Scientific and Technical References

This very large relatively unsorted file contains the consulting engineers' analysis of the condition of the embankments, informed by site visits that took place in late summer - fall of 2017.   

This is the document that the CC highlights in its new videos (Spring 2019).  It has lots of information about how erosion starts and progresses, but the only statement about the risks of vegetation is pasted below. Note that the document uses FEMAs   guidelines from above -- again, those are not supported by scientific research -- and uncited anecdotes  ("numerous incidents..." ) as evidence that vegetation is problematic on dams.  

"3.3 Biologic Activity (Animal and Vegetation) 

Many State dam safety agencies and FEMA have published guidance on maintenance of vegetation on embankment dams (FEMA 473, 2005; FEMA 534, 2005). Numerous incidents have occurred due to improper maintenance of vegetation and animal activity on levees and embankment dams. For example, the Truckee Canal10 is thought to have failed as a result of burrowing animals. Burrowing animals weaken embankments by providing shortened seepage paths and voids. Heavy vegetative cover provides habitat for burrowing animals, and the root systems can also lead to internal erosion. Large trees can topple and also result in breach of a dam or levee. Proper vegetative and animal maintenance is required to address these potential failure modes."

New Army Corps implementation guidance for vegetation management on levees (October, 2017).   Relevant text highlighted.

Key quote from the Memorandum: "The results of the study indicate that the number of performance records that discuss vegetation influence on levee performance is small (95 records or1.4% of 6,970 records). Of these, a small number of records (11 records or 12% of the 95 records that indicate vegetation influence on levee performance) indicated that vegetation played a role in the levee performance. The majority of these performance records discussed the influence of vegetation on the levee O&M. Generally it described the inability to visually determine levee performance during high water events."

Key quote from the Synthesis of Levee Vegetation Research Results:  "The effects of trees on levee soil properties, slope stability, and seepage-induced piping are small and hard to measure or simulate. Impacts can be beneficial or slightly detrimental depending on conditions and situation. A field experiment in California showed that seepage was slower in a levee slope containing a decaying tree stump than in an adjacent section of the same levee slope with only herbaceous vegetation."

Chapter 4 from the "International Levee Handbook" (2013) describing the state of the art in woody vegetation management on levees in different countries.  Different countries take different approaches to managing established trees, no approach is described as "safest". (Information about the handbook here) The relevant section starts on page 33 of this PDF.    

Engineering report on 1912 canal breach in Bushnell's Basin.   Note that the problem was caused by modification of the culvert bringing Irondequoit creek under the canal, and that the structure of the embankment is described as sand. The woody vegetation on the Great Embankment is likely doing a very important service for us by stabilizing this sandy soil.  At the Jan 18 public meeting, the CC engineers did not seem to know that this embankment was in fact mostly sand, raising serious questions about how well they studied this problem before taking action.

Comprehensive look at all dam breaches caused by internal erosion up to 2007.  Bottom line: of 267 failures described and recorded since 1868 due to internal erosion, only 5 were reported to have to do with tree root growth.  

What this tells us is that 98% of dams that have failed due to internal erosion did so without trees being mentioned as a possible problem.  Removing trees will not make the dam safe from failure. 

If what we need is a better inspection regime, let's try to figure out how to do that while retaining the trees that provide other values.

Similar to the above paper, this one published in 2016. 

The biggest cause of failures in earthen dams is connections between built structures (like culverts) and soil.  We already knew this is the case on the Canal based on historical failures.

In addition, they find that the older the embankment, the less likely it is to fail. This suggests that mature vegetation could be helping -- not causing instability.

Another review.  This does indicate that "homogenous earth fill" embankments (which is what -- I think?-- most of the Erie Canal embankments are) are the most prone to failure.  

But the risk of failure of these types of dams after 5 years of operation is 0.02% per year. 

The risk is increased when there is "limited internal erosion control." Unless the PA intends to completely rebuild the embankments, one could easily argue that healthy tree roots are now PROVIDING internal erosion control by preventing the formation of a continuous "pipe".

In light of all of this research, California updated its levee management strategy in 2017 to leave much established vegetation in place.

Remember when the huge Oroville Dam in CA nearly collapsed in winter 2017?  The vegetated portion was not a problem then, nor has it ever been.

OK, this isn't about embankment stability, but it does have survey data on why people visit or are interested in the Canal.