Snow Load Code Development and Building Failures and Investigations

Snow Load Code Development and Building Failures and Investigations

“Winter is Coming”   OK, just making sure all you GOT fans are paying attention…

State College hasn’t seen much in the way of snow flakes so far this season but there is no question that winter weather is quickly approaching in many areas of the country in addition to Pennsylvania.  Some locations with higher elevations have already experienced several inches of snow on the ground.  With more snow on the way and the inevitable roof damage and collapses that seem to follow at some point, Paul Rouis, PE, CEO of Ryan Biggs Clark Davis  in Albany, NY  participated in the Visiting Practitioner Failures Lecture Series to Penn State AE students on the topic of Snow Load Codes and their evolution over the years.  Mr. Rouis also discussed several snow collapse case studies including pointing out lessons learned so that hopefully you won’t make the same mistakes as others.

Structural collapse and roof damage from snow can occur from a variety of sources and often results from a combination of events including snow drifting, roof geometry, storm overloads from ice and snow combinations, design deficiencies  and construction defects just to name a few.   A good background article on this topic can be found in the November 2008 issue of Structure titled “Structural Collapse from Snow Loads” by Michael O’Rourke.  Another good summary of snow collapses can be found from the Building Failures Forum guest post by Alyssa Stangl titled Snow-Induced Roof  Failures and Prevention Methods which also leads to a detailed discussion on the topic on the Failures Wiki site.

On a related topic, the Whole Building Design Guide contains a section titled Considerations for Building Design in Cold Climates: Avoiding Falling, Sliding or Windblown Ice and Snow from Buildings and Structures.  In addition, no snow collapse discussion would be complete without first looking at FEMA P-957 Snow Load Safety Guide.

While snow collapses are nothing to kid about, MKev does occasionally find humor or at least irony in the topic as can be seen in the photo below.

I guess this means I shouldn’t be standing here but how was I supposed to read the sign?

And to start out the commentary, remember the comment by Mr. Rouis about how before the early 1970’s many roofs didn’t have much in the  way of insulation so the escaping heat pretty much took care of long-term snow build up?   See my comments  below from last year about snow melting on an industrial scale.   And, ask me in class about the snow melt idea for one of the campus buildings that fortunately never came to pass.   MKev

Tags: , , , ,

48 Responses to “Snow Load Code Development and Building Failures and Investigations”

  1. Ishan Uppal
    November 8, 2016 at 9:04 am #

    The article “Structural Collapse from Snow loads” was a great read d clearly explains the procedure for clearing off accumulated roof snow in a safe manner. I had various doubts regarding the methods adopted for the same that were cleared.

    The various talks and presentations by industry experts and highlight the importance of consideration of snow loading while design. It was made evident that snow loading conditions at a particular location depended upon various factors suck as the geographical location, the exposure to the elements at that site, the elevation etc. Also during the class case studies and assignments such as Rostraver Ice Arena snow collapse it became apparent that the geometry of the roof, the span between supports and the height of the adjoining structures if any play vital roles in snow load calculations. The assignment for calculation of Pg for Portage, PA showed how highly probabilistic the method of determination of snow load values for a particular area can be.
    All these factors prove how tasking a job authoring Tue Snow Codes really is.

  2. Mehrzad
    November 8, 2016 at 8:45 am #

    Apart from having roof drains that are not clogged, prevention of ice dam occurance at roof edges is important. It happens when the heat somehow manages to melt some portion of the roof snow and the running water is freezed at the edges. This may cause water leakage into the building envelope, collapse of the overhangs, and fatalities. To avoid ice dam, heat transfer to the roof snow should be minimized. The newer designs have more thermal insulation of the roof snow from the roof of the building envelope. The article below talks in simple terms about why a 2-inch airspace under roof sheathing for ventilation (vs 1-inch code-specified) or a combined usage of the vented space and a “compact roof” in areas with ground snow load more than 50 psf is recommended.

  3. Shubham
    November 8, 2016 at 3:03 am #

    Another take-away from the guest lecture is that, when the owner feels of making any structural changes (or even if he/she isn’t sure), a word with the original designer is most important. Changing existing conditions changes the statics and need to be compared with existing design.

  4. Brendan B
    November 8, 2016 at 12:32 am #

    Reading the “Structural Collapse from Snow Loads” article, it’s interesting to see how many roof collapses are due to drifting. The article states that about 70% of the failures in the author’s practice have been due to drifting of one kind or another. This means the majority of failures are the result of the engineer not accounting for a special type of snow drift on the roof. Due to wind being unpredictable, it can certainly be tricky to consider every possible drift scenario, especially situations where drift loading is combined. An example the article provides of combined drift is a parapet on a gable roof that receives a pure parapet wall drift and a gable roof drift for wind in the perpendicular direction. This is where years of experience and knowledge of several different roof collapses can help prevent failure due to drift. A proficient understanding of the code as well as experience in the industry is essential for preventing these failures.

  5. Yingzhe You
    November 7, 2016 at 9:18 pm #

    I found a reference reading material talks about the snow-induced building failure incidents, and I want to share with you all.
    From the report we can see we still suffer a nation-wide snow collapse.

  6. Alec B
    November 6, 2016 at 11:17 pm #

    I found a case study involving a partial roof collapse due to snow drift on a pre-engineered metal building in Oklahoma City, Oklahoma (link below). The case study gives a great in-depth analysis of the failure and includes pictures documenting the failed Z-purlins of the roof structure (the failure mode of the Z-purlins seems to be just as Professor Parfitt described in general during class). What I found most interesting about the case study was that the building was designed to code, but the code-predicted design snowdrift height was significantly lower than the observed height of snowdrift after the storm event.

    • mkev
      November 7, 2016 at 9:42 am #

      Yes, a article from our forensic buddies at Nelson. They have been known to recruit PSU AE students. Note in particular the photo showing the buckling of the Z purlin. Pretty much as discussed in class!

    • Joe H
      November 7, 2016 at 9:04 pm #

      An example like this definitely proves what the code is: a bare minimum. To design to code means do have reached the absolute minimum requirements for the case the code specifies. Naturally, designing for this minimum is the most cost effective and efficient way to design, so how can an engineer determine when the code is not good enough, i.e. when they should exercise engineering judgment.

      It also shows the flaws in the code’s use of statistics. Although statistically it is very unlikely that the drift height was supposed to exceed 4 feet, in this case it reached 10 feet. Unfortunately, it is events like these that cause codes to be reconsidered and altered. The changes are all reactive, not progressive.

      This brings up an interesting point of debate. Now that a 10 foot drift is seen in this part of Oklahoma, does the code need to be changed to address this issue? Or is this event so random that taking the extra design considerations for this snow would be superfluous on almost all occasions? These are the difficult questions that need to be answered, and unfortunately you can never really be proven right in these situations. If a building stands up, that’s expected. But a code change can be proven wrong if a failure happens because of it.

  7. Yamile R
    November 6, 2016 at 6:50 pm #
    This is an interesting article about why there were so many roof failures due to snow in 2010-2011 in the Northeast.
    In those years the ground snow load did not exceed the code, but other factors influenced these cases. One is that when the codes were adapted, most of the buildings were already built and another is that significant loading reveals hidden structural defects (detailed in the article).

    • mkev
      November 7, 2016 at 9:40 am #

      Yes, one of those “cycles” we mentioned. There is a similar article out there about the big storms of the 1990’s.

  8. philr
    November 5, 2016 at 5:49 pm #

    Here is an additional resource I found, that provides additional and illustrative detail. It has a lot of good discussion on how to (or not) remove snow from roofs:

    • mkev
      November 7, 2016 at 9:37 am #

      The author of this version of the presentation is a PSU AE grad from 2000! MKev may or may not have contributed some ideas to this via a former professor from Penn State. At the very least, you all should note that your project is shown in the presentation!

      You are all doing a good job of answering the questions this time. Keep it up! We will fill in some of the gaps in class later this week.

    • Di W
      November 7, 2016 at 9:12 pm #

      This article actually solved my question. I was wondering with the increase of the snow depth, the density of snow may increase as well. This article showed me a diagram that indicates the snow density will not change even if the snow depth increases. However, the specific gravity changes over time.
      We should consider these when we do a case study.

  9. Prateek Srivastava
    November 4, 2016 at 10:08 am #

    On reading the article on “Consideration for Building Design in Cold Climates”, which evidently state the problems of falling snow and ice from high rise buildings which are very prominent in big cities like New York, Chicago and many more.

    In cities like these, the interaction between the pedestrians and vehicles is of few meters which thereby takes in account the safety of these people who are non occupants besides the safety of people using the building facility. My question being, are there any special codes which define safety measures for such conditions?

    The article also mentions about the design considerations which can be taken into account to prevent ponding, freezing/falling of snow. These design considerations go hand in hand with the micro-climate studies as they to a large extent define what is essential for the area in consideration because there have been failure cases due to improper weight-age of local weather conditions while designing.

    So engineers have to design considering both macroscopic and microscopic effects, in order to ensure the longevity and serviceability of a building, at the same time complying with the safety measures.

    • philr
      November 5, 2016 at 5:54 pm #

      Here is a New York law related to icicle removal.

      • Prateek Srivastava
        December 1, 2016 at 2:17 pm #


        Thank you for your response.

    • ErikS
      November 7, 2016 at 4:35 pm #

      Prateek, I am not aware of other specific codes defining safety measures; however, similar to falling snow or ice from building roofs is falling snow or ice from smaller horizontal projections such as fenestration trim covers, watertables, or similar as well as falling cladding elements themselves. When working with WJE I had begun investigating several instances of deep trim covers (snap covers) falling from buildings, especially during heavy snow events. There is little to no code regulations in the IBC covering cladding attachments and these cladding elements have fallen from many stories to occupied (or often unoccupied during storm events luckily). I am not aware of many injuries but they have occurred. IBC discusses “component and cladding loads” but this is generally understood to be the structural framing for fenestration itself and not necessarily the decorative “trim” attached to the fenestration frames.

      What we were finding was that the deep trim covers, which were sometimes 6 inches to 18 inches deep, were insufficiently mechanically attached (without fasteners usually) and were either being knocked loose when snow was falling from above or when the snow melted it would run between within the raceway that connected the trim cover to the window frame and then would re-melt opening the connection between the two elements, which would lead to failure of the connection and displacement of the trim cover. The language was ambiguous and stated that the trim was required to withstand maintenance impact from workers, but generally nothing more specific than that.

      There are firms that specialize in investigation and designing for snow and ice “accretion” or the building up of snow or ice on objects exposed to freezing precipitation. One such company (which I just found out has closed their doors as of October 31, 2016) is Northern Microclimate ( and they were working on ASTM standards to research and develop design guidelines to minimize this risk. They also helped write articles for the Whole Building Design Guide ( Their site is still active and they have links to many of their articles they wrote and research they conducted.

      • Prateek Srivastava
        December 1, 2016 at 2:16 pm #


        Thank you for an elaborate response. It is really helpful to learn from someone else’s experience.

  10. Mehrzad
    November 3, 2016 at 11:41 am #

    Paul’s lecture was interesting in showing what failures snow can cause and what preventive design and modification considerations should be taken into account. Various shapes of snow load diagrams are necessary in structural calculations. Proper drainage system design by the architects are equally important. Structural engineers of record are sometimes asked for the weight of the snow that they designed for as average pound per cubic foot by owners. This allows owners / facility mangers to measure weight of the actual snow settled and compare to the design value while speeding up removal of the snow. As Paul mentioned, it also is important to include snow design of the components that are not part of the main structural system, such as awnings.

    On the other hand, the non-technical FEMA document below, intended for owners and facility managers, in chapter 5 explains measures needed to be taken to minimize the snow and ice effects. It also talks about snow guards on roof, and snow drift and slides in simple figures in other chapters.

  11. Prateek Srivastava
    November 3, 2016 at 8:34 am #

    Mr. Rouis gave a good presentation about the snow load determination and usage of codes. As he clearly depicted the improvement of snow loads from the historic times till date.
    He touched upon various technical terms like awning drift, sliding plus drifting, unbalanced plus drifting and many more which provided a better insight in understanding the topic.
    He gave his insights from the case studies some of which he was a part, thereby, it was interesting to know the practical ways and methods to actually handle snow situations. The best part of the presentation was the structure of case studies, i.e., it started from the problem statement went to investigation and analysis, conclusions and recommendations and finally culminated with lessons learned, which was an important aspect as we need to actually enumerate the points which we can carry forward as experience or wisdom to utilize it on other similar cases.
    So, overall it was a holistic source of information.

  12. Shubham
    November 3, 2016 at 5:05 am #

    Very informative session by Paul Rouis. Helped us understand the reasoning behind standards mentioned in snow load Codes, mechanism that unbalanced loading can lead to, sliding and drifting combined case, roof steps in series case, and discuss the interesting (cool) case studies.

    In the Dome failure case study, the fabrication issue was major after the original design was flawed. Instead of square-cut, the vertical members should have been cut for accordingly accommodating variable angle dome shape for a good bondage with connecting members.
    Such small and specific construction details, if disregarded, may trigger/ contribute to failure and prove fatal. The design engineer should specify such detailing.

    One important take-home lesson is that, along with theoretical knowledge an engineer should accommodate practical site conditions during the analysis (hence in modelling too). Because the structure needs to be safe for practical (not-ideal) conditions.

  13. Ishan Uppal
    November 3, 2016 at 3:11 am #

    The presentation on the Snow load Code development was very informative and made me realize the importance if taking into consideration all the various loading conditions arising for snow loa, at the design stage. These could easily be overlooked by a designer at his/her own peril. In that vein, it was comical for me in 2016 to imagine that at some point of time there were just 3 values of snow load for the entire United States. Speaks volumes about how the Snow code has evolved.

    Mr Rouis’s lecture was more important as it is directly linked to our course projects.

  14. Shane M
    November 3, 2016 at 2:28 am #

    The snow load presentation was very informative and I am glad we had the opportunity to gain some knowledge for our semester projects. Every time I hear more about snow load analysis, it reveals itself to be more complex and important than I had previously thought. I am beginning to realize how easy it is for designers to overlook some of the more complex load situations.
    One particular case I seem to come across much too often is the snow collapse of pre-engineered metal buildings. These buildings are always being changed over their life time with additions and renovations. These additions are very quick and easy to add so they might not get the special attention from an engineer. The addition of a lower structure is obviously going to cause drifting issues, so I wonder if there are any requirements for the owners to get their structures checked before even small additions.

  15. Brendan B
    November 3, 2016 at 2:11 am #

    Mr Rouis’ presentation was interesting to hear as I have never really considered the history of snow loads, and why they are what they are today. Obviously, todays codes are developed based off past events. What caught my eye was when Mr Rouis mentioned how older buildings did not have much insulation which helped take care of snow build up on the roof. Why do we no longer try to melt the snow on the roof from heat loss? Is it cheaper to use stronger roof framing members instead of producing enough heat to melt the snow? Snow drift can be very unpredictable, and fighting long term snow build up from a mechanical approach could help reduce the risk of roof collapse.

  16. Di W
    November 2, 2016 at 10:39 pm #

    Very instructive lecture provided by Mr. Rouis. I learned the history of Snow Load codes changing and some classic snow overload failure cases. And now I am starting to realize why so many old buildings collapsed due to heavy storms.

    There is another snow-induced failure case you might be interested in. This article shows snow collapse in different kinds of structures such as wood and steel.

  17. Yingzhe You
    November 2, 2016 at 10:30 pm #

    Mr. Rouis’s presentation was very informative for me.
    The first part about the development of the snow code was very interesting. I thought the code should have been well established long time ago since we had a long history of building houses, but the fact was that the drifting, unbalanced load part was introduced several years before.
    And we also need to pay attention to the surcharge of the rain on the snow which may lead to additional load for the roof.

  18. YusufA
    November 2, 2016 at 9:21 pm #

    Mr Paul Rouis’s lecture may be one of the most informative session this semester. As it touches base on our two ongoing projects for the semester. I am sure that many of us would at some point have many more questions we would have hoped we had asked in the lecture as we get deeper into the projects. The Rec Hall and the Ice Rink Arena both have similar features to the case studies which were mentioned in the lecture.

    I for one is familiar with the use of solar panels in buildings and know quite well that this works best in the sunny climates or rather the summer season here in America. However, for a household or a building which uses this alternative power source which is erected on the roof which must have been calculated to hold the gravity load of it, what happens when the winter season sets in?

    This case was briefly touched upon in the lecture and it looks obvious now that the panels would just rather hold off most of the snow load and cause an excessive extra load for the roof to hold which in turn could lead to roof failure or collapse depending on the snow load. With this kind of difficulty however, I would like to know if an alternative solution to this problem has been found because otherwise this could discourage the use of this solar panels in buildings in the area where snow is expected because of the stress of maybe removing and replacing during the summer seasons.

  19. MichaelB
    November 2, 2016 at 6:13 pm #

    Mr. Rouis lecture was a good overview of snow load code has evolved over the years. It is interesting to learn how little snow was considered early on in building engineering. I definitely got the sense that it is better to be conservative for snow drift rather than strictly economical as drift can be very unpredictable.

    This is especially interesting for roofs with large spans. It was mentioned in class that some buildings have failed due to ponding caused by deflection. I think it would be worth considering a stricter deflection limit for roofs especially for long spans if the engineer thinks that the weather could cause snow to thaw then accumulate more snow later. In addition roofs are typically designed with a looser deflection limit than floor members so stiffening the roof members would not be a huge jump.

    • Yemi O
      November 8, 2016 at 12:19 am #


      I totally agree with you. I always wondered why there wasn’t a stricter deflection limits on roofs as you mentioned, but it seems there are other controlling factors.

  20. Joe H
    November 2, 2016 at 5:12 pm #

    When hearing that long term snow load used to be almost a non issue, my first thought is how could a technology be implemented to recreate this scenario? Is there essentially a way to “turn off” insulation when a large amount of snow is on the roof? I know most insulation is a permanent material, so it would be unreasonable to remove or modify these permanent features, unless a method were developed to easily take out and reinstall the insulation.

    I understand that this idea raises a lot of questions, more from a mechanical issue. How would the building perform if the snow were used as the insulation as opposed to roofing insulation? How much snow would be needed to be effective insulation? How long would this scenario take place for? By how much could the snow load be reduced on the structure? Would this allow a wothwhile size reduction?

    I know that this idea is kind of wild, but if we had prior success over a certain issue with a specific method, I think it is worth looking into if that method can be reapplied in a different manner and integrated into another building system.

    • Joshua Z
      November 6, 2016 at 10:19 pm #


      With regards to keeping the snow on the roof for insulation, I think it would be interesting as you’d have to keep the roof at the right temperature so that the right amount of snow melts and/or stays on the roof. Additionally, as you mentioned, it’d be necessary to determine the proper amount of snow required to make it act as effective insulation. However, this is not the only variable that needs to be considered. Snow can vary in terms of moisture content and density as well, and finding a way to control these seems like a difficult task. Additionally, I’m curious as to how mechanically controlling the snow would impact the thermal comfort of the rest of building.

    • MichaelB
      November 7, 2016 at 12:58 pm #


      Interesting idea. I sounds similar to one of the solutions Professor Parfitt was talked about for the indoor football practice field where someone proposed heating the roof to melt the snow. As Professor Parfitt mentioned the electricity costs would be extreme for this solution and I imagine that temporarily reducing building insulation would have a similar effect.

      Maybe there could be a way to direct all waste heat from a building to the roof in a manner that melts the snow. I am pretty sure there are examples of server buildings that use the heat from processors to control the water temperature of a nearby swimming pool such as this example:

      • Joe H
        November 7, 2016 at 9:07 pm #

        That’s a pretty neat technology! For use in this situation it would definitely have to be the appropriate building type to produce this heat waste. Another factor is then could the heat be used to serve a different purpose. I’m hesitant to say a “better” purpose because different people have a different opinion of a “better” use, but this is a unique potential solution to the issue.

  21. Rebecca M
    November 2, 2016 at 3:28 pm #

    After having some background of the snow codes based on general academic situations, Mr. Rouis’s discussion of the codes’ evolution and application gave a deeper understanding to what was originally interpreted. When only reading ASCE 7-05 and 7-10 snow codes, it was hard to imagine a building being designed in a heavy snowfall area without those provisions.

    In these older buildings where the snow that gathered on the roof melted via heat loss, were structures ever renovated and reinforced to handle a larger roof load after the insulation had been upgraded due to the energy crisis? Or did it take a failure or full collapse for design professionals to realize what was happening to these buildings in larger snow fall regions?

    Historically under the codes of the time, did designers realize that snow typically melted off of occupied building roofs, so they only designed to minimum loading? Would the situation ever turn into negligence on the design professionals part for not considering that the load was lighter due to envelope inefficiency? This is especially curious for designers originally from areas that never experienced larger or even “case study” snow loading and for those that followed the minimum code design.

  22. Ommar E
    November 2, 2016 at 3:15 pm #

    Snow loads proved to be unforgiving to faulty roof designs or poor building techniques. Mr. Rouis’s lecture on Snow Load Code Development and Building Failures and Investigations revealed yet another snow loads induced roof failure where the roof geometry is nontraditional one. Going all the way back to Ms. Sonja Hinish’s presentation of Solutions for the built World and the case study for Tension Fabric Roof Collapse to Rostraver ice rink collapse case and the Dome collapse case study in Mr. Rouis presentation the common thread is snow loads causing collapse of a nontraditional roof structure. It was quite clear in the latter case the inadequate design was the culprit. Understanding the vulnerabilities of such types of roofs geometry and the peculiar nature of how they interact with loads and stresses would help remedy and avoid altogether many of future collapses. Using conventional design approaches and traditional factors of safety might not be enough.

    • Yamile R
      November 6, 2016 at 6:19 pm #

      You are totally right. All the roofs that you mentioned are of irregular geometry and fabric. Reading section 7.6 of the code, it mentions snow loads for flat, hip, gable, curved, barrel vault and dome roofs. But it does not specify how to approach those with other characteristics, such as the collapses on the tension fabric roof and the dome.
      So, like you said, it might be considered further analysis to determine snow load on this type of structures and not only what is included in the code.

      • Ommar E
        November 7, 2016 at 3:26 pm #


        If you add to roof geometry another factor of the spans being on the long side, then unless careful analysis and design in addition to very close construction supervision the risk of collapse becomes even higher. There are many examples of that in sports arenas in snowy regions of The U.S. Metrodome Collapse being one, located in downtown Minneapolis, Minnesota. , another one is HARTFORD CIVIC roof collapse CENTER , By the way this one was authored by Ryan G. Johnson, BAE/MAE, Penn State, 2009.
        Both facilities were built after the code improvements of the early 1970s were already there. Obviously, the code provides for minimum requirements and most common types of construction and it is not feasible or even possible to include every single situation out there. So, it is incumbent on the structural engineer or the responsible design official in these cases to do his do diligent and provide a structure that ensure the safety of the occupant and the public and get it approved by authorities, to quote from the IBC [A] 104.11 “The provisions of this code are not intended to prevent the installation of any material or to prohibit any method of construction not specifically prescribed by this code, provided that any such alternative has been approved”

  23. Joshua Z
    November 2, 2016 at 2:46 pm #

    Mr. Rouis gave a very informational presentation about the history and development of snow codes through time, as well as how they are relevant in construction and failures. One thing that I found intriguing was his discussion of collapse related to drift. In case 2, the building in question, a pre-engineered metal building, collapsed because of snow overload. The code did not account for drifted load moving as a solid mass. And, as he mentioned, failure due to this was not uncommon. I understand that the building was erected in 1991, and that the code has changed since then. I also understand that you’re supposed to treat drifting snow and sliding snow separately, as they are both intended to be 50 year events. However, how does one design for failure due to drifted snow moving as a solid mass? Is there a special provision in the code that deals with this? Is it widely accepted that you design a roof for snow normally and then assume that this condition won’t happen? Or is there a specific factor of safety or consideration that should be accounted for in order to prevent this kind of failure?

    • Shubham
      November 8, 2016 at 2:59 am #

      Interesting question Joshua. I did look up for the provisions, but didn’t find any (I might have missed it). Prof Parfitt will be able to tell us, if there are any!

      Without any snow provisions, I would design the roof and the columns at the roof edge (or point of obstruction)(edge is most critical for Snojax), for snow load equal to the ground snow load depth observed in that region. As the roof is sloped, one can apply some reduction to this load, based on coefficient of friction and compactness of snow.

  24. Yemi O
    November 2, 2016 at 12:27 pm #

    I have always wondered how buildings back in the day dealt with snow loads, especially seeing how the codes have evolved over the years. However, snow load collapses are still an issue.

    As I have come to learn more about buildings, it becomes more and more apparent that you never really know what to expect with regards to failure, even when all areas seem to taken care of.This is why the industry is still growing and coming innovative solutions to tackle these issues i.e. performance based engineering methodologies. You can find some more information on this online.

    Furthermore, after doing some research on the topic, I found an article that assists with prevent of snow load roof collapses. The article link can be seen below

    • YusufA
      November 2, 2016 at 9:57 pm #

      Yemi, I tried to access the link which you posted but I was unable to. Perhaps you could revise this.

      But I completely agree with you that there is much more to learn and design for in a structure than the structural components. Attention needs to be paid to little details here and at the same time many a times an engineer needs to work on finding more innovative ideas to solve problems which are being faced in the industry to avoid failure and possible live savings.

      I believe a lot of continuous work has been done in this area and this can be evident from the evolution and improvement of the snow code over the years. Looking at the Snow Load Safety Guide in the link above, more insight was given to this and how many building codes were used before the current acceptance of the International Building Code, IBC.

      Many potential hazards are naturally been encountered as a result of snow on roof in a building and many of it is gradually been tackled using various means. With different geometry of roof available, there is almost none which doesn’t experience an accumulation of snow or cause drift to adjoining or lower structures.

      Lastly, the rather unconscious method of solving this issue which was used prior to 1970 is a method which I believe could still be the most effective and reduce the amount of snow at a roof at any given time. By this I mean to continuously heat the roof during the event of snow. Knowing that this would only lead to an increase in the cost of the heating, a design engineer needs to put preventive measures in place and make all necessary calculations of the maximum expected possible roof snow load in the area and design a roof to withstand this while the other methods of snow collection is been put in place.

      • ErikS
        November 2, 2016 at 10:11 pm #

        Yusef, the link text is correct but the hyperlink had inadvertently cut off the “.pdf” from the end of the link (which may happen with mine as well – likely the ‘(3)’ confused it. If my link does not work either just copy the full link below and copy it – it should work.

        • YusufA
          November 6, 2016 at 11:37 am #


          Thank you for the heads up. I did not realise that, just went for the link straight.

          I have been able to now access the pdf file. Good information generally for home owners.

  25. Alec B
    November 1, 2016 at 11:43 pm #

    Mr. Rouis gave an excellent talk about very relevant topics including the evolution of snow codes, applying/interpreting snow codes, and building failures related to snow loading. One thing I found particular interesting was the process of removing snow from roofs. I did some additional investigation related to snow removal in the FEMA Snow Load Safety Guide. I was surprised to find so many “instructions” related to snow removal, and many times snow removal requires the knowledge/insight of a Professional Engineer and the experience of a licensed, insured professional roofing contractor. There are many things that can go wrong when removing snow including creating unbalanced snow loads that can lead to overstressing of some structural members.

    The Snow Load Safety Guide (and common sense) suggests that when existing snow load is close to the capacity of the roof structure, snow removal is prudent. Of course, roof live load or snow load governs during design. When people and snow removal equipment are on the roof when the snow load is close to the capacity of the roof structure, isn’t there load on the structure that was not accounted for in design since snow load governed over roof live load? Does anyone know of any collapses/partial collapses as a result of people being on the roof when the snow load was close to the capacity of the roof structure?

    • ErikS
      November 2, 2016 at 10:33 pm #

      Alec, I cannot specifically reference any cases where a full roof/building collapse has occurred due to snow removal but they have occurred…hence the strong recommendation for professional engineering oversight. Generally if there is a note/comment/requirement about something it must have happened…similar to how we are instructed in the CE Survey class to not throw your surveying poles like javelins (or at least they warned us that when I took it…).

      Anyway, snow removal is a very important process and must be considered logically, and often with the assistance of a Professional (engineer or architect) knowledgeable in snow removal. After Snowmageddon in the Mid-Atlantic area in 2010, WJE and other similar firms, were kept quite busy assisting owners with snow removal procedures as it can become dangerous very quickly.

      I am aware of projects where, as Mr. Ruis discussed, the awnings/porticos, were overloaded and collapsed due to snow overload; however, this was not due to the snow event itself but the snow removal process. Contractors pushed snow to the edge of the roof then over the side but did not account for what might be located below, and in these cases, a small overhang, portico, roof structure was overloaded and collapsed. I have also seen the base of walls collapse due to the same procedure where the snow piled up against the wall at ground level after being removed from the roof and induced a lateral load the wall was not designed for.

      Unbalanced loadings become of great concern, as you mentioned, as snow is removed from one side of a beams tributary area but not the other and, as Mr. Ruis commented on, the loading reversed and the members were damaged.

      Back to you original question there is some capacity (or at least there should be) remaining beyond the design load from a typical snow event. Even when an extraordinary event (Snowmageddon) occurs there is some additional capacity that should not be overloaded from properly considered and planned snow removal procedures. However, with that said, placing snow removal equipment (snow blowers) or numerous crew to remove the snow can quickly exceed that additional capacity; making that engineered snow removal procedure critically important especially in extreme snow events.

  26. mkev
    October 26, 2015 at 6:08 pm #

    I should also mention that in a few very specialized structures heating is sometimes still used regardless of cost. The Minn. Metrodome had a system of steam heating hoses used to clear snow but the problem was it was manual and people had to go on the roof to use it. You can see how well that turned out by reading about the many collapses of that dome: and at failures wiki

    The Carrier Dome in Syracuse is a little simpler. They track the weather and if it even looks like a heavy snow fall is coming, they turn up the heat inside which then escapes via the roof and helps to keep the dome clear of snow (or at least limit it) See this article from Feb. of 2014:

Leave a Reply