Skip to main content

CROSS Safety Report

Snow loading in Scotland - summary of reports

Report ID: 264 Published: 1 July 2011 Region: CROSS-UK

This report is over 2 years old

Please be aware that it might contain information that is no longer up to date. We keep all reports available for historic reference and as learning aids.

Overview

There were heavy snowfalls over the entire United Kingdom in the winters of 2009/2010 and 2010/2011 and CROSS received a number of reports, mostly from Scotland where a significant number of buildings suffered roof collapses.

Key Learning Outcomes

For civil and structural design engineers:

  • The fact that there are so many records of increased snow loading is a reminder that all loadings are statistical projections and there is always a chance that they will be exceeded

  • Careful consideration should be given to the risk of snow loads especially for areas prone to heavier snow falls

  • Consider what reasonably foreseeable loads could be applied beyond the code minimum values

For all built environment professionals:

Full Report

Find out more about the Full Report

Our secure and confidential safety reporting system gives professionals the opportunity to share their experiences to help others. If you would like to know more, please visit the reporting to CROSS-US page.

There were heavy snowfalls over the entire United Kingdom in the winters of 2009/2010 and 2010/2011 and CROSS received a number of reports, mostly from Scotland where a significant number of buildings suffered roof collapses. The reports have been summarised to give an overview from the reporters, but individual reports can be found on the web site data base.

Are current codes adequate?

A common thread is that the snow in Scotland fell vertically, in low temperatures, and in still air conditions which resulted in significant depths. These were reported as being from 600mm up to 1,200mm with ice underlying snow in some cases. This led to reporters questioning whether loads calculated from BS codes are adequate in such circumstances and how indeed would buildings fare if designed to Eurocodes.

One reporter encountered collapses of two modern well built structures – one was a 1,000m2 portal frame shed and another was a large domestic building with a flat roof carried on glulam beams. In both cases the reporter observed a build-up of ice on the roofs ranging in depth from 75 to 150mm overlain by around 400mm of snow. 

Arguably, and very likely they said, the loadings were even higher when the failure occurred. Another report was about a lightweight pressed metal framed building that completely collapsed with a snow depth of approximately 500mm on the roof.

Warehouse roof collapses

A significant report was about forty whisky storage warehouse buildings that suffered from partial or full roof collapses. These were generally steel frames with steel lattice roof trusses from 20-40 years old and 2,000 -2,500m2 in area.

Another reporter says that nine agricultural buildings collapsed (Figure 1) in a region where there was between 1,100 and 1,200mm of snow with an icy layer at the base of each snow storm and more powdery snow above each basal ice layer.

A collapsed steel frame structure in snowy conditions.
Figure 1: steel framed agricultural building in northern Scotland

Purlin failures

There was a further report of twenty five cases of agricultural buildings which collapsed when the depth of snow was 1,200mm. Most were steel framed portals with spans around 20 – 30m and generally the purlins failed first (Figure 2). The reporter recommends that in such buildings purlins should be at no more than 1.2 – 1.3m centres and that frame centres should be at 6m.

A collapsed steel frame structure and roof purlins in snowy conditions.
Figure 1: typical purlin failures

Most of these buildings were clad with cementitious sheeting which is favoured by farmers because there are no drips, particularly onto cattle, from condensation. In one case a farmer had lost two sheds and had a newly completed shed which was at risk of collapsing so he took in a number of straw bales and set fire to them to melt the snow.

Buckling of rafters

One reporter noted evidence of lateral torsional buckling instability of rafters between rafter restraints due to overload. Instability of the rafter in the apex haunch area suggests to him that it would be good detailing practice to fit rafter stays at each purlin position on either side of the rafter and at the two penultimate purlin positions either side of the apex, as these are in the vicinity of plastic hinge positions which are very likely to form should the roof be overloaded.

What is not known is how the return period of the recent snow event relates to the assumption is Codes of Practice. What is apparent, says another reporter, is that farmers want their buildings designed for real loads. Similar issues are raised in report 264 and also reports 183, 193, 195, & 248. You can search for safety information on snow loads and failures on the CROSS website.

Expert Panel Comments

Find out more about the Expert Panel

An Expert Panel comment on the reports we receive. They use their experience to help you understand what can be learned from the reports. If you would like to know more, please visit the CROSS-US Expert Panel page.

These are general comments on all of the above reports

The fact that there are so many records of increased snow loading is a reminder that all loadings are statistical projections and there is always a chance that they will be exceeded. Designers should be generally alert that safety is not prejudiced by any single assumption (in strength or loading). They should avoid situations where, should the assumption be in error, the result will be a significant safety hazard.

This might be the case if structures are very light weight and the only dominant loading is snow. Where there are significant amounts of dead and live loading, it is unlikely that both will be in error at the same time. Safety should not be sensitive to any one assumption. The example of the farmer burning straw to melt snow reflects the practice at some sporting events where there are marquees of having space heaters available to prevent snow from lying on the roofs.

Agricultural buildings

Agricultural type buildings may be designed with lower safety levels for economy and because safety to humans is not likely to be prejudiced. Designers (and their clients) should also consider carefully the commercial consequences of failure. It might be worth accepting a higher risk of failure if the structure is just protecting stored fodder, but if the protected contents are valuable livestock, it might be thought prudent to look for a more robust structure. The marginal cost increase of a safer more robust design may well be a sensible investment. It is not known however what design loads were used in any of the reported cases. 

Designers (and their clients) should also consider carefully the commercial consequences of failure.

Adapting to changes

The industry should also beware of changing practice and inadvertently rendering inherent assumptions invalid. Thus, the demand in many structures for increased roof insulation renders it less likely that snow will quickly melt or slide. A number of portal frames in these reports failed under direct overload which might be tolerable in terms of safety if the form of failure were a standard plastic collapse mechanism (i.e. excessive displacement short of unstable failure), but from the pictures this appears not to be so.

An investigation of forms of failure might show less risk of catastrophic damage simply by boosting connection capacity or paying greater attention to avoiding failure by instability: benefits that might be achieved at minimum cost. These examples and other evidence has been collated by the Scottish Government and used to issue guidance to the agricultural community and designers, and a report has been sent to BSI so that the data, along with other information, can be considered with reference to Eurocodes and their UK National Annexes.

Recent SCOSS alerts

Alerts were also issued by the Standing Committee on Structural Safety (SCOSS) in 2010 and 2011. The situation emphasises that design is a risk management process; it needs thought, and some argument perhaps, to decide if the relaxations currently permitted for agricultural buildings are as acceptable as once thought, when weighed against the cost of failure – albeit mostly economic in these cases.

For other buildings, it may also be that designing for increased snow loads and making additional effort to avoid build-up/drifting opportunity may be a worthwhile investment when set against business disruption should they present a high risk of failure.

Industry guidance

However, many of the collapses were probably of buildings which were designed to superseded Codes and it is possible that some buildings may not have been built as designed and that there were cases of construction defects and of maintenance issues.

Designers and constructors of new buildings, particularly agricultural buildings, in areas potentially subject to heavy snow falls, should refer to:

The winters of 2009/2010 and 2010/2011 were very severe in other countries and in the USA, for example, there were numerous reports of failure due to snow loads and it will be interesting to find out whether these are being regarded as abnormal. Of course, snow falls may be related to climate change, but the science, and work to support it, is not sufficiently advanced to provide an answer.

Share your knowledge

Your report will make a difference. It will help to create positive change and improve safety.

Our secure and confidential safety reporting system gives professionals the opportunity to share their experiences to help others.