Content uploaded by Farshid Vahedifard
Author content
All content in this area was uploaded by Farshid Vahedifard on Nov 09, 2021
Content may be subject to copyright.
A Dataset of Levee Overtopping Incidents
Stefan Flynn, P.E., M.ASCE1; Farshid Vahedifard, Ph.D., P.E., F.ASCE2;
and David Schaaf, P.E.3
1Geotechnical Engineer, US Army Corps of Engineers, Rock Island District (corresponding
author). Email: Stefan.G.Flynn@usace.army.mil
2CEE Advisory Board Endowed Professor and Associate Professor, Richard A. Rula School of
Civil and Environmental Engineering, Mississippi State Univ., Mississippi State, MS.
Email: farshid@cee.msstate.edu
3Senior Structural Engineer, US Army Corps of Engineers, Risk Management Center.
Email: David.M.Schaaf@usase.army.mil
ABSTRACT
The most common mode of levee failure, breach due to overtopping, is generally
considered as a function of a complex set of hydraulic and geotechnical factors. The US
Army Corps of Engineers (USACE) actively tracks and compiles data related to such events
across the United States within their portfolio of levee systems. Recent experience suggests
that frequency of flood events is increasing, leading to both increased stress on levees, as
well as the probability of overtopping. As such, data related to this failure mode is constantly
evolving. This paper presents and discusses a dataset of more than 200 overtopping events of
USACE levee systems throughout the nation. The dataset is employed to examine the effects
of various factors and to establish correlations among parameters using both breach and
nonbreach events. Generating a discussion based on empirical evidence, this research is
aimed at furthering the profession’s collective understanding of levee behavior when
overtopping occurs. The dataset can provide a valuable basis for planning, design, and risk
assessment of levees systems.
INTRODUCTION
The U.S. Army Corps of Engineers (USACE) maintains a semi-quantitative dataset,
which documents reported loading events and historic performance associated with levee
segments across the USACE national levee portfolio, known as the USACE Levee Loading
and Incident Dataset (LLID). The LLID considers most components that can be part of a
flood risk management system (i.e., levee embankment, floodwall, pump station, and closure
structures). The dataset contains information on many distress incidents, which are generally
tied to commonly-assessed potential failure modes. While the LLID contains information on
multiple failure modes, the topic of research presented herein is the dataset failure mode of
breach due to levee overtopping. Overtopping is a critical concern for flood risk management
systems, and a logical dataset to consider first, given that breach due to overtopping is the
most common mode of levee failure (Hui et al. 2016; USACE 2018). Assessing levee
overtopping requires a complex combination of hydrological and geotechnical parameters.
As such, it is imperative that datasets such as the LLID continue to be refined and calibrated.
Validation of this data will allow for future creation of models to achieve better insight into
the uncertainty of the overtopping phenomenon.
Geo-Extreme 2021 GSP 330 99
© ASCE
Geo-Extreme 2021
Downloaded from ascelibrary.org by Farshid Vahedifard on 11/08/21. Copyright ASCE. For personal use only; all rights reserved.
The overtopping data subset is semi-quantitative in that it contains both qualitative and
quantitative variables. The qualitative data for each system documented includes, but is not
limited to, location, operational and maintenance responsibilities, generalized material
classification, construction entity, protective structure type (i.e. levee or floodwall),
evacuation considerations and anecdotal evidence. Quantitative and semi-quantitative data
regarding overtopping events includes date of event, breach locations, number of breaches,
loading conditions, embankment geometry, breach dimensions, and evacuation time range
data. Quantitative data is often presented in ranges due to a lack of precise measurements.
Currently, the dataset includes 230 levee overtopping incidents that range in date from 1948
to 2019. Data is derived primarily from an exhaustive research effort of USACE construction
documentation, flood fight reports, and post event repair documentation provided by
individual districts. Additional overtopping information comes from dated aerial videos and
photographs showing locations of overtopping and breaches. Additionally, general aerial
images from software such as Google Earth and Bing Aerial also provide information
relative to breach locations after the floodwaters have receded. Overtopping data is compiled
from the USACE districts across the nation including Buffalo, Louisville, Pittsburgh, New
Orleans, St. Paul, Rock Island, St. Louis, Baltimore, Philadelphia, Kansas City, Portland,
Seattle, Walla Walla, Omaha, Jacksonville, Mobile, San Francisco, Fort Worth, and Tulsa.
Figure 1 shows a map of all USACE district boundaries for reference.
Figure 1. Map of USACE Districts (Source: U.S. Army Corps of Engineers, Headquarters
Website).
While a great deal of information has been collected and compiled regarding
parameterized compositional and spatial data throughout the USACE levee portfolio, it is
critical to further that effort to include refined performance data for a more complete
Geo-Extreme 2021 GSP 330 100
© ASCE
Geo-Extreme 2021
Downloaded from ascelibrary.org by Farshid Vahedifard on 11/08/21. Copyright ASCE. For personal use only; all rights reserved.
assessment of these levees. The overarching goal of gathering and analyzing this data is to
inform decision making with regard to design and evaluation of flood risk management
systems in an effort to better inform guidance, policy and risk assessment. As more data is
collected and assessed, it is the intent of the author to expand efforts to further assess other
areas of the LLID related to additional failure modes.
The dataset discussed in this paper is expected to add to the collective field of failure
analysis. Several others (Gui et al. 1998; Isola et al. 2020; Kamalzare et al. 2013; Ozer et
al.2020) have made attempts to create databases and models to better understand levee
overtopping failures. Ozer et al. (2020) presented a review of various flood risk databases in
dam and levee safety. Gui et al (1998) detailed an earlier attempt of creating reliability
models for riverine levee segments. Several models have since been created, including a
bivariate methodology which looks at hydrological characterization of levee overtopping
(Isola et al. 2020). Overtopping based on surface erosion has been investigated Kamalzare et
al. (2013), where erosion parameters are modeled in a controlled setting and applied to
computational analysis. In addition to those discussed, several other research endeavors have
considered database analysis, field and scaled testing, and parametric study. Common to all
of these models is the need for data, which the LLID presents in a manner than can be
applied to various statistical models.
USACE LEVEE OVERTOPPING DATASET
The focus of current research is to evaluate the USACE dataset specifically related to
overtopping events. Levee breach caused by overtopping is a common failure mode and is
considered in most, if not all, risk assessments of USACE levee systems. Levee overtopping
occurs when flood water elevation exceeds the height of a levee at any given point along its
alignment. Therefore, it is necessary to understand two critical inputs when assessing the
likelihood that overtopping is going to occur for a given system. These two inputs are the
elevation of the levee and the probability of flood water exceeding this elevation.
Levees within the USACE portfolio are typically surveyed on a periodic basis, with
elevation data stored within USACE district offices and the National Levee Database
(USACE 2020). When assessing overtopping probability, it is important to locate extended
areas along the alignment that are likely to overtop for a given hydrological event such that
efforts to raise isolated low points no longer is practical. This is often referred to as the
incipient overtopping location(s). These locations can occur anywhere along a levee system,
can change over time, and may be present due to a number of reasons including, but not
limited to, settlement, rutting from vehicular traffic, manmade crossings, rodent borrows and
distress during a previous flood event. For the purposes of understanding the LLID,
overtopping events are considered where widespread overtopping occurred that could not be
contained by flood fight measures.
Flood loading is considered in terms of the frequency of occurrence of a flood event and
is often described as the annual chance of exceedance (ACE) or annual exceedance
probability (AEP). The ACE, or AEP, is given a probability value based on a hydrological
interpretation of the likelihood of occurrence. For example, a 1% event correlates to an event
that is expected to have a 1% chance of occurring each year. More colloquially, this might be
described as a 100-year flood event, as it statistically has a 1-in-100 chance of occurring in a
given year (USACE 2018).
Geo-Extreme 2021 GSP 330 101
© ASCE
Geo-Extreme 2021
Downloaded from ascelibrary.org by Farshid Vahedifard on 11/08/21. Copyright ASCE. For personal use only; all rights reserved.
The probability of overtopping as a finite frequency is provided to risk assessment teams as
part of the general background information of a levee system. This frequency of occurrence is
the starting point for levee risk assessment when considering overtopping leading to breach. The
distinction between breach and non-breach overtopping is critical in that consequences of
overtopping without breach are generally lower, in terms of both life and financial loss, than the
same event leading to overtopping with breach. Understanding the contributing factors that act as
a cumulative tipping point between the two scenarios are critical to what is being investigated in
this assessment of the LLID.
What comes next in the assessment is the evaluation of resiliency of the levee when
subjected to that given overtopping event. Along with technical assessment, the effects of related
floods at corresponding elevations are discussed with the local entity or group responsible for
flood fighting the levee system to assess the levees resiliency. Resiliency, in this sense, is the
levee’s ability to withstand overtopping loading and subsequent breach. Considerations in
assessing levee resiliency when overtopping include embankment material type, duration of
overtopping, depth of overtopping, embankment slope protection, embankment height, and
steepness of the embankment slope. When evaluating resiliency, subjectivity becomes critical in
risk assessment and design evaluation, and it is a goal of this study to better understand how
levees perform when subjected to overtopping loading as supported by empirical evidence.
Finally, some data has been inferred or elicited from district or levee personnel, therefore
portions of the qualitative data rely on human recollection and judgment. As a result of
documenting events from decades ago, information gaps exist in many of the earlier noted
overtopping events. It should also be noted that some breach locations were reported as
“multiple”, or without having a real sense for exact number as this information wasn’t available
in the post flood repair reports. For the purposes of this analysis, these recorded events were
considered as a single event.
CATEGORIZATION OF VARIABLES
To practically analyze the overtopping data set, incidents are organized based upon physical
differences. The data set is broken down for the purposes of initial research based on a flood load
source, generalized erosion resistance classification, and construction and maintenance
responsibility. General trends are observed related to these general categorizations of data and
future efforts in refining the dataset will serve to better correlate and utilize additional available
data.
Flood load source refers to the type of loading the levee system experiences, i.e. riverine,
canal, or coastal loading events. Riverine levee loading refers to any inland water source loading
and is generally considered in terms of steady-state (static) or transient (dynamic) loading. In
regard to the evaluation of overtopping events, transient loading is the more critical consideration
because duration of overtopping is related to the dynamic process of the water rising over the
crest and receding below the crest. Canal levee segments are typically highly regulated, therefore
dynamics involved with canal loading differ from riverine loading. Canal loading makes up a
very small percentage of the LLID, with only one documented event which was included in the
coastal dataset. As more canal loading events are documented, this information will likely be
distributed. Coastal levee loading refers to any loading related to surge or tidal action. These
events are always dynamic, and often related to tropical storms and hurricanes. Of the 230 levee
embankment overtopping events documented, 214 are riverine events (93%) and 16 are coastal
Geo-Extreme 2021 GSP 330 102
© ASCE
Geo-Extreme 2021
Downloaded from ascelibrary.org by Farshid Vahedifard on 11/08/21. Copyright ASCE. For personal use only; all rights reserved.
events (7%). Of the coastal levee events, 14 of 16 (87.5%) are directly related to 2005 gulf coast
hurricane events, which have been heavily evaluated (Briaud et al 2008; Seed et al 2005, 2008;
Sills et al 2008; Ubilla et al 2008). This distribution agrees well with the overall USACE levee
portfolio which breaks includes approximately 95% percent riverine levees and 5% coastal
levees (USACE 2018).
Each embankment contained within the LLID is categorized by erosion resistance related to
descriptions found within available documents held by individual districts. Descriptions of levee
system materials where breaches occur vary from broad material type descriptions to laboratory
classification and are subject to engineering interpretation. Erosion resistance categories within
the LLID are defined as “low”, “moderate”, or “high” relative erosion resistance with
embankments having little to no associated material data being classified as “other” or “no
classification”. Material descriptions included within the low relative erosion resistance category
include sand, silty sand, silty sand with gravel, sand/silt mix, sand/gravel mix, sand/gravel mix
with silt, sandy silt, and sandy gravel. Material descriptions included within the moderate relative
erosion resistance category include silt, clayey silt, silt with sand/clay, silt/clay mix with sand,
silty loam, silty/clayey loam, sand/silt mix with clay, sand with silty/clay, and clayey sand.
Material descriptions included within the high relative erosion resistance category include clay,
clay/silt mix, clay with sand/silt, zoned embankment with impervious cover, and clay
enlargement of an existing sand levee. Figure 2 shows a breakdown of all embankments
classified within the LLID overtopping dataset where 25% of levees are classified as low relative
erosion resistance, 33% are classified as moderate relative erosion resistance, 38% are classified
as high relative erosion resistance and 4% are classified as other or not classified.
Figure 2. Summary of Erosion Resistance Classifications.
Levee systems are also categorized by quality of construction and maintenance associated
with the levee embankments. This distinction is separated into two categories, “locally
constructed/maintained and re-classified federal levees” and “federally constructed/improved
38%
33%
25%
2% 2%
High Relative Erosion Resistance Moderate Relative Erosion Resistance
Low Relative Erosion Resistance Other
No Classification
Geo-Extreme 2021 GSP 330 103
© ASCE
Geo-Extreme 2021
Downloaded from ascelibrary.org by Farshid Vahedifard on 11/08/21. Copyright ASCE. For personal use only; all rights reserved.
levees”. The differences in these two designations are centered on construction authorization,
quality of original design/construction, available data and observed maintenance actions. A “re-
classified” federal levee is one which has known design/construction or widespread historical
maintenance deficiencies. Of all embankments classified within the LLID overtopping dataset,
55.7% were classified as locally constructed/maintained and re-classified federal levees and
44.3% were classified as federally constructed/improved levees. Given that several levee systems
throughout the country were constructed long before federal construction authorizations and
appropriations for levees existed, the difference in distribution is considered reasonable.
LEVEE OVERTOPPING PERFORMANCE
The remainder of presented analysis will focus on overtopping events relating the
aforementioned categorizations of relative erosion resistance and construction and maintenance
designation. First, overall breach rates are considered for overtopping events. Breach rate (Rb) in
this analysis is simply the ratio of the number of overtopping events resulting in breach (Nb) to
the total number of overtopping events in a single category (N). The total number of overtopping
events is equal to the sum of breach events (Nb) and non-breach (Nn) events per given category.
This can be shown mathematically as:
Figure 3 shows the distribution of overtopping events that resulted in breach versus non-
breach for each construction and maintenance designation. As can be interpreted from the
histogram, breach rates of levees when overtopped are significantly lower when comparing
federally constructed/improved levees (46%) to locally constructed/maintained re-classified
federal levee segments (80%). The cumulative breach rate of all systems in the levee overtopping
data set is 65% given that overtopping of the embankment occurs, regardless of classification.
Figure 3. Summary of Levee Overtopping Breach Rates by Construction and Maintenance
Designation.
80%
46%
65%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Locally
Constructed/Maintained
and Re-Classified Federal
Levee Segment/Reaches
Federally
Constructed/Improved
Levee Segment/Reaches
All Embankments
Levee Breach Rate
Geo-Extreme 2021 GSP 330 104
© ASCE
Geo-Extreme 2021
Downloaded from ascelibrary.org by Farshid Vahedifard on 11/08/21. Copyright ASCE. For personal use only; all rights reserved.
Figure 4 further refines the breach rate analysis, where overtopping rates are shown for
both construction and maintenance designation and relative erosion resistance. As shown in
the figure, likelihood of breach due to overtopping increases as relative erosion resistance
decreases, which is expected. Considering only the relative erosion resistance of the
embankment material, levees with low relative erosion resistance breach at a rate of 84%
when overtopped, this figure decreases to 74% for levees with moderate relative erosion
resistance, and to 45% for levees with high relative erosion resistance. Significantly, it is
noted that as relative erosion resistance categorization of the levee improves, construction
and maintenance designation play a major role in breach rates. While all embankments with
low relative erosion resistance breach at a rate between 83-85%, levees with moderate
relative erosion resistance show a breach rate disparity of 30%, and those with high erosion
resistance show a disparity of 43%, when considering locally constructed/maintained and re-
classified federal levees versus federally constructed/improved levees. This is a strong
indicator that federally constructed and maintained levees are typically more resilient than
those that are locally constructed and maintained. When considering riverine versus coastal
levees, a similar relationship is observed. Figure 5 shows that, when coastal events are
excluded from the data set, relative erosion resistance has a slightly increased effect on
breach rate for federally constructed and maintained levees. Low relative erosion resistance
levee breach rate is unchanged while moderate and high relative erosion resistance levees
breach at a rate 2% and 6% lower, respectively. Table 1 includes a summary of all levee
overtopping breach and non-breach data.
Figure 4. Summary of LLID Levee Breach Rates by Relative Erosion Resistance.
85% 84%
71%
83%
54%
28%
84%
74%
45%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Low Moderate High
Levee Breach Rate, Rb
Locally Constructed/Maintained and Re-Classified Federal Levee Segment/Reaches
Federally Constructed/Improved Levee Segment/Reaches
All Embankments
Geo-Extreme 2021 GSP 330 105
© ASCE
Geo-Extreme 2021
Downloaded from ascelibrary.org by Farshid Vahedifard on 11/08/21. Copyright ASCE. For personal use only; all rights reserved.
Figure 5. Summary of LLID Riverine Levee Breach Rates by Relative Erosion Resistance.
Table 1. Summary of LLID Overtopping Data
85% 84%
71%
86%
52%
22%
85%
74%
44%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Low Moderate High
Levee Breach Rate, Rb
Locally Constructed/Maintained and Re-Classified Federal Levee Segment/Reaches
Federally Constructed/Improved Levee Segment/Reaches
All Embankments
ALL EMBANKMENTS - ALL RESISTANCE CATEGORIES No. of OT Events OT with Breach OT w/o Breach
Local and Re-Classified Federal Segment/Reaches 128 103 25
Federal Constructed, Well Maintained 102 47 55
All Embankment Overtopping Events 230 150 80
HIGH RELATIVE EROSION RESISTANCE EMBANK No. of OT Events OT with Breach OT w/o Breach
Local and Re-Classified Federal Segment/Reaches 34 24 10
Federal Constructed, Well Maintained 53 15 38
Higher Erosion Resistance Embank OT Events 87 39 48
MODERATE RELATIVE EROSION RESISTANCE EMBANK No. of OT Events OT with Breach OT w/o Breach
Local and Re-Classified Federal Segment/Reaches 50 42 8
Federal Constructed, Well Maintained 26 14 12
Moderate Erosion Resistance Embank OT Events 76 56 20
LOW RELATIVE EROSION RESISTANCE EMBANK No. of OT Events OT with Breach OT w/o Breach
Local and Re-Classified Federal Segment/Reaches 40 34 6
Federal Constructed, Well Maintained 18 15 3
Low Erosion Resistance Embank OT Events 58 49 9
OTHER EROSION RESISTANCE EMBANKMENT No. of OT Events OT with Breach OT w/o Breach
Local and Re-Classified Federal Segment/Reaches 0 0 0
Federal Constructed, Well Maintained 5 3 2
Other Erosion Resistance Embank OT Events 5 3 2
NO EROSION RESISTANCE CLASSIFICATION EMBANKMENT No. of OT Events OT with Breach OT w/o Breach
Local and Re-Classified Federal Segment/Reaches 4 3 1
Federal Constructed, Well Maintained 0 0 0
No Erosion Resistance Embank OT Events 4 3 1
Geo-Extreme 2021 GSP 330 106
© ASCE
Geo-Extreme 2021
Downloaded from ascelibrary.org by Farshid Vahedifard on 11/08/21. Copyright ASCE. For personal use only; all rights reserved.
In addition to analysis of breach rates of embankments, current efforts are investigating physical
properties collected for each breach event. One example of this is breach top width, or the length of
the levee that has breached during for a particular event. This set of analysis considers all levees in
the LLID with a defined number of breaches given an overtopping event. Breach top width is
assessed for each segment given that a breach occurs. Analysis of this dataset does not include levees
which had an undefined or unknown number of breaches for a given overtopping event. The number
indicated on the horizontal axis is the top end of the range, whereas the value to the left of a given
horizontal axis values is the bottom end of the range. For example, if 30 events occur with a Breach
Width of 250 feet, these 30 events are between 200 feet and 250 feet in length. Figure 6 shows a
summary of the frequency of all overtopping event breach widths contained within the LLID dataset.
Breach width data indicates a heavy grouping of collected measurements less than 400 feet, with
individual breach widths greater than 400 feet being less common. As part of future research and
analysis of the LLID, physical performance parameters such as breach width will be further
investigated.
Figure 6. Summary of LLID Breach Width Data.
CONCLUSION
The Levee Loading and Incident Dataset’s overtopping subset can best serve to inform risk
assessment of overtopping failure modes for levees both in the design and post-construction phases
of a project. Contained within the presented analysis are just two general trends derived from the
overtopping subset Levee Loading and Incident Dataset which serve as insight into valuable
correlations that can be derived from future efforts to analyze the LLID. Overtopping breach
probability based on construction and maintenance designation and relative erosion resistance for a
limited dataset is presented as an example of data that can be used when assigning likelihoods of
breach occurrence given levee overtopping. It is not recommended that exact values be used as
standard values to be inserted in risk failure mode event trees, but rather as general range values to be
considered when assessing risk. Results presented in this paper synthesize a portion data contained
within the overtopping data set of the LLID only, and would benefit from additional data collection
to further define the existing overtopping events within the dataset, as well as to expand the dataset to
include additional, previously undocumented overtopping events. With further data collection and
refinement, greater confidence can be placed in the various subsets within the LLID, including the
overtopping dataset analysis, which will allow for more guided use in the levee design and risk
assessment processes.
0
10
20
30
40
50
60
50 150 250 350 450 550 650 750 850 950 1100 1300 1500 1700 1900 2250 2750 3250 3750 4500 5500
Count (Failures/Range)
Breach Width (range between each number on horizontal axis) (ft)
Geo-Extreme 2021 GSP 330 107
© ASCE
Geo-Extreme 2021
Downloaded from ascelibrary.org by Farshid Vahedifard on 11/08/21. Copyright ASCE. For personal use only; all rights reserved.
REFERENCES
Briaud, J. L., Chen H. C., Govindasamy, A. V., and Storesund, R. (2008). “Levee erosion by
overtopping in New Orleans during the Katrina Hurricane.” Journal of Geotechnical and
Geoenvironmental Engineering, ASCE, 134: 618-632.
Gui, S., Zhang, R., and Xue, X. (1998) “Overtopping Reliability Models for River Levee”
Journal of Hydraulic Engineering, ASCE, 124(12): 1227-1234.
Hui, R., Jachens, E., and Lund, J. (2016). “Risk-based planning analysis for a single levee”.
Water Resources Research, AGU. 52.
Isola, M., Caporali, E., and Garrote, L. (2020). “River Levee Overtopping: A Bivariate
Methodology for Hydrological Characterization of Overtopping Failure”. Journal of
Hydrological Engineering, ASCE, 25(6).
Kamalzare, M., et al. (2013) “Computer Simulation of Levee Erosion and Overtopping” Geo-
Congress 2013. San Diego, CA. USA.
Ozer, I. E., van Damme, M., and Jonkman, N., (2020) “Towards an International Levee
Performance Database (ILPD) and Its Use for Macro-Scale Analysis of Levee Breaches and
Failures” Water 2020. 12., 19.
Seed, R. B., et al. (2008). “New Orleans and Hurricane Katrina. I: Introduction, Overview, and
the East Flank.” Journal of Geotechnical and Geoenvironmental Engineering, ASCE, 134:
701-717.
Seed, R. B., et al. (2005). “Preliminary report on the performance of the New Orleans levee
system in Hurricane Katrina on August 29, 2005.” Rep. No. UCB/CITRIS-05/01, National
Science Foundation.
Sills, G. L., Vroman, N. D., Wahl, R. E., and Shwanz, N. T. (2008). “Overview of New Orleans
Levee Failures: Lessons Learned and Their Impact on National Levee Design and
Assessment.” Journal of Geotechnical and Geoenvironmental Engineering, ASCE, 134(5):
556-565.
Ubilla, J., Abdoun, T., Sasanakul, I., Sharp, M., Steedman, S., Vanadit-Ellis, W., and Zimmie,
T., “New Orleans Levee System Performance during Hurricane Katrina: London Avenue and
Orleans Canal South.” Journal of Geotechnical and Geoenvironmental Engineering, ASCE,
134: 668-680.
US Army Corps of Engineers. (2018) Levee Portfolio Report Levee Safety Program,
Headquarters, Washington, D.C.
US Army Corps of Engineers. (2020) “National Levee Database” Website. Headquarters,
Washington, D.C. <https://levees.sec.usace.army.mil/#/>.
US Army Corps of Engineers. (2020) “Where We Are” Website. Headquarters, Washington,
D.C. <https://www.usace.army.mil/locations.aspx>.
Geo-Extreme 2021 GSP 330 108
© ASCE
Geo-Extreme 2021
Downloaded from ascelibrary.org by Farshid Vahedifard on 11/08/21. Copyright ASCE. For personal use only; all rights reserved.
