir that we've created. 67 00:02:56.170 --> 00:02:58.340 And again, that reduces the amount of time 68 00:02:58.340 --> 00:03:02.600 before we get equilibrium on both sides of the barrier and can reopen. 69 00:03:02.600 --> 00:03:05.050 - The barrier cannot be allowed to fail when it's needed. 70 00:03:05.050 --> 00:03:08.210 So, each gate can be moved by one of four separate motors. 71 00:03:08.210 --> 00:03:10.490 That service tunnel I was in? There are two of them, 72 00:03:10.490 --> 00:03:12.250 a few metres apart, identical, 73 00:03:12.250 --> 00:03:15.220 with completely independent power and control systems. 74 00:03:15.220 --> 00:03:18.120 The barrier has three connections to the UK power grid, 75 00:03:18.120 --> 00:03:20.060 for both the north and south of the river. 76 00:03:20.060 --> 00:03:22.660 And if the entire British national power grid fails, 77 00:03:22.660 --> 00:03:24.470 which has never happened, but if it does, 78 00:03:24.470 --> 00:03:27.500 the barrier has three separate diesel generators, 79 00:03:27.500 --> 00:03:29.910 any one of which could operate the entire thing. 80 00:03:29.910 --> 00:03:31.950 - We have a number of modes of operation, 81 00:03:31.950 --> 00:03:34.520 so that if, when we're using Mode 1 operation, 82 00:03:34.520 --> 00:03:38.390 we have a problem of a technical nature, we go to Mode 2. 83 00:03:38.390 --> 00:03:40.360 Problem with Mode 2, we go to Mode 3. 84 00:03:40.360 --> 00:03:43.890 We don't necessarily have time to sort out any technical issues, 85 00:03:43.890 --> 00:03:46.800 but we've got other modes of operation that we can fall back on. 86 00:03:46.800 --> 00:03:49.790 The barrier has to close when we need it to. 87 00:03:49.790 --> 00:03:51.800 I always say that if the Prime Minister 88 00:03:51.800 --> 00:03:53.960 was stood in the House of Commons in his wellingtons 89 00:03:53.960 --> 00:03:57.310 and we said, "Sorry, PM, we'll get it right next tide," 90 00:03:57.310 --> 00:03:59.870 that would probably be fairly career-limiting(!) 91 00:03:59.870 --> 00:04:01.880 - So, that's the technical resilience. 92 00:04:01.880 --> 00:04:03.750 What about the physical construction? 93 00:04:03.750 --> 00:04:06.330 Well, these piers are made of Portland cement, 94 00:04:06.330 --> 00:04:08.440 so, they'll stand for a couple of centuries at least. 95 00:04:08.440 --> 00:04:12.450 The steel gates are protected by what's called a sacrificial anode. 96 00:04:12.450 --> 00:04:15.400 Corrosion in water like this is an electrochemical process. 97 00:04:15.400 --> 00:04:16.650 So, if you connect a chunk of metal 98 00:04:16.650 --> 00:04:19.170 that's more easily corroded to something like the gate, 99 00:04:19.170 --> 00:04:22.150 that chunk of metal will take the corrosion damage instead, 100 00:04:22.150 --> 00:04:24.570 and you can replace it fairly easily whenever you need to. 101 00:04:24.570 --> 00:04:27.466 So, the gates will also be fine for a century or so. 102 00:04:27.466 --> 00:04:29.830 What if a ship hits the Barrier? 103 00:04:29.830 --> 00:04:31.810 Well, ships are designed to float. 104 00:04:31.810 --> 00:04:34.170 Relatively speaking, they're not that solid. 105 00:04:34.170 --> 00:04:35.740 These piers are thick concrete, 106 00:04:35.740 --> 00:04:38.516 and each of the main gates is 3,000 tonnes of steel. 107 00:04:38.516 --> 00:04:40.570 So, the ship's going to come off worse. 108 00:04:40.570 --> 00:04:42.580 Collisions have actually happened several times. 109 00:04:42.580 --> 00:04:46.380 In 1997, a ship called the Sand Kite hit one of the piers. 110 00:04:46.380 --> 00:04:48.310 The only damage to the Barrier was some scratched paint 111 00:04:48.310 --> 00:04:51.120 and a broken ladder, but the ship sank. 112 00:04:51.120 --> 00:04:52.930 The trouble was that it sank where it hit 113 00:04:52.930 --> 00:04:55.580 and sat broken on top of one of the open gates. 114 00:04:55.580 --> 00:04:58.320 That would've blocked that part of the Barrier from closing. 115 00:04:58.320 --> 00:05:00.770 The contingency plan, if there had been a flood tide, 116 00:05:00.770 --> 00:05:03.020 would have been to close all the other gates 117 00:05:03.020 --> 00:05:05.290 and use the force of the water trying to get through 118 00:05:05.290 --> 00:05:07.133 to blast the wrecked ship clear. 119 00:05:08.160 --> 00:05:12.760 But also, even if one gate was completely broken somehow, 120 00:05:12.760 --> 00:05:14.610 the Barrier would still not technically fail. 121 00:05:14.610 --> 00:05:17.520 The margin for error would be gone, but that's why the margin's there. 122 00:05:17.520 --> 00:05:21.630 It would still hold back enough water to keep London safe. 123 00:05:21.630 --> 00:05:24.340 - We have a series of forecast models 124 00:05:24.340 --> 00:05:27.590 and we run what we call an ensemble forecast 125 00:05:27.590 --> 00:05:29.280 that looks many tides ahead. 126 00:05:29.280 --> 00:05:32.040 With any models, particularly forecast models, 127 00:05:32.040 --> 00:05:34.350 you will always get a degree of error. 128 00:05:34.350 --> 00:05:38.210 So, the forecaster here will make an adjustment to the forecast 129 00:05:38.210 --> 00:05:40.420 and that data will be passed to the controller 130 00:05:40.420 --> 00:05:44.070 from which to make a decision as to whether to close or not. 131 00:05:44.070 --> 00:05:46.980 - So, the barrier is highly reliable and strong. 132 00:05:46.980 --> 00:05:48.620 What if there's a really high tide, 133 00:05:48.620 --> 00:05:50.830 a storm surge that's too much for this to cope with? 134 00:05:50.830 --> 00:05:53.170 Well, the design odds of that are one in a thousand per year. 135 00:05:53.170 --> 00:05:56.360 Not technically failure, just not being specified high enough. 136 00:05:56.360 --> 00:05:59.264 The barrier was designed in 1970 for a 60-year life, 137 00:05:59.264 --> 00:06:00.980 which would be 2030, 138 00:06:00.980 --> 00:06:03.380 which must've seemed a very long way away then. 139 00:06:03.380 --> 00:06:05.950 The good news is that with modern simulations and modelling, 140 00:06:05.950 --> 00:06:08.920 the team here are confident that it'll last another 40 years at least. 141 00:06:08.920 --> 00:06:12.100 But part of the reason for that is because those 1970s engineers 142 00:06:12.100 --> 00:06:14.230 built in a huge margin of safety. 143 00:06:14.230 --> 00:06:17.830 They didn't cut corners then, and that's given us breathing space now. 144 00:06:17.830 --> 00:06:19.410 - The gates themselves, 145 00:06:19.410 --> 00:06:22.150 with inspections and remedial works, 146 00:06:22.150 --> 00:06:25.250 will probably be good to go until 2070, 147 00:06:25.250 --> 00:06:27.250 but a lot of the supporting infrastructure, 148 00:06:27.250 --> 00:06:32.000 the fire systems, the security systems, even the cranes on the piers 149 00:06:32.000 --> 00:06:35.270 will all have to be updated in the years to come. 150 00:06:35.270 --> 00:06:39.140 - But what if the tide went high enough to actually overtop the barrier? 151 00:06:39.140 --> 00:06:42.040 Well, those gates are built strong enough that any water they can't hold back 152 00:06:42.040 --> 00:06:45.040 can flow under into the low water on the other side. 153 00:06:45.040 --> 00:06:48.160 Remember, they close the barrier early, when the river's at its lowest point, 154 00:06:48.160 --> 00:06:51.750 so, yes, if some apocalyptic storm comes along, 155 00:06:51.750 --> 00:06:53.900 the barrier should still work well enough. 156 00:06:53.900 --> 00:06:55.540 If this was almost anything else, 157 00:06:55.540 --> 00:06:57.960 you could say the redundancy was over-specified, 158 00:06:57.960 --> 00:06:59.670 that requirements like this were ridiculous, 159 00:06:59.670 --> 00:07:01.970 but with so much at stake, 160 00:07:01.970 --> 00:07:04.470 the Thames Barrier is a marvel not just of engineering, 161 00:07:04.470 --> 00:07:08.653 but of how you have to build when something cannot be allowed to fail.