More about messy, lumpy Mother Nature and how we deal with her effect on our forensic engineering and insurance investigations

Very brief summary of what we do

We do this by “zooming” in on the details of the natural environment and adjusting our investigations accordingly.  That is, apply the principles of the semi-empirical – observational, sciences and also that of simple geometry.

This underlies what we do, but we’re not thinking science and geometry as we examine the site of a failure or accident getting our hands dirty and mud on our boots.

An interesting, relevant article

I recently came across an article on fractions, like in arithmetic.  More correctly, the article was on fractals.  Fractals are an attempt in mathematics to come to terms with the fractured roughness of everyday life, including the natural environment. (Refs 1 and 2)

The article was very relevant to the nature and methods of forensic engineering where the natural environment is involved.  For me, learning about this was enjoyable coming as it did shortly after I posted a blog on messy Mother Nature.  And how this messiness sometimes results in difficult, “messy” forensic and insurance investigations. (Ref. 3) 

Problems in the real world of nature

High school geometry deals with straight lines – think triangles, squares and rectangles, and perfect circles.  But in the real world none of these exist.  Clouds, trees, soil, and rocks are fragmented, jagged, and fractional.  Nature is rough and lumpy – messy. (modified after Ref. 1, page 110)

Describing such real-world things has historically defeated mathematicians.  It has also – as I blogged recently – caused problems for forensic engineers investigating failures and accidents involving the natural environment.  The devil, it appears, is in the details. (modified after Ref. 1, page 110)

Detailed observation of nature

We have dealt with this problem in engineering with increasingly detailed observation of what we find in nature.  For example, more test holes and laboratory tests to characterize the irregular layers of soil existing beneath the site of a failure. (Ref. 3)

Some of our sciences have actually come to be recognized as semi-empirical in nature.  You could also say “semi-observational”.  For example, soil mechanics, rock mechanics, snow mechanics, ice science, hydromechanics, etc.  Medicine is a semi-empirical science.  Our analytical procedures and formula are based partly on what we theorize takes place in rough and lumpy nature and partly on what we observe actually takes place there.

Zooming in on the rough details of Nova Scotia

Fractal mathematics is also an attempt to come to terms with the fractured, lumpy roughness of everyday nature by also looking at the details.

An example of its use – possibly unknowingly, would be the measurement of the length of the coast of Nova Scotia.

On a small-scale map, say the highway map of Nova Scotia, the coastline of the province is very uncomplicated.  Because the level of detail is low, long stretches of the coastline can be represented with straight lines.  To get more detail and a longer, more accurate measure of the coastline you need a larger-scale map – to zoom in, in a sense, and use a ruler with smaller graduations.  (after Ref. 1)

This was done by someone, and at the level of detail finally accepted, it was found that Nova Scotia’s coastline is 15% longer – 7,400 km, than Canada is wide – 6,422 km.  (N.S. is 575 km long overland from one end to the other).  (Ref. 4)  Of course, the measurement could have zoomed in closer still and got an even greater distance.

There is really no limit to applying this process of “zooming in” and observing and studying at any level of detail any problem involving the messy, lumpy natural environment.  This is often necessary to complete a thorough forensic engineering or insurance investigation.  For example, in problems like:

  • Characterizing the contour and topography of the site of a failure or accident,
  • The depth, contour and physical properties of the layers of soil beneath the site,
  • How water drains across the terrain or flows below the surface
  • How a plume of contamination migrates across the site of a fuel oil spill

In a forensic engineering or insurance investigation of problems like these and others, the engineer would decide on an adequate level of detail in dealing with rough and lumpy Mother Nature.  His decision would be based on the evidence that comes in on completion of the different stages and tasks of the investigation.


  1. Jackson, Tom, ed., Mathematics: An Illustrated History of Numbers, page 110, Shelter Harbour Press, 2012
  2. Crilly, Tony, 50 Mathematical Ideas You Really Need to Know, page 100, Quercus, 2007
  4. Internet, Wikipedia, April 21, 2015, 2:00 pm

The “messiness” of some forensic engineering and insurance investigations is illustrated by messy snowbanks

Forensic engineering and insurance investigations can sometimes get quite “messy”.

We can easily identify the different stages in standard investigations – the protocol we must follow to carry out a thorough investigation and to conform to good practice.

But we can’t necessarily identify all the tasks involved at each stage and the length of time to complete each one.  Nor where the evidence will lead us and the need for follow-up investigations.  All of this impacts the ease/difficulty estimating costs which adds to the “messiness”.

This is often the case when the natural environment is involved in the failure or accident.  Such as in foundation and earthworks’ failures, landslides, floods, fuel oil spills, traffic accidents, and occasionally slip and fall accidents.

There’s pattern in nature that we often need to look for and characterize in determining the cause of these types of failures and accidents.  But Mother Nature doesn’t reveal herself readily.

This is particularly the case when a failure involves:

  • the terrain at the scene of a failure,
  • the foundation soils below, and/or
  • the surface water and the ground water below the surface.

Problems like this require us to identify:

  • the different layers of soil and rock beneath a site and
  • the physical properties of the different layers.
  • Also, how the water drains across the terrain’s surface and
  • how the ground water flows beneath the surface.

There’s often pattern in these elements of a site.  Based on a review of published topographic, geologic, and hydro-geologic mapping, we hypothesize its nature at the start of a forensic engineering or insurance investigation.

But, Mother Nature lies in wait.  If there’s one thing I learned as a civil engineer specializing in geotechnical and foundation engineering for a number of years, it was to “expect the unexpected” in attempting to characterize the foundation soil conditions beneath a site.  You can’t see the thing you’re investigating.  Also, “if in doubt go deeper” with your investigation.  I acquired the former bit of wisdom in the U.K. and the latter in Australia where I practised for a time.

“Mess” came to mind in connection with this difficulty when I was walking my dog recently just as a last snow storm was starting – “last” would be nice, but we’re not holding our breath in eastern Canada.

I was struck by the pattern – quite messy at some locations, in the layers of snow plowed up along the sides of the streets.  The different layers would be different snow storms and we’ve had quite a few.  What I saw in the snow was illustrative of layers of soil.

Soil is deposited in layers like snow and can be just as uniform or irregular like that seen in the snow banks.  And just as “messy”, irregular, and difficult to describe and characterize, and difficult to cost.

I took a few pictures and include a selection of these below – Figs 1 to 5 in the Appendix..  The pictures are two-dimensional.  It’s very important to remember that the irregular, messy pattern continues in the third dimension as well.  The layered patterns seen in Figs 1 and 2 are quite uniform.  Those in Figs 3, 4 and 5 are irregular and messy.

Similar “messiness” must be expected – as an initial hypothesis, to permeate all forensic work where the natural environment is an element in the problem.  For certain, the environment below the ground surface.  But often enough that above as well even though we can see the surface.

And this “messiness” must be expected to confound our efforts to identify all the tasks that will be necessary during a forensic engineering and insurance investigation, the time to carry out these tasks, and their cost.


Fig. 1  Layers of snow at side of road

Fig. 1 Layers of snow at side of road



Fig. 2 Layers of snow at side  of road

Fig. 2 Layers of snow at side of road


Fig. 3 Fairly irregular layers of snow.  If this were foundation soil at the site of failure it would be fairly easy to characterize

Fig. 3 Fairly irregular layers of snow. If this were foundation soil at the site of failure it would be fairly easy to characterize


Fig. 4 Somewhat more irregular layers of snow. If this were foundation soil it would be more difficult to characterize and would introduce some "messiness" into the forensic engineering investigation

Fig. 4 Somewhat more irregular layers of snow. If this were foundation soil it would be more difficult to characterize and would introduce some “messiness” into the forensic engineering investigation




Fig. 5 Quite a bit more irregular layers of snow. If this were foundation soil it would be quite  difficult to characterize and would introduce a lot of "messiness" into the forensic engineering investigation

Fig. 5 Quite a bit more irregular layers of snow. If this were foundation soil it would be quite difficult to characterize and would introduce a lot of “messiness” into the forensic engineering investigation

Why, in a recent blog, didn’t I seem to consider foundation failure as a possible cause of the Edmonton bridge failure?

Readers in Edmonton and Halifax commented on an item I posted last Friday about the Edmonton bridge failure – and my initial hypothesis as to the cause of the failure. (Ref. 1)

I concluded the bridge failed because a crane supporting a middle section of one of the bridge beams moved in the wind pulling the beam sideways and causing it to buckle.  Cross-bracing was not adequate enough to prevent this movement.

Gary in Halifax

Gary in Halifax wondered why I didn’t give links to the photographs and video that I studied and mentioned in my blog.  He thought this would make it easier for the reader.  He is quite right because we learn more visually than we do verbally or from text.

Basically, I was anxious to get the item out there while it was current news – and while some readers might still have the newspapers around that carried the story.

Googling “Edmonton bridge failure” quickly took me to a good number of sites with text, photographs and at least one good video.  I looked at a number of them in forming my initial hypothesis.

It would have been an exercise in itself reviewing all this material and selecting three or four links, and in the process possibly omit some that might have been included.  It seemed easier to let readers google.  But, I might have suggested readers google in lieu of providing specific links rather than unconsciously assuming they would.  I’ll do that next time.

Albert in Edmonton

Albert in Edmonton noted the lack of comment in my blog about the possibility of foundation failure causing the beams to buckle.  He knew that as a civil engineer I had specialized in geotechnical and foundation engineering for a number of years.  Why didn’t I mention the foundations?

Why I didn’t mention the foundations

Typical structural damage due to foundation failure

I didn’t mainly because I’ve seen a lot of structures damaged by foundation failure and what I saw in the pictures and video didn’t fit what I’ve learned over the years.

Most foundation failures – not all, mind you, result in typical damage to the structure above.  You come to know the character of this damage after seeing quite a lot of it.  Also, after seeing the different structures, foundations and sub-surface soils and rocks involved.

This damage is due to marked vertical movement of parts of the structure – inches, and a little horizontal movement – fractions to maybe inches.  It’s characterized by cracking and distortion of the structure.  The damage is differential in nature – more here and less there, not uniform throughout the structure.

(For certain, there are exceptions to this typical failure and distortion of a structure.  I investigated a 54 foot long structure one time that had “settled” – moved vertically, 11 inches from one end to the other.  Turned out it was mistakenly built this way)

Edmonton bridge damage

The middle sections of three of the Edmonton bridge beams moved sideways several feet, not inches.  The end sections where the foundations are located did not move sideways hardly at all.  And all of this movement was quite uniform, not differential.  Quite unlike the typical damage associated with foundation failure.

Did the two end sections of each beam move towards one another and cause the beams to buckle and to do this uniformly?  If so, where did the force come from to push on the ends of the beams?  There’s nothing showing in the pictures and video.

Did the foundations beneath the two end sections move towards one another causing the beams to buckle, uniformly?  I can’t see the foundations but I’ve never known foundations to move sideways several inches causing the structure above to move sideways too.  Landslides and retaining wall failures possibly excepted, but that’s not the situation here.  And where did the forces come from to push on the sides of the foundations?  To the extent you can see the soil near the level of the foundations in some of the pictures, there’s nothing there.

Intuition?  Engineering experience?

Sub-conscious thoughts like these would have been running through my head as I looked at the pictures.  They resulted in me not considering the foundations as the cause of the problem.  Intuition?  Engineering experience?

But Albert in Edmonton made me think and next time I’ll draw attention to the fact that something in what I was seeing didn’t fit, didn’t look right.

Similarly, the structural engineer I chatted with about this bridge failure quickly dismissed wind forces on the sides of the beams as being anywhere near strong enough to cause the beams to buckle.  That would be his intuition, his engineering experience kicking in.

Initial hypothesis of bridge failure validated to some extent by others 

Barry Belcourt, manager of Edmonton’s road design and construction branch, was reported recently in the Globe and Mail as saying it was a construction procedure failure.  This would suggest the failure is not due to inadequate design.  I don’t know what Mr. Belcourt is basing that opinion on but he’s on the ground in Edmonton and all I’ve got are pictures and video.

Mr. Belcourt’s comments to some extent validate my initial hypothesis of the cause of the bridge failure.

Albert, who is also a civil engineer, also considered “careless handling during construction” as one of several possible causes

Reader’s questions are important

Questioning is how we revise hypothesis so keep them coming.


  1. Wind, construction crane and inadequate cross-bracing caused Edmonton bridge failure: An initial hypothesis