What makes modern buildings less resilient to flooding?

With rainfall in much of the UK now arriving in shorter, heavier bursts, flood risks are increasing dramatically.

River flooding is being exacerbated by saturated ground and the failure of aging drainage systems such as culverts, while flash floods are being made worse by changes in ground cover, especially in cities. For coastal regions, additional problems include sea level rise, storm surge and erosion.

Meanwhile, insurance companies report that flood claims are even more common due to escape of water, that is, leaks from plumbing and water supplies. This problem is exacerbated as supply and sewer systems age or reach maximum capacity.

Yet traditional and modern building envelopes – or greatcoats and raincoats – behave differently when exposed to flood-water, just as they differ in terms of their vulnerability to rain.

This was confirmed by testing at Sheffield Hallam University, where full-scale sections of frogged-brick and block wall were placed on a balance before being subjected to a head of water.

The walls proved no barrier: indeed, the flood-water passed so quickly through the exterior frogged-brick wythe that it was higher in the cavity than in the tank.

Cavity-wall sections built tightly of high-fired engineering brick in contrast proved very resistant, though because of the small size of the sections tested this result must be interpreted with some caution.

Engineering bricks gain their resistance to water by being fired at high temperatures for long periods. When shrinkage cracks appear, the bricks continue to be heated until those cracks are compressed and sealed, turning them into raincoat materials.

Cement mortars are also hard and impermeable. In full-scale walls, however, this combination can be counterproductive: settlement and other building movement often causes cracks to appear at the interfaces between mortar and brick, which become perfect capillaries for siphoning flood-water through the wall.

Another weak point for flood entry in raincoat buildings can be damp-proof courses (DPCs), and even injected DPCs have been found to act as conduits for flood-water. Membrane-based systems create capillary paths that bridge the walls, and because the membrane often continues through the floor slab flood-water can be carried right through to the heart of the building. Subsequent drying is difficult and very slow, especially when the floors are concrete or cement.

Remediation will often necessitate opening up the wall. This can be challenging and damaging, especially where there are proprietary building systems and the original materials are no longer available. Where the cavities have been used as conduits for pipework and electrical wiring, remediation can be dangerous as well as difficult.

Not least because of these problems, it is sadly not unknown for rainscreen buildings to be demolished when they sustain flood damage, even when they are listed. This was the case with Carlisle's 1964 civic centre, flooded in 2015 and subsequently razed.

To make raincoat envelopes more resilient, potential entry points for water must be identified and sealed in advance of any flooding; but this process is not generally straightforward.

For cavity wall construction, at least, recent testing on the Sheffield Hallam rig suggests a way forward: rendering the frogged-brick sections with lime-based mortars appeared to provide much-needed protection from flooding

This evidence includes the many buildings deliberately constructed in flood-prone areas to take advantage of the proximity to water, from mills to river terraces in trading ports. Even after serious flood events these buildings are able to dry out, with only prolonged or repeated exposure to standing water leading to failure.

The flood resistance of greatcoats derives from the same source as their resistance to driving rain, that is, from their materials and structure, and from the air trapped in the pore system.

To understand why, it helps to compare what happens when a single brick is rested on a shallow tray of water to the outcome should the same brick be dropped into a full water bucket.

When only the bottom surface touches the water, that water will rise into the brick, pushing out the air in the pores ahead of it. As a result, the brick in the tray quickly saturates.

By contrast, a brick dropped into a bucket will take many weeks or months to saturate, since the air in the pores cannot readily escape. This trapped air resists the uptake of water.

When a real wall is flooded, a large area of surface is in contact with the water. As with the brick in a bucket this water stops the air in the pores escaping easily, so the wall resists the flood-water.

Surveyors sometimes use Rilem tubes, also known as Karsten tubes, to try to assess the vulnerability of walls to moisture penetration. However, because these expose only a very small area of surface to water they are not a reliable tool. 

As with the brick on the tray, the air being displaced can easily move into neighbouring pores and out through the surrounding wall surfaces, allowing the water in the tube to be absorbed many times more quickly and easily than would be the case with flood-water.

There was almost no leakage even though a change in colour could be seen, the internal surface became damp to the touch and the balance confirmed that the wall was slowly taking up some water.

A particularly interesting observation was that the lowest-fired of the hand-made bricks were the most resistant to water uptake. The researchers attributed this to the absence of the shrinkage cracks that start to appear as firing continues, and which would make effective capillaries for drawing water into the brick.

The overall result is that, for greatcoats, a single flood is not usually serious. Historic England's field research in York and Cumbria has confirmed that buildings take up much less water than expected, and often all that is needed after a flood is to clean them down and let the permeable materials dry out naturally. This they appear to do surprisingly quickly.

The drying of porous materials is divided into two stages, as described by researchers.

Stage 1 is when water moves from the interior to the surface as a liquid flowing through capillaries: the continuity of the liquid pathway from core to surface allows evaporation to draw out the moisture from the wall. This is governed by pore structure and the environmental conditions outside the wall, and is fast and effective.

But to dry the bulk of the wall, water transfer must not be too quick. If evaporation is removing water from the surface faster than it can be replenished by the water traveling through the convoluted pore system, the liquid pathways will then start to break up and the system enters stage 2.

In this second stage, water in some part of the capillaries will be in vapour rather than liquid form. Unlike liquid flow, the movement of any vapour molecule is random, and almost entirely governed by the shape and materials of the pore surrounding it.

Vapour movement within a wall is therefore completely independent of the temperature, humidity and air movement outside the wall, even in pores close to the surface.

Bulk drying then effectively stops until condensation and seepage in the pores has once again been able to establish new, unbroken liquid pathways to the surface. This can take some time.

This will quickly make the surface dry to the touch, but once the heating and fans have been turned off the moisture in the core will start to redistribute. After a few weeks, pathways back to the surface have formed once again and the system returns to stage 1. The surface will again become wet, and mould may suddenly appear.

For maximum effectiveness, drying after a flood must be kept at stage 1 for as long as possible. The best approach is gentle air movement coupled with good ventilation, seeking to remove water at a gradual pace that keeps the liquid flow paths intact. The optimum conditions will depend on the pore structure of the masonry, but essentially – like the hare and the tortoise – slow and steady will win the race.

It is useful for surveyors to recognise that, when drying is proceeding at the right pace, the surface will stay wet to the touch right until the end of the process. Once the last of the liquid in the core has been carried to the surface, it will start to feel dry and remain dry.

Like the application of raincoat systems to try to increase rain resistance, such coatings will have little impact on the amount of water absorbed in a flood event but will significantly impede subsequent drying.

Moreover, whenever air and salts are trapped under a raincoat finish it is likely to cause serious failure. It is little surprise that blowing and spalling of treated surfaces are commonly observed by surveyors after floods.

Although greatcoats are flood-resistant by nature, prolonged exposure to flood-water or repeated floods before the wall has had the chance to dry also cause problems. Sheffield Hallam is currently investigating the impact of repeated flooding, and testing has already proved that solid walls will absorb considerably more water during a second flood.

This makes sense: the wetting and drying of the wall as a result of the first flood will have created a network of water-filled capillaries to the surface. These will be able to absorb water from a subsequent flood, and indeed will also make the wall more susceptible to absorbing rain.

If flooding causes unexpectedly serious problems, it can be helpful to consult the long-term records for the building. Has it flooded regularly in the past, or are there other moisture problems such as poor maintenance that may have allowed liquid-filled pathways to form in the wall?

This could well be a result of guidance being copied from one local authority to the next, since many councils lack the technical knowledge that would prompt them to question what others have published.

For example, guidelines often specify retrofitting DPCs as a way of flood-proofing walls, although in practice these are likely to introduce pathways for water penetration.

Other perverse recommendations include applying waterproof coatings, or installing perimeter trenches filled with gravel. Gravel fill has very large pores that make it easy for water and mud to soak down into the bottom of the trench, while at the same time preventing evaporation.

Perimeter trenches should therefore be back-filled with ordinary clay-rich soil instead, which has small pores that resist water penetration but encourage evaporation.

Current guidance also gives little recognition to some critical pathways for flood-water penetration: for example, the introduction of below- and above-ground conduits for pipes and cables and equipment such as heat pumps.

It is also always worth bearing in mind that although insurance companies report that plumbing leaks remain the most common cause of flood damage – leaks that can reduce buildings' resistance to subsequent floods – advice still focuses almost entirely on weather-related flooding.

It would therefore be helpful for RICS and other technical bodies to draw attention to such problems in guidance that tends otherwise to be useful, and perhaps also to offer a broader definition of flooding.

To enable this, we urgently need some means of allowing practitioners and researchers not simply to learn from observation and experience, but to share their findings with each other and with policymakers quickly and openly.

Coordinating this exchange of knowledge and synthesising it into improved guidelines for flood resilience is a role that RICS would seem ideally placed to assume.