Need for Updating Alluvial Fan Floodplain Delineation Guidelines: A White Paper
Response to Comments
The Arid Regions Committee would like to sincerely thank all those to took the time to
offer comments on our White Paper (now called a “Discussion Paper”) on alluvial fan
floodplain delineation guidelines.
As of December 22, 2010, we believe we addressed all of the relevant comments at the
joint FMA-ASFPM Arid Regions Conference in Henderson, Nevada in November 2010,
and have revised the discussion paper accordingly. Unofficial meeting notes from the
FMA-ASFPM discussion sessions are attached as an appendix to this document. We have
also attached all of the review comments received to date. We do not believe that it
would be productive to provide a formal response to each and every comment received,
since most of the comments were addressed verbally at the FMA-ASFPM discussion
sessions. However, because some of the pre-conference review comments were
distributed to a wider audience, the following general responses are provided for
clarification:




Objectives. Two reviewers asserted that the main objectives of the paper were to
“maximize areas that can be developed,” “easing of FEMA’s current regulations,”
or to “minimize areas that require costly mitigation for development.” Those
assertions are incorrect, and in fact are opposite of the Committee’s objectives.
The objectives of the paper are clearly stated – to update the current alluvial fan
delineation methodology to more accurately reflect flood risk. The members of
the Arid Regions Committee have dedicated their careers to improving floodplain
regulations and effectively enforcing FEMA regulations.
Focus: One reviewer stated that the paper focuses “on just one Arizona flood
control agency,” was “produced at the behest of one flood control district,” and
“presents the agenda of only one Arizona flood control agency.” Those statements
are untrue. The paper is the product of the ASFPM Arid Regions Committee,
whose members represent communities throughout the western United States.
Input. One reviewer suggests that “collaborative input from all communities in”
Region IX is needed. The fact that these reviewers are commenting on the paper
is sufficient evidence that we are indeed seeking collaborative input. Some of the
Committee’s efforts to distribute the paper are summarized on the following
pages. In fact, the paper was reviewed and vetted by communities throughout
FEMA Region IX, as well as by communities outside of Region IX that are
affected by alluvial fan flooding. The Arid Regions Committee went to, and
continues to go to, great lengths to assure that all interested communities and
affected parties had the opportunity to comment.
Methodology. Several reviewers commented that the paper advocates for one
particular methodology. We do not agree. The paper merely calls for the
floodplain management regulatory community to improve the existing
methodologies. None of the recommendations reference a particular delineation
methodology. We look forward to discussing this issue in depth to better
understand these reviewers’ concerns

Deterministic vs. Stochastic Analysis. Several reviewers concluded that the
paper recommends that stochastic methods be abandoned in favor of deterministic
models. Again, we find no support for that comment in the text of the discussion
paper. In fact, the paper specifically calls out the importance of addressing flow
path uncertainty when considering alluvial fan flooding hazards. However, we do
note that many inherently stochastic processes (e.g., prediction of the 1%
discharge) are routinely evaluated using deterministic models (e.g., HEC-HMS).
Therefore, use of deterministic models to evaluate the flood risk associated with
flow path uncertainty should not be dismissed out of hand.
While we sincerely appreciate the comments, insights and passions the various reviewers
have brought to this conversation, we conclude that some reviewers have “jumped the
gun” and have commented not on content of the paper itself, but instead on anticipated
technical discussion items that may (or may not) arise if the recommendations of the
discussion paper are adopted. This discussion paper merely calls out the need to improve
the methodologies. If the floodplain management community agrees that improvements
are needed and are feasible, then a much longer, much more complex process of
identifying those improvements can begin.
We invite all interested parties to read the current version of the discussion paper,
consider it carefully, and to actively and cooperatively participate in the upcoming
discussions of the paper, so that we can all improve our ability to identify and regulate
alluvial fan flood risk.
Arid Region Committee Efforts to Solicit Input on the Discussion Paper
The White Paper has been an Arid Regions Committee Agenda item since 2008.
 2008 ASFPM Conference – Reno
o Committee Meeting – White Paper initially proposed.
 2009 ASFPM Conference – Orlando
o Committee Meeting – Update on white paper status.
 2010 ASFPM Conference – Oklahoma City
o Committee Meeting - Presentation of draft White Paper
o Early Bird Session – Presentation of draft White Paper
 Also on the agenda of every committee conference call & update since 2008.
Comments were formally solicited from the Arid Region Committee members on
6/10/2010 via email from Jeanne Ruefer. The email distribution list included:
jeanne.m.ruefer@accenture.com
genie@cwsd.org
mitchell.blum@hdrinc.com
robert.l.davies@amec.com
brian.fry@jefuller.com
tcl37cfm@gmail.com
bmacguire@rbf.com
ann.moynihan@rfcd.pima.gov
nelsonje@pbworld.com
patti.sexton@tetratech.com
mtruby@sjcounty.net
swaters@cals.arizona.edu
Dianna.woods@co.yakima.wa.us
Amanda.G.Wilson@aecom.com
lbond@labond.com
david.jensen@kimley-horn.com
becky@floods.org
craig.sellers@yumacountyaz.gov
gbrownell@mbakercorp.com
remmet_degroot@urscorp.com
jon@jefuller.com
trl@mail.maricopa.gov
dmallory@udfcd.org
tmm@mail.maricopa.gov
kimberley_pirri@urscorp.com
vas@mail.maricopa.gov
bvarrella@fcgov.com
westhoff@aquagroundwater.com
williams-lanza@reno.gov
Edward.Curtis@dhs.gov
patricia.rippe@dhs.gov
cadair@mt.gov
mbishop@mbakercorp.com
alex.coronel@amec.com
keubanks@regionalflood.org
tsgeorge@mbakercorp.com
scott@lyleengineering.com
barbara.mcevoy@idwr.idaho.gov
mandarnangare@mail.maricopa.gov
george@floods.org
mteal@westconsultants.com
tomwalker.jazz@gmail.com
john.wise@stantec.com
sranderson@pbsj.com
cadair@mt.gov
marc_mcintosh@urscorp.com
The White Paper was proposed a panel discussion topic for the November 2010
FMA/Arid Regions Conference at the first conference planning meeting on February 22,
2010. Those noticed for those meetings included:
Alejandro.Gutierrez@amec.com
'Brian Varrella'
'Desai, Harshal'
'Fox, Mike'
jeanne.m.ruefer@accenture.com
'Jon Fuller'
'Kim Groenewold'
Mark.Seits@hdrinc.com
'Mike Nowlan'
'Pineda, Ricardo'
robert.l.davies@amec.com
'Schaefer, Kathleen'
Steven.Mano@hdrinc.com
thomas@civilsolutions.com
'Vince Geronimo'
'Lenaburg, Raymond'
'Greg Jones - FCDX'
Gerry.Hester@snwa.com
khouse@unr.edu
Sergio.vargas@ventura.org
'Andrew Trelease'
Carol.L.Tyau@hawaii.gov
Eric.S.Clyde@us.mwhglobal.com
Gerry.Hester@snwa.com
jhays@co.santa-cruz.az.us
'Judy M Soutiere'
'Maria Lorenzo-Lee'
'Martin Teal'
'Pal Hegedus'
reese.lundgren@taic.net
rvillalobos@rbf.com
slongvil@csusb.edu
SWEBER@rbf.com
'Tom Smythe'
'Woolam, Susan'
'Massoud Rezakhani'
'Kathryn Gross - FCDX'
'Susan Longville'
lisaas@unr.edu 'Jon Fuller'
'Mike' 'Fox'
'Bob Battalio'
'Chris Bowles'
Eric.Simmons@dhs.gov
'Javier (Alex)' 'Yescas'
'John Wise'
'Kevin Eubanks'
Mark.forest@hdrinc.com
'Messano, Lisa'
Patti.Sexton@tetratech.com
'Ric Reinhardt'
'Sara' 'Agahi'
SParrish@regionalflood.org
'Syndi Dudley'
vas@mail.maricopa.gov
'Lisa Beutler'
'Plasencia, Doug'
keubanks@regionalflood.org
jprice@unr.edu
'Curtis, Edward'
jtlancaster10@hotmail.com
The FMA-ASFPM discussion session in November 2010 was very well attended, with
well over 100 participants in the session. Participants included representatives from
FEMA Region 9, Flood Control Districts from Southern California, Arizona, and
Nevada, members of the ASFPM Arid Regions Committee, members of the California
Alluvial Fan Task Force, and numerous private consultants. The discussions were
productive, with a clear consensus expressed for each of the discussion paper items (as
revised).
Appendix A: Comments Received
Comments from Ray Lenaburg/FEMA Region 9 (via email on 10/14/2010)
From: Lenaburg, Raymond [mailto:raymond.lenaburg@dhs.gov]
Sent: Tuesday, October 12, 2010 10:24 PM
To: George Riedel
Cc: jeanne.m.ruefer@accenture.com; robert.l.davies@amec.com; Curtis, Edward; Pietramali,
Ryan; Ike, Ryan; Chen, Wen; Norton, Beth; Schaefer, Kathleen
Subject: RE: Discussion Paper on Alluvial Fan Floodplain Delineation
Importance: High
George,
Thank you for your request to comment on the Discussion Paper and its impact on the
health of the National Flood Insurance Program (NFIP). I’ve been contacted by many
flood control district officials in Southern California, who have expressed concerns about
this paper, its focus on just one Arizona flood control agency, and its applicability to their
jurisdictions in Southern California. In fact, the Southern California agencies are faced
with the same pressures from engineering consulting firms assisting developers trying to
build homes in areas that are located in active alluvial fan areas.
FEMA Region IX has also been contacted by many engineers and geologists in Southern
California Counties, and they are also of the opinion that the flow path uncertainty issues
cannot be handled by deterministic methods and that the avulsion mechanism is very hard
to predict in practice. To summarize their concerns,
1. Even though other things are mentioned in the Discussion Paper, the focus is really on
minimizing areas subject to flow path uncertainty and maximizing areas that can be
developed without the flood risk mitigation requirements in 44 CFR 65.13(c).
2. Improving the science of avulsion frequency is a good idea, but will be very hard to
accomplish in practice.
3. New age dating techniques do indeed exist, but are extremely expensive and generally
not practical for use by industry at this time.
4. FEMA’s existing protocols provide a buffer on the numerous uncertainties that surround
alluvial fan flooding (not only flow path uncertainty). It might be one thing to run 2-D
models if we really had a good handle on the 100-year storm event, but how well is this
really known in most watersheds? Also, how well does anyone really understand 100year sediment production? And, what if a debris flow occurs at the beginning of a flood,
changing flood pathways that were previously mapped in the 2-D topographic model?
5. It seems to me that any improvement in the procedure has to involve parallel advances
in understanding the entire flow-frequency relationship, 100-year storm runoff, 100-year
sediment production, surface age dating, understanding the frequency of debris flows
and flow modeling, not just the latter.
I grew up in Southern California and have worked for over 40 years in Federal and local
government agencies dealing with flooding and disasters and their consequences. I’ve
seen first-hand devastation resulting from underestimating flood hazard risks. Almost
any of the Southern California communities will tell you what their experiences have
been with alluvial fan flooding. For example in the Coachella valley, developers want to
remove flood zone designations developed by FEMA’s FAN analyses in the Oasis area
by any means to build homes – build out and get out before their flood protection is
meaningfully tested.
Regarding the Discussion Paper’s stated concern about the health of the NFIP, I will
repeat what is provided in detail below: The “accurate delineation of flood hazards” in
the context of the Discussion Paper implies decrease of active alluvial fan surfaces that
are mapped as high hazard areas. If the true objective is the health of the NFIP, then
more and not less alluvial fan surfaces should be mapped as high-hazard AO Zones.
FEMA’s engineering consultants and I have reviewed the Discussion Paper, and offer the
following commentary with constructive suggestions.
Introduction. The Discussion Paper reinforces the concerns expressed in the NRC 1996
document, “Alluvial Fan Flooding”[1] that development pressure will try to reduce special
flood hazard areas and costly mitigation requirements and maximize areas where
elevation of fill, or less, can be offered to mitigate the underestimated risk. The
Discussion Paper further illustrates that the cautions in “Alluvial Fan Flooding” have
been ignored or sidestepped by offering deterministic mathematical modeling (2dimensional in this case) and geomorphic analysis in active alluvial fan areas where
elevation on fill will not reliably mitigate flood risk.
“Alluvial Fan Flooding” states time and time again that some areas on alluvial fans with
uncertain flow path flooding cannot be reliably analyzed with deterministic models.
Deterministic models applied to the existing condition alluvial fan topographic surface
provide only a snapshot based on pre-flood topography, of the surface that represents the
sum of past floods. On active alluvial fan surfaces, the flood itself changes the
topography by erosion, deposition, and avulsion processes that are themselves
unpredictable. “Alluvial Fan Flooding” clearly states: “For risk assessment under
alluvial fan flooding, existing channels cannot be relied on to convey the 100-year peak
flow, so their role is ignored. For riverine floodplain management, however, the
channels are significant.“ Further, “. . .floodwater surface elevations computed using
preflood topography are not a comprehensive indicator of true hazard for alluvial fan
flooding situation as they are for riverine flooding.”
FEMA’s policy remains that stochastic analysis methods (e.g. “FAN”) are the only
acceptable methods to assess flood risk in areas “. . .characterized by flow path
uncertainty so great that this uncertainty cannot be set aside in realistic assessment of
flood risk or in the reliable mitigation of the hazard.” FEMA’s hazard mitigation policy
is clearly set forth in 44 CFS 65.13(c) which requires sound engineering practices to
effectively eliminate alluvial fan flood hazards by removing flow path uncertainty.
Implementation of FEMA’s policy implies that the cost of flood risk mitigation will be
borne by the developer, not the flood control agency or the public. This last statement is
the driving force behind the Discussion Paper – to minimize areas that require costly
mitigation for development. By the way, it would be nightmare for the regulatory
agencies and people who buy homes built on alluvial fans. Until we know more about the
physical processes on the fan during an extreme or design flood event, we simply have to
[1]
National Research Council, 1969. “Alluvial Fan Flooding”.
be conservative to protect the public, the flood protection agencies, and FEMA. The
Federal and local governments have a duty to place public health and safety first.
Summary. The Discussion Paper was reviewed in anticipation that it would provide
collaborative, balanced, and inclusive information related to fundamental improvement in
knowledge of alluvial fan flow processes – particularly the parts of alluvial fan surfaces
that exhibit flow path uncertainty (active alluvial fan surfaces). In the context of
“floodplain delineation”, it was anticipated that FEMA’s Base Flood would be
mentioned. The Discussion Paper exhibits none of these attributes, but instead presents a
point of view biased toward wholesale reduction of Special Flood Hazard Areas by
underestimating alluvial fan flood hazards. Further, the Discussion Paper presents the
agenda of only one Arizona flood control agency as interpreted in the context of
deterministic two-dimensional modeling and deterministic geomorphologic analysis. It is
suggested that the Discussion Paper be rewritten with additional collaborative input from
all communities in RIX (CA, AZ, NV) and other Regions within the Western states
before being submitted as the policy of the ASFPM or FMA.
FEMA’s Regional office staffs, along with its engineering contractors, have been
involved in alluvial fan floodplain analysis mapping for decades. Our collective
evaluation of the mapping problems and attempted solutions point out the principal
unknown factor in the flood mapping process is “flow path uncertainty”. When “flow
path uncertainty” is combined with the inherent and unavoidable statistical uncertainties
in hydrology, the analysis and mapping process becomes very complex, and are
overwhelmingly dominated by stochastic processes that cannot be easily addressed by
deterministic analyses. In fact, “flow path uncertainty” means that a natural system’s
subsequent state is determined by a combination of unpredictable actions and elements of
randomness, it is innately non-deterministic.
Over the years, we have tried to narrow the random elements and reinforce the understanding
of predictable actions. FEMA’s 2003 Guidelines and Specifications (Appendix G, Guidance for
Alluvial Fan Flooding Analyses and Mapping) which took several years to develop, describes the
methodology for mapping alluvial fan surfaces. FEMA explicitly identifies the “FAN” model
approach as the method to use for active alluvial fan surfaces where stochastic processes
predominate. This Discussion Paper appears to offer another method which is not included in
FEMA’s Guidelines and Specifications.
FEMA’s FAN handles avulsion, erosion, sedimentation, flow path uncertainty, and it considers
the dynamic environment, such as alluvial fan flooding, to produce AO flood insurance Zones
with defined depths and velocities. The AO Zone with depths and velocities is the method that
FEMA uses to identify to the public areas that are subject to alluvial fan flooding, and is also
used for flood insurance rating purposes.
FEMA and the affected communities would not be well served by a methodology that
decreases Special Flood Hazard Areas because the results could conflict with FEMA’s
adopted mapping process and lead to inconsistent land use planning by affected
communities. FEMA’s methodology and FAN are more reliable in areas of “flow path
uncertainty”, are conservative, and are defendable in Court.
The affected communities should be particularly sensitive to this last point, as
underestimation of flood risk can lead to fatalities and costly reparations. A case in point
is the Paterno Decision, which cost the State of California about $500 Million. The
Paterno Decision included a requirement for a “reasonable plan” for flood protection. In
general, applying FEMA’s methods to assess flood risk is legally defensible and passes
the “reasonableness test”. Applying a methodology that underestimates flood risk is not
likely to pass such a test.
For years our FEMA staff along with our engineering contractors, have looked into the
technical merits of geomorphology and 2-dimensional modeling, compared with FEMA’s
FAN methodology. Deterministic methods share the limitation that they cannot address
all the unknowns related to “flow path uncertainty” on alluvial fan surfaces. 2-D
modeling, in particular the 2-D model that appears to be proposed to the exclusion of
other 2-D models, is no panacea, as it does not directly address avulsion, erosion, or
deposition. River basin geomorphologic forms “…are stochastic processes, or processes
governed by laws of chance.” Age dating techniques, in the absence of Bulletin 17B
paleohydrologic analysis, are disconnected from the alluvial fan flow-frequency
relationship and FEMA’s Base Flood. Increasingly complex 2-D modeling and
geomorphologic analysis have been offered by proponents of the methodology in this
Discussion paper. The proponents have alleged that their methodology covers all the
eventualities of alluvial fan flow, resulting in demonstrable decreases in alluvial fan flow
discharges, areas subject to flooding, and excluding probability theory. This reasoning is
fundamentally flawed. “If the causative factors are random variables, which they are
more often than not in hydrology, these cause-effect deterministic relationships…are
nothing else than functions of random variables.”
FEMA’s alluvial fan flow methodology, as evaluated and refined by the 1985 DMA
study adequately analyzes “flow path uncertainty”. The DMA study results have been
implemented in the FEMA FAN model. FEMA has not seen any recent breakthroughs in
the technology of predicting avulsion, erosion and disposition implied in the Paper. It is
FEMA’s understanding that the science of water and sediment transport haven’t
fundamentally changed since Hans Albert Einstein, except that computers make applying
the complex modeling equations easier to solve than before.
Overall: The document and its recommendations focus on the easing of FEMA’s current
regulations and guidance. There is no reference to the possible impacts to the safety of people
and property. It appears that if the recommendations of the document were adopted it would
allow more development within high risk areas of alluvial fan flooding than there would be
allowed currently. There is no explanation of the potential adverse impacts of easing regulation
on development within these hazardous areas, or who would ultimately be responsible for the
cost of any potential negative impacts.
Many of the recommendations present one side of issues that are still being actively debated.
Most of these ongoing debates will not be concluded until there is sufficient evidence to support
a consensus viewpoint. The document and its recommendations clearly demonstrate the need
for analyses and justification to support any recommendations made with respect to these
important issues.
It was noted that Recommendation #4 argued against the prohibition of using
[deterministic] “hydraulic analytical methods” on active alluvial fans, contrary to FEMA
policy in FEMA (2003)[2] Appendix G. In many other locations, it was noted that twodimensional modeling was implied in the context of modeling active alluvial fan flows, to
the exclusion of “FAN”. Further, it was noted that two-dimensional modeling, and not
one-dimensional modeling, was recommended for modeling inactive alluvial fan flows.
In regions of flow path certainty, either one-dimensional or two-dimensional
methodology with proper boundary conditions would produce similar results.
The ASFPM document that details the creation of this Discussion Paper
(http://www.floods.org/acefiles/documentlibrary/committees/Arid_Regions_WorkPlan_2010.pdf) outlines that a portion of
the original intent of this document was to produce development guidelines for alluvial fan
areas. That would seem to imply developing and communicating a clear understanding of
current FEMA regulations and guidance. However that does not appear to be included in the
current document, which appears to focus mainly on the need to reduce regulation of
development in these areas.
Introduction: The “accurate delineation of flood hazards” in the context of the
Discussion Paper implies decrease of active alluvial fan surfaces that are mapped as high
hazard areas. Health of the NFIP, on the other hand, would be decreased by reducing
areas mapped as high-hazard AO Zones. If the true objective is the health of the NFIP,
then more and not less alluvial fan surfaces should be mapped as active.
History of Success: There are examples where the methodology proposed in this paper
has not resulted in conservative or even accurate depiction of alluvial fan flood hazards.
See Figure 1 and 2 attached at the end of these comments, showing active flow paths
passing into and through an area determined by 2D modeling to be “inactive”.
It was noted that only two of the three basic references on alluvial fan flow were
mentioned: Dawdy (1979)[3] and NRC (1996)[4]. The Discussion Paper skipped the
intermediate paper that provides the computational basis linking Dawdy theory to FAN
implementation: DMA (1985)[5]. Instead of comprehensive clarification, it was noted
that there was a conspicuous absence of any mention at all of FEMA (1990)[6] “FAN”
stochastic mathematical model for floodplain delineation on active alluvial fan surfaces.
This section provides the impression that NRC (1996) has constrained active alluvial fan
surfaces to the point that two-dimensional modeling, to the exclusion of any other
modeling methodology, could be used exclusively for floodplain delineation, and that
two-dimensional modeling would be appropriate for active alluvial fans.
[1]
FEMA, 2003. “Appendix G: Guidance for Alluvial Fan Flooding Analyses and Mapping”.
[2]
Dawdy, D.R., 1979. “Flood Frequency Estimates on Alluvial Fans”, Journal of the Hydraulics Division,
ASCE, Vol. 105, HY11, Proc. Paper 15001, p. 1407-1413.
[3]
National Research Council, 1969. “Alluvial Fan Flooding”.
[4]
DMA, 1985. “Alluvial Fan Flooding Methodology – An Analysis”, FEMA EMW-84-C-1488, July.
[5]
FEMA, 1990. “FAN – An Alluvial Fan Flooding Computer Program”, Risk Studies Division, Office of Risk
Assessment, September.
Item (3) geologic data as applied in the context of this paper removes the flow of water from the
picture. The Base Flood is not linked to geomorphologic findings. The science of Stochastic
Geomorphology is well developed but ignored in this paper. “Erosion is highly stochastic in time
and space because it is strongly influenced by rainfall…” (Benda, 1994). “For example,
hydrologists can make relatively accurate predictions of changing peak flows in a watershed
using new…models. Using this information, geomorphologists can make only relatively
inaccurate predictions of sediment transport and cannot say anything definitive regarding bed
scour or changes in channel form…” Moglen and Beighley, 2002).
Recent Developments. This Paper was produced at the behest of one flood control
district, and nationwide applicability of its guidance and recommendations should only be
considered based upon the concurrence of the communities that would be affected.
Need to Update. . . . Nothing in FEMA’s procedures prohibits accurate depiction of
flood hazards. It is the duty of the community to insure that flood hazards are accurately
depicted by conservative application of existing guidelines. Third bullet - It should be
considered that over this same timeframe, while the current regulations were being
enforced, and the current methodologies implemented, there continue to be catastrophic
floods that damage property and endanger the lives and safety of the general population
within these areas. There also continues to be significant and costly legal settlements
associated with this type of flooding. With that in mind, it should be clearly explained
how a reduction, and not increase, in the protections currently afforded is warranted. It
should be shown how the current methodologies and regulations have failed at protecting
persons and property, and how the recommendations will correct these shortcomings.
New and Improved Tools – Two Dimensional Modeling. As noted elsewhere, the
implication is that deterministic two-dimensional models can be applied to active fan
surfaces. Deterministic two-dimensional models cannot calculate the physical processes
of alluvial fan flow which carves its own path(s) over the fan surface and also suddenly
relocates the path(s) by avulsion. Where there is flow path uncertainty, only stochastic
methods (FEMA “FAN”) can be applied.
New and Improved Tools – Prediction and Modeling Tools. It is implied here and
elsewhere that new and improved tools have reduced flood threat uncertainty and can
somehow predict future flood events. New and improved tools might reduce, but by no
means eliminate, areas of flow path uncertainty. Hydrologic uncertainty remains,
regardless of tools. Flow path uncertainty does not vanish by ignoring “…the statistical
nature of the problem”[7].
New and Improved Tools – Geologic Dating Techniques. Dating techniques establish
past flooding with no context to FEMA’s Base Flood. Past flooding must be considered
in the context of historic or paleo-flood events with date(s) and magnitude(s) in
accordance with Bulletin 17B (USGS (1982)[8]). To provide insight as to future flooding
[6]
[7]
Gumbel, E.J., 1958. “Statistics of Extremes”, Dover.
USGS, 1982. “Guidelines for Determining Flood Flow Frequency”, Bulletin 17B of the Hydrology
Subcommittee.
probability, past flood events must be assigned an exceedence probability. Past absence
of flooding is no guarantee of future absence of flooding.
Recommendation #1. For active alluvial fan flooding the designations are AO w/depth and
velocity. For inactive areas it would appear that BFEs or AO with depths would apply. We need
to distinguish debris generated inside the steep canyon which is brought down by the flood
flows onto the area near the apex of the fan. This is the main mechanism of avulsion. Because
these materials are usually fairly large, they remain near the apex (e.g. Deep Canyon in Sept of
1976). There is another source of debris which is generated on the fan surface by the flowing
floodwaters. By definition, an alluvial fan is relatively erodible. That is why it is dynamic process
coupled with avulsion. Erosion-generated debris will rapidly diminish in the downstream
locations. The basic problem is that we will never see the actual condition afterwards unless we
have instrumentation such as buried scour cords, etc. in the fan surface beforehand. We will
need a lot of field work and investigation.
Recommendation #2. AO zones have varying depth and velocity combinations. It is not
apparent how geomorphic methods applied by themselves could result in an A Zone – may need
further explanation. It should be noted that A Zone designations for alluvial fans should be used
with care since the flood insurance rates are the highest and future development may think they
can get a LOMR-F.
It is acknowledged that the current hazard zones given by the FEMA FAN model need
refinement. That is why FEMA and good engineering practice include a site reconnaissance and
walk-down to map the fan boundaries to ascertain actual contour widths. There could be high
ground or ridges within a fan surface which would never be flooded. These might be out of the
Zone A, but would still affect contour widths used with FAN.
Recommendation #3. Sheet flow areas are not necessarily low hazard areas. It all depends
where the sheet flow locations are relative to the channelized flow area. The key issue is to be
able to clearly define the demarcation of the channelized and sheet flow area. Assuming that in
“real” sheet flow area far away from channelized flow area, the erosive power of the flow is
greatly diminished so it might be possible to use a 2-D model to estimate the flow depth and
velocity for design. The 2-D model upstream boundary would have to be at the demarcation
line between channelized flow and sheet flow zones. The flow onto the sheet flow zone would
have to consider dynamic effects with equal probability that it could occur at any location along
this demarcation line. The key issue appears to be the definition of the demarcation line where
"real" sheet flow begins. The 1985 DNA report needs to be updated to include more recent field
observations. This will be difficult, as the field observations are taken after the fact. The
principal difficulty with Fan Flood Hydrology is not getting good data during the passage of the
flood as it occurs.
Separate definitions of high and low threshold do not appear to be mutually exclusive.
Re elevation on fill, this appears to be a risky proposal on active fans. Risk mitigation should be
substantiated within the text, or supporting documentation or data should be referenced.
The quoted words are the language currently in place; it should be shown how the removal of
this language would better protect persons and property.
Recommendation #4. Improvement of technical guidance is easier said than done as alluvial fan
hydrology, like flood hydrology, is site specific. It is difficult to provide written technical
guidance because there is no one-fit-all process and every alluvial fan is unique. Hydrologic
analysis should never to be a cook-book one.
Re the prohibition of using “hydraulic analytical methods” on active alluvial fans, it is not
contradictory. There are individuals within the engineering community and within FEMA who
currently provide clear justification as to why this is the case. Their input appears to have been
overlooked in the drafting of this Discussion Paper. This Discussion Paper outlines the
recommendations of one side of an apparent disagreement, and does not address the
underlying conflict of opinion.
Flow path uncertainty does not vanish by applying deterministic “hydraulic analysis
methods” on active fan surfaces, no matter how detailed or complex the deterministic
methods are. “Outcomes in observation of a random variable [say avulsion] are affected
by an extremely large number of causative factors. Determinists in natural sciences
sometimes contend that the search for deterministic relations will include with time more
and more causative factors to explain effects on a random variable. Following this
reasoning, it could be expected with time to continuously extend the deterministic
explanation and description of present-day hydrologic random phenomena and
substantially restrict the application of probability theory in hydrology. This reasoning
has a fundamental weakness. A finite number of causative factors taken into any
deterministic equation could rarely exhaust all causative factors involved in the outcome
of a hydrologic random phenomenon. If the causative factors are also random variables,
which they are more often than not in hydrology, these cause-effect deterministic
relationships, as stated previously, are nothing else than functions of random
variables.“[9]
Re deficient submittals, this is a good point and brings to light that local communities signing off
on CLOMR/LOMR prior to sending to FEMA and their need to know what is and isn’t accepted.
Part of the confusion might be alluvial fans mapped as A zones preceding a CLOMR/LOMR. As it
is being implied that current LOMR submittals do not meet the current requirements, that
would seem to imply that there is neither a current consensus or clear understanding of what
the requirements of the NFIP for this sort of development are. This issue could be addressed by
obtaining that consensus and educating the community.
Recommendation #5. Re flow attenuation, justification should be provided or referenced.
Exactly how the same level (or an increased level) of risk reduction would be achieved through
the implementation of this recommendation should also be included. As worded, this appears
to indicate that the calculated flows are less conservative.
FEMA’s FAN theory and application is based on steady flow, while 2D modeling as presented in
this Paper is based on unsteady flow. 2D modeling can exploit this difference alone to produce
a decrease of flow down the fan. The amount of decrease depends on the shape of the
hydrograph and depth and width of the flow path. High intensity, short duration rainfall would
produce the greatest apparent dissipation using 2D unsteady flow modeling.
Alluvial fan flows do not necessarily spread out and dissipate, but carve channels that can be
narrower than the width of the fan surface, and narrower than the topographic depressions that
2D models rely on to define the flow paths and convey flows.
[8]
Yevjevich, V., 1997. “Probability and Statistics in Hydrology”, WRP.
Inappropriate use of 2D modeling can produce inappropriate results that exploit high-intensity
short-duration hydrograph and topographic spreading and dissipation of the flow by internal
loss schemes.
Allowing infiltration on the fan may result in understating the risk for different types of
storms. For instance it is raining on the fan surface creating run-off, then the storm
spreads to the watershed feeding the apex – would the apex Q’s be stored on the fan
surface? If the flood flows are confined in a channel as it would be in single channel
region, the attenuation would be small. It is reasonable to allow for some storage in the
multiple channel region. Then, one would have to be very careful as to where the
demarcation line is between the two regions. The 1985 DMA figure related to
bifurcation point might need to be updated. If on-fan precipitation is included, there
would not be as much flow attenuation. If losses are recommended to be considered,
then gains must also be considered.
Re “setting aside” uncertainty, again, it appears that others in the engineering community and
at FEMA have voiced conflicting viewpoints. The sheet flow region might be best handled by 2D
model as the flow paths would be more or less following the topography and more certain, but
only if uncertainty is removed from the upstream boundary conditions. The upstream boundary
condition for 2D modeling need to be carefully considered in light of flow path uncertainty and
probability that unattenuated apex discharge could impinge onto 2D model in unpredictable
location(s).
Recommendation #6. This portion of the recommendation, aside from the advances in
technology, would be beneficial if it was specifically based upon promoting the understanding of
a clear explanation and practical application of current FEMA guidance and regulations. Regular
training should be devoid of any commercial and business interests, or self promotions. We
have seen again and again that certain individuals are using training as a vehicle for selfpromotion for a proprietary model which may not necessarily apply to the evaluation of flood
hazards on alluvial fans. How to do so will be a challenge. Many of communities have expressed
concerns over this issue to Region IX office.
As current regulations prohibit the use of two dimensional models on active fan areas, some
clear justification (with consensus) should be provided to make this recommendation. If we
recommend these methods and they are not allowed, we may be sending an unclear message.
Recommendation #7. This is a jurisdictional problem with local agencies.
Recommendation #8. The study that is recommended is the exact type of justification that
should be referenced to support the recommendations of this Discussion Paper. Unfortunately,
as this recommendation suggests, it appears that the necessary studies to justify the
recommendations have not yet been performed. This recommendation implies that the risk of
avulsion can be accurately and reliably assessed, however there appear to be conflicting
opinions within the engineering community that may not have been considered. The default
recommendation in FAN is that an avulsion event occurs every other Base Flood event, or P =
0.5 and avulsion coefficient A = 1 + P or 1.5. It is implied that the probability of an avulsion can
take values other than 0.5, but no guidance is provided. It is a simple matter to do multiple FAN
analyses with P from 0 to 1 to investigate the entire range of frequency of alluvial fan avulsions.
Recommendation #9. The study that is recommended is the exact type of justification that
would help the local agencies and communities obtain a good understanding of the flood hazard
Zones and the insurance rates applied to them. For instance, the insurance rates in A Zones are
very expensive, in an A zone if the Finish Floor is elevated 5 feet above adj. grade the rate is still
a little more than a FF elevated at or above a Zone AO.
The effort so far seems to be focused more on extracting “inactive” fan surfaces for
development without interference from FEMA, rather than looking at the whole fan surface or
whether “active” fan surfaces might ever encroach into other areas. This is complicated by use
of certain geomorphologic and proprietary modeling tools that give a false sense of security.
Overall, the statement about understanding fan flooding is correct, however none of the
recommendations stated previously clearly define how a better understanding would be
achieved through their implementation.
Figure 1. Alluvial fan areas.
Figure 2. Comparison of 2D flow modeling flood boundaries (blue lines) with active
flow paths determined by past flooding (1977).
1 National Research Council, 1996. “Alluvial Fan Flooding”.
2 FEMA, 2003. “Appendix G: Guidance for Alluvial Fan Flooding Analyses and
Mapping”.
3 Dawdy, D.R., 1979. “Flood Frequency Estimates on Alluvial Fans”, Journal of the Hydraulics Division,
ASCE, Vol. 105, HY11, Proc. Paper 15001, p. 1407-1413.
4 National Research Council, 1996. “Alluvial Fan Flooding”.
5 DMA, 1985. “Alluvial Fan Flooding Methodology – An Analysis”, FEMA EMW-84-C-1488, July.
6 FEMA, 1990. “FAN – An Alluvial Fan Flooding Computer Program”, Risk Studies
Division, Office of Risk Assessment, September.
7 Gumbel, E.J., 1958. “Statistics of Extremes”, Dover.
8 USGS, 1982. “Guidelines for Determining Flood Flow Frequency”, Bulletin 17B of
the Hydrology Subcommittee.
9 Yevjevich, V., 1997. “Probability and Statistics in Hydrology”, WRP.
Regards, Ray
Raymond T. Lenaburg
Chief, Risk Analysis Branch
Mitigation Division
DHS-FEMA Region IX
1111 Broadway, Suite 1200
Oakland, CA 94607
(510) 627-7181
Comments from Georgia Celehar-Bauer/Coachella Valley WCD (via email 9/9/2010)
(Comments in MS Word)
Hi Jon,
Please excuse the delay getting comments back to you. I sent the paper for review and
comment to Sam Hui with Bechtel (or former consultant of over 30 years) – Sam and his
team were instrumental in preparing and reviewing H&H studies and flood control
schemes for the Coachella Valley, most of which were on alluvial fans. The attached
document has both our comments (Sam is in Red).
Thank you for allowing us to review and comment on the paper.
Sincerely,
Georgia
Need for Updating Alluvial Fan Floodplain Delineation Guidelines: A White Paper
By Jonathan Fuller, PE, RG, CFM
ASFPM Arid Regions Committee
Introduction
Up to one third of the remaining developable land area in western United States lies on
alluvial fan landforms (Add REF). Therefore, accurate delineation of flood hazards on
these unique landforms is critically important to the health of the National Flood
Insurance Program. The goal of this white paper is to encourage ASFPM member
communities and FEMA to continue their long history of continuously improving tools
for delineating and managing alluvial fan floodplains.
A History of Successes
Hazards associated with alluvial fan flooding were brought into the national spotlight
following catastrophic debris flows on alluvial fans near Rancho Mirage, California in
the 1970’s. These disastrous floods highlighted the fact that alluvial fan flooding was
substantively different than riverine flooding, and therefore required different mapping
and management tools. FEMA responded by implementing a probabilistic methodology
developed by the USGS (Dawdy, 1979). The so-called Dawdy method offered two key
improvements in alluvial fan floodplain delineation technology. First, the method
recognized that alluvial fan flooding is very different from riverine flooding. Second, the
method recognized that uncertain flow paths are a key aspect of alluvial fan flooding.
From 1979 to late 1980’s, the Dawdy method was used to map alluvial fan floodplains
throughout the West. The Dawdy method is now implemented as the FAN model and is
described in Appendix G of the FEMA Guidelines.
In 1995, FEMA contracted with the National Research Council (NRC) to evaluate their
alluvial fan floodplain delineation practices. After deliberating for nearly a year, visiting
numerous alluvial fans, and interviewing dozens of communities, the NRC technical
committee issued their report Alluvial Fan Flooding (1996), which suggested a number
of improvements to FEMA’s alluvial fan floodplain delineation methodology. The NRC
recommendations included the following: (1) recognition that not all alluvial fans are
alike, (2) distinction of active and inactive alluvial fan landforms, (3) incorporation of
geologic data into the fan identification and mapping process, and (4) use of a three stage
delineation process. FEMA adopted many of the key NRC recommendations and codified
them in the NFIP Regulations as well as in Appendix G of the FEMA Guidelines.
Recent Developments
Since publication of the NRC report in 1996, and the revised FEMA Guidelines in 2002,
the three-stage alluvial fan floodplain delineation process has been applied to alluvial fan
landforms throughout the United States. The Flood Control District of Maricopa County
(FCDMC) published its draft Piedmont Flood Hazard Assessment Manual (PFHAM) in
1998, which still stands as the benchmark for implementation of the FEMA three-stage
alluvial fan delineation process. More than 30 alluvial fan floodplain delineations have
been completed in Maricopa County alone using the FCDMC manual. More recently, the
State of California commissioned an Alluvial Fan Task Force, which adopted an
integrated planning and analysis approach which closely parallels the approach
recommended in the NRC Report. FEMA has also recently transitioned from MapMOD
to RiskMAP. Some of the main themes of the RiskMAP program include: (1) better
technology, (2) better mapping, (3) risk-based mapping, and (4) use of multi-disciplinary
approaches.
Need to Update Alluvial Fan Floodplain Delineation Techniques
Given FEMA’s call for better technology and mapping as part of their RiskMAP
program, and the experienced gained by local communities like the FCDMC and the
State of California in alluvial fan floodplain delineation, the Arid Regions Committee
now calls on its members to encourage FEMA to update its alluvial fan floodplain
delineation procedures. ASFPM’s report, National Flood Programs and Policies in
Review (2007), specifically noted that the existing alluvial fan floodplain delineation
techniques inaccurately depict the actual flood hazard. The following are reasons that
now is the time for such an update:

Time. It has been 14 years since publication of the NRC report and eight years since
FEMA adopted the current version of Appendix G. We have learned much over the
past 14 years that could be applied to alluvial fan flood hazard evaluations. Every
technical methodology needs periodic updates. The alluvial fan floodplain
delineation methodology is no exception.

Experience. The current three-stage alluvial fan delineation methodology has been
applied many times over the past decade. As with any new procedure, the more it is
applied, the better we know its short-comings and needs. We should leverage our
experience to shore up existing gaps and weaknesses.

New and Improved Tools. The engineering tools available for evaluation of alluvial
fan flooding have been improved, and new tools have been developed in the past 14
years. These tools include improvements in two-dimensional modeling, new geologic
dating techniques, and new debris flow prediction and modeling tools.
Management & Mitigation Issues
An update of FEMA Appendix G would also be consistent with the larger goals of the
NFIP. Better floodplain delineation tools for active alluvial fans not only enables NFIP
member communities to better manage their flood hazard, but also it provides
opportunities for FEMA to determine whether the hazards are properly rated for
insurance purposes. Furthermore, better hazard identification tools for alluvial fans will
allow FEMA and NFIP member communities to deal directly with coverage for erosion
and debris flow hazards which occur on active alluvial fans.
Recommendations
The Arid Regions Committee strongly believes that we should build on FEMA’s long
history of continuous improvement of their alluvial fan floodplain delineation tools, and
therefore offer the following recommendations for updating the current alluvial fan
floodplain delineation methodology:
Recommendation #1: The methodology should recognize that there are different types
of active alluvial fans. Currently, Appendix G recognizes that there are different types of
flooding on alluvial fan landforms (G.2.2.4), but the methodology does not include any
finer distinctions among types of active alluvial fans. Furthermore, the current
methodology does not distinguish different types of flood zones that correspond to the
differing types of flooding on alluvial fans. Specifically, the guidelines should
distinguish the following:
 Debris flow fans vs. fluvial fans. Debris flow fans have significantly greater flood
hazards than fans subject only to water floods.
 Channelized flow vs. sheet flooding. Fans subject to avulsive, channelized flow have
significantly greater flood hazards than fans which are dominated by shallow sheet
floods.
Recommendation #2: The methodology should distinguish high hazard portions of fans
from lower hazard portions of fans. Currently, the effective methodology does not
adequately distinguish degrees of hazard within an alluvial fan floodplain. Either the
hazards are mapped a parallel bands of equivalent hazard (FAN model) or the entire
alluvial fan floodplain is mapped as an unnumbered A Zone (Geomorphic method),
giving the impression that the hazard level is equal throughout the entire 100-year
floodplain. We now know that the degree of hazard is a function of watershed
characteristics (drainage area, flow rate, flow frequency, sediment production, wild fire
risk, debris flow potential, etc.), fan surface characteristics (slope, sediment size, fan area,
topographic relief, etc.), and the hydraulic characteristics of flow (depth, velocity, etc.).
The following types of hazard levels should be distinguished on active alluvial fans:
 Debris flow hazard areas (See Recommendation #1)
 Avulsion risk areas
 High flood depth & velocity areas
 Shallow sheet flooding areas
Recommendation #3: The terminology used in Appendix G and the NFIP should be
clarified and made internally consistent. An example of contradictory or confusing
language is in the definition and descriptions of active alluvial fans. According to the
language in Appendix G, active alluvial fans can either be areas in which “elevation on
fill would not reliably mitigate the flood risk” (high threshold), or areas of sheet flooding
(low threshold), or areas that would have been inundated in the past 100 or 1,000 years
(low threshold). Areas of shallow sheet flooding on fluvial, non-debris flow fans,
experience low flow depths and velocities. In such areas, it is likely that elevation on fill
would reliably mitigate the risk, even though the fan may be “active” in geologic time
(i.e., aggrading, albeit slowly). In addition to clarifying such potential contradictions, the
text should be edited to remove, to the extent that is feasible, non-quantitative adjectives
from definitions, such as “high,” “abrupt,” “so severe that,” etc.
Recommendation #4: Improve the technical guidance for delineation of active alluvial
fan flood hazards. Specifically, better guidance on what constitutes a composite method
(Table G-1) is needed, as well as how use of a composite method fits with the apparent
(and contradictory) prohibition of using “hydraulic analytical methods” on active alluvial
fans. In addition, more detailed description of the types of analyses required to support a
LOMR request for an active alluvial fan floodplain. Many LOMR/CLOMR submittals
are woefully deficient because of the lack of clarity regarding what is required, resulting
in multiple reviews and frustrated applicants.
Recommendation #5: The methodology should formally recognize specific key
processes that are known to occur on active alluvial fans. These processes include the
following:
 Flow attenuation between the hydrographic apex and the toe of the fan.
Because of this attenuation, use of the full apex discharge downstream of the
apex is overly conservative, particularly near the toe of an alluvial fan. Also,
because of the degree of flood storage that occurs on active alluvial fans, the
impact of development in such storage areas drastically increases peak
discharges, adversely impacting downstream areas, a fact which is not
adequately captured by using the full apex discharge.
 Sheet flooding is the dominant flood processes on the mid- and distal-fan
areas of many active (fluvial) fans. Use of floodplain delineation models that
do not recognize the presence of sheet flooding will inaccurately depict the
true flood hazard. These sheet flooding areas are subject to flow path and
flow rate uncertainty, but generally do not present the same level of risk as
channelized flooding and/or debris flow hazards. That is, any uncertainty in
sheet flooding areas can “be set aside” in the assessment of the hazard.
Recommendation #6: Conduct regular training to increase awareness of alluvial fan
floodplain hazards and keep abreast of advances in technology. This training should be
targeted at FEMA and local community floodplain reviewers, floodplain management
agencies, and development engineers. The target audience should be taught to recognize
alluvial fan landforms, as well as how to delineate flood hazards on active alluvial fans.
In addition, training in the use of new tools needed to quantify active alluvial fan flood
hazards is needed, such as two-dimensional modeling, geomorphic mapping, and
recognition of debris flow hazards.
Recommendation #7. Initiate a consistent review process to assure that the presence of
alluvial fans is not missed in the review and floodplain delineation process. This could
be applied only in communities known to have active alluvial fans or could be applied on
a broader scale. Currently, there is a high potential for alluvial fan flood hazards to be
missed in the delineation and review process. Conversely, there have been a number of
well-documented cases where the threat of alluvial fan flooding existed, but was missed
entirely in the delineation and review process.
Recommendation #8. Investigate the frequency of alluvial fan avulsions. FEMA and
NFIP member communities should specifically investigate the physical mechanisms that
lead to alluvial fan avulsion, and quantify the frequency with which avulsions occur on
debris flow and fluvial fans. This investigation will determine whether avulsion risk is
sufficient to warrant consideration in floodplain delineation and engineering mitigation
evaluations. This investigation could also be used to assess the accuracy and applicability
of statistical/probabilistic approaches.
Recommendation #9. FEMA should explore linkages between improved floodplain
delineation tools for alluvial fans and insurance/management concerns. Better
understanding of the nature and type of alluvial fan flooding will result in better
assignment of risk for insurance purposes and better management of the actual hazards.
Comments from Anthony Barry/San Diego County (via email on 9/9/2010)
The County of San Diego’s comments on the White Paper are attached.
Need for Updating Alluvial Fan Floodplain Delineation Guidelines: A White Paper
By Jonathan Fuller, PE, RG, CFM
ASFPM Arid Regions Committee
Introduction
Up to one third of the remaining developable land area in western United States lies on
alluvial fan landforms (Add REF). Therefore, accurate delineation of flood hazards on
these unique landforms is critically important to the health of the National Flood
Insurance Program. The goal of this white paper is to encourage ASFPM member
communities and FEMA to continue their long history of continuously improving tools
for delineating and managing alluvial fan floodplains.
A History of Successes
Hazards associated with alluvial fan flooding were brought into the national spotlight
following catastrophic debris flows on alluvial fans near Rancho Mirage, California in
the 1970’s. These disastrous floods highlighted the fact that alluvial fan flooding was
substantively different than riverine flooding, and therefore required different mapping
and management tools. FEMA responded by implementing a probabilistic methodology
developed by the USGS (Dawdy, 1979). The so-called Dawdy method offered two key
improvements in alluvial fan floodplain delineation technology. First, the method
recognized that alluvial fan flooding is very different from riverine flooding. Second, the
method recognized that uncertain flow paths are a key aspect of alluvial fan flooding.
From 1979 to late 1980’s, the Dawdy method was used to map alluvial fan floodplains
throughout the West. The Dawdy method is now implemented as the FAN model and is
described in Appendix G of the FEMA Guidelines.
In 1995, FEMA contracted with the National Research Council (NRC) to evaluate their
alluvial fan floodplain delineation practices. After deliberating for nearly a year, visiting
numerous alluvial fans, and interviewing dozens of communities, the NRC technical
committee issued their report Alluvial Fan Flooding (1996), which suggested a number
of improvements to FEMA’s alluvial fan floodplain delineation methodology. The NRC
recommendations included the following: (1) recognition that not all alluvial fans are
alike, (2) distinction of active and inactive alluvial fan landforms, (3) incorporation of
geologic data into the fan identification and mapping process, and (4) use of a three stage
delineation process. FEMA adopted many of the key NRC recommendations and codified
them in the NFIP Regulations as well as in Appendix G of the FEMA Guidelines.
Recent Developments
Since publication of the NRC report in 1996, and the revised FEMA Guidelines in 2002,
the three-stage alluvial fan floodplain delineation process has been applied to alluvial fan
landforms throughout the United States. The Flood Control District of Maricopa County
(FCDMC) published its draft Piedmont Flood Hazard Assessment Manual (PFHAM) in
1998, which still stands as the benchmark for implementation of the FEMA three-stage
alluvial fan delineation process. More than 30 alluvial fan floodplain delineations have
been completed in Maricopa County alone using the FCDMC manual. More recently, the
State of California commissioned an Alluvial Fan Task Force, which adopted an
integrated planning and analysis approach which closely parallels the approach
recommended in the NRC Report. FEMA has also recently transitioned from MapMOD
to RiskMAP. Some of the main themes of the RiskMAP program include: (1) better
technology, (2) better mapping, (3) risk-based mapping, and (4) use of multi-disciplinary
approaches.
Need to Update Alluvial Fan Floodplain Delineation Techniques
Given FEMA’s call for better technology and mapping as part of their RiskMAP
program, and the experienced gained by local communities like the FCDMC and the
State of California in alluvial fan floodplain delineation, the Arid Regions Committee
now calls on its members to encourage FEMA to update its alluvial fan floodplain
delineation procedures. ASFPM’s report, National Flood Programs and Policies in
Review (2007), specifically noted that the existing alluvial fan floodplain delineation
techniques inaccurately depict the actual flood hazard. The following are reasons that
now is the time for such an update:

Time. It has been 14 years since publication of the NRC report and eight years since
FEMA adopted the current version of Appendix G. We have learned much over the
past 14 years that could be applied to alluvial fan flood hazard evaluations. Every
technical methodology needs periodic updates. The alluvial fan floodplain
delineation methodology is no exception.

Experience. The current three-stage alluvial fan delineation methodology has been
applied many times over the past decade. As with any new procedure, the more it is
applied, the better we know its short-comings and needs. We should leverage our
experience to shore up existing gaps and weaknesses.

New and Improved Tools. The engineering tools available for evaluation of alluvial
fan flooding have been improved, and new tools have been developed in the past 14
years. These tools include improvements in two-dimensional modeling, new geologic
dating techniques, and new debris flow prediction and modeling tools.
Management & Mitigation Issues
An update of FEMA Appendix G would also be consistent with the larger goals of the
NFIP. Better floodplain delineation tools for active alluvial fans not only enables NFIP
member communities to better manage their flood hazard, but also it provides
opportunities for FEMA to determine whether the hazards are properly rated for
insurance purposes. Furthermore, better hazard identification tools for alluvial fans will
allow FEMA and NFIP member communities to deal directly with coverage for erosion
and debris flow hazards which occur on active alluvial fans.
Recommendations
The Arid Regions Committee strongly believes that we should build on FEMA’s long
history of continuous improvement of their alluvial fan floodplain delineation tools, and
therefore offer the following recommendations for updating the current alluvial fan
floodplain delineation methodology:
Recommendation #1: The methodology should recognize that there are different types
of active alluvial fans. Currently, Appendix G recognizes that there are different types of
flooding on alluvial fan landforms (G.2.2.4), but the methodology does not include any
finer distinctions among types of active alluvial fans. Furthermore, the current
methodology does not distinguish different types of flood zones that correspond to the
differing types of flooding on alluvial fans. Specifically, the guidelines should
distinguish the following:
 Debris flow fans vs. fluvial fans. Debris flow fans have significantly greater flood
hazards than fans subject only to water floods.
 Channelized flow vs. sheet flooding. Fans subject to avulsive, channelized flow have
significantly greater flood hazards than fans which are dominated by shallow sheet
floods.
Recommendation #2: The methodology should distinguish high hazard portions of fans
from lower hazard portions of fans. Currently, the effective methodology does not
adequately distinguish degrees of hazard within an alluvial fan floodplain. Either the
hazards are mapped a parallel bands of equivalent hazard (FAN model) or the entire
alluvial fan floodplain is mapped as an unnumbered A Zone (Geomorphic method),
giving the impression that the hazard level is equal throughout the entire 100-year
floodplain. We now know that the degree of hazard is a function of watershed
characteristics (drainage area, flow rate, flow frequency, sediment production, wild fire
risk, debris flow potential, etc.), fan surface characteristics (slope, sediment size, fan area,
topographic relief, etc.), and the hydraulic characteristics of flow (depth, velocity, etc.).
The following types of hazard levels should be distinguished on active alluvial fans:
 Debris flow hazard areas (See Recommendation #1)
 Avulsion risk areas
 High flood depth & velocity areas
 Shallow sheet flooding areas
Recommendation #3: The terminology used in Appendix G and the NFIP should be
clarified and made internally consistent. An example of contradictory or confusing
language is in the definition and descriptions of active alluvial fans. According to the
language in Appendix G, active alluvial fans can either be areas in which “elevation on
fill would not reliably mitigate the flood risk” (high threshold), or areas of sheet flooding
(low threshold), or areas that would have been inundated in the past 100 or 1,000 years
(low threshold). Areas of shallow sheet flooding on fluvial, non-debris flow fans,
experience low flow depths and velocities. In such areas, it is likely that elevation on fill
would reliably mitigate the risk, even though the fan may be “active” in geologic time
(i.e., aggrading, albeit slowly). In addition to clarifying such potential contradictions, the
text should be edited to remove, to the extent that is feasible, non-quantitative adjectives
from definitions, such as “high,” “abrupt,” “so severe that,” etc.
Recommendation #4: Improve the technical guidance for delineation of active alluvial
fan flood hazards. Specifically, better guidance on what constitutes a composite method
(Table G-1) is needed, as well as how use of a composite method fits with the apparent
(and contradictory) prohibition of using “hydraulic analytical methods” on active alluvial
fans. In addition, more detailed description of the types of analyses required to support a
LOMR request for an active alluvial fan floodplain. Many LOMR/CLOMR submittals
are woefully deficient because of the lack of clarity regarding what is required, resulting
in multiple reviews and frustrated applicants.
Recommendation #5: The methodology should formally recognize specific key
processes that are known to occur on active alluvial fans. These processes include the
following:
 Flow attenuation between the hydrographic apex and the toe of the fan.
Because of this attenuation, use of the full apex discharge downstream of the
apex is overly conservative, particularly near the toe of an alluvial fan. Also,
because of the degree of flood storage that occurs on active alluvial fans, the
impact of development in such storage areas drastically increases peak
discharges, adversely impacting downstream areas, a fact which is not
adequately captured by using the full apex discharge.
 Sheet flooding is the dominant flood processes on the mid- and distal-fan
areas of many active (fluvial) fans. Use of floodplain delineation models that
do not recognize the presence of sheet flooding will inaccurately depict the
true flood hazard. These sheet flooding areas are subject to flow path and
flow rate uncertainty, but generally do not present the same level of risk as
channelized flooding and/or debris flow hazards. That is, any uncertainty in
sheet flooding areas can “be set aside” in the assessment of the hazard.
Recommendation #6: Conduct regular training to increase awareness of alluvial fan
floodplain hazards and keep abreast of advances in technology. This training should be
targeted at FEMA and local community floodplain reviewers, floodplain management
agencies, and development engineers. The target audience should be taught to recognize
alluvial fan landforms, as well as how to delineate flood hazards on active alluvial fans.
In addition, training in the use of new tools needed to quantify active alluvial fan flood
hazards is needed, such as two-dimensional modeling, geomorphic mapping, and
recognition of debris flow hazards.
Recommendation #7. Initiate a consistent review process to assure that the presence of
alluvial fans is not missed in the review and floodplain delineation process. This could
be applied only in communities known to have active alluvial fans or could be applied on
a broader scale. Currently, there is a high potential for alluvial fan flood hazards to be
missed in the delineation and review process. Conversely, there have been a number of
well-documented cases where the threat of alluvial fan flooding existed, but was missed
entirely in the delineation and review process.
Recommendation #8. Investigate the frequency of alluvial fan avulsions. FEMA and
NFIP member communities should specifically investigate the physical mechanisms that
lead to alluvial fan avulsion, and quantify the frequency with which avulsions occur on
debris flow and fluvial fans. This investigation will determine whether avulsion risk is
sufficient to warrant consideration in floodplain delineation and engineering mitigation
evaluations. This investigation could also be used to assess the accuracy and applicability
of statistical/probabilistic approaches.
Recommendation #9. FEMA should explore linkages between improved floodplain
delineation tools for alluvial fans and insurance/management concerns. Better
understanding of the nature and type of alluvial fan flooding will result in better
assignment of risk for insurance purposes and better management of the actual hazards.
Comments from Marty Teal/WEST Consultants (received via email on 7/6/2010)
Jon,
I thought that the white paper was very well done. I have just a few comments/suggestions that
I put in using the track changes feature in Word.
Good job!
Need for Updating Alluvial Fan Floodplain Delineation Guidelines: A White Paper
By Jonathan Fuller, PE, RG, CFM
ASFPM Arid Regions Committee
Introduction
Up to one third of the remaining developable land area in western United States lies on
alluvial fan landforms (Add REF). Therefore, accurate delineation of flood hazards on
these unique landforms is critically important to the health of the National Flood
Insurance Program. The goal of this white paper is to encourage ASFPM member
communities and FEMA to continue their long history of continuously improving tools
for delineating and managing alluvial fan floodplains.
A History of Successes
Hazards associated with alluvial fan flooding were brought into the national spotlight
following catastrophic debris flows on alluvial fans near Rancho Mirage, California in
the 1970’s. These disastrous floods highlighted the fact that alluvial fan flooding was
substantively different than riverine flooding, and therefore required different mapping
and management tools. FEMA responded by implementing a probabilistic methodology
developed by the USGS (Dawdy, 1979). The so-called Dawdy method offered two key
improvements in alluvial fan floodplain delineation technology. First, the method
recognized that alluvial fan flooding is very different from riverine flooding. Second, the
method recognized that uncertain flow paths are a key aspect of alluvial fan flooding.
From 1979 to late 1980’s, the Dawdy method was used to map alluvial fan floodplains
throughout the West. The Dawdy method is now implemented as the FAN model.
In 1995, FEMA contracted with the National Research Council (NRC) to evaluate their
alluvial fan floodplain delineation practices. After deliberating for nearly a year, visiting
numerous alluvial fans, and interviewing dozens of communities, the NRC technical
committee issued their report Alluvial Fan Flooding (1996), which suggested a number
of improvements to FEMA’s alluvial fan floodplain delineation methodology. The NRC
recommendations included the following: (1) recognition that not all alluvial fans are
alike, (2) distinction of active and inactive alluvial fan landforms, (3) incorporation of
geologic data into the fan identification and mapping process, and (4) use of a three stage
delineation process. FEMA adopted many of the key NRC recommendations and codified
them in the NFIP Regulations as well as in Appendix G of the FEMA Guidelines.
Recent Developments
Since publication of the NRC report in 1996, and the revised FEMA Guidelines in 2002,
the three-stage alluvial fan floodplain delineation process has been applied to alluvial fan
landforms throughout the United States. The Flood Control District of Maricopa County
(FCDMC) published it’s draft Piedmont Flood Hazard Assessment Manual (PFHAM) in
1998, which still stands as the benchmark for implementation of the FEMA three-stage
alluvial fan delineation process. More than 30 alluvial fan floodplain delineations have
been completed in Maricopa County alone using the FCDMC manual. More recently, the
State of California commissioned an Alluvial Fan Task Force, which adopted an
integrated planning and analysis approach which closely parallels the approach
recommended in the NRC Report. FEMA has also recently transitioned from MapMOD
to RiskMAP. Some of the main themes of the RiskMAP program include: (1) better
technology, (2) better mapping, (3) risk-based mapping, and (4) use of multi-disciplinary
approaches.
Need to Update Alluvial Fan Floodplain Delineation Techniques
FEMA needs to update its alluvial fan delineation procedures. Given FEMA’s call for
better technology and mapping as part of their RiskMAP program, and the experienced
gained by local communities like the FCDMC and the State of California in alluvial fan
floodplain delineation, the Arid Regions Committee now calls on its members to
encourage FEMA to update its procedures. ASFPM’s report, National Flood Programs
and Policies in Review (2007), specifically noted that the existing alluvial fan floodplain
delineation techniques inaccurately depict the actual flood hazard. The following are
reasons that now is the time for such an update:

Time. It has been 14 years since publication of the NRC report and eight years since
FEMA adopted the current version of Appendix G. We have learned much over the
past 14 years that could be applied to alluvial fan flood hazard evaluations. Every
technical methodology needs periodic updates. The alluvial fan floodplain
delineation methodology is no exception.

Experience. The current three-stage alluvial fan delineation methodology has been
applied many times over the past decade. As with any new procedure, the more it is
applied, the better we know its short-comings and needs. We should leverage our
experience to shore up existing gaps and weaknesses.

New and Improved Tools. The engineering tools available for evaluation of alluvial
fan flooding have been improved, and new tools have been developed in the past 14
years. These tools include improvements in two-dimensional modeling, new geologic
dating techniques, and new debris flow prediction and modeling tools.
Management & Mitigation Issues
An update of FEMA Appendix G would also be consistent with the larger goals of the
NFIP. Better floodplain delineation tools for active alluvial fans not only enables NFIP
member communities to better manage their flood hazard, but also it provides
opportunities for FEMA to determine whether the hazards are properly rated for
insurance purposes. Furthermore, better hazard identification tools for alluvial fans will
allow FEMA and NFIP member communities to deal directly with coverage for erosion
and debris flow hazards which occur on active alluvial fans.
Recommendations
The Arid Regions Committee strongly believes that we should build on FEMA’s long
history of continuous improvement of their alluvial fan floodplain delineation tools, and
therefore offers the following recommendations for updating the current alluvial fan
floodplain delineation methodology:
Recommendation #1: The methodology should recognize that there are different types
of active alluvial fans. Currently, Appendix G recognizes that there are different types of
flooding on alluvial fan landforms (G.2.2.4), but the methodology does not include any
finer distinctions among types of active alluvial fans. Furthermore, the current
methodology does not distinguish different types of flood zones that correspond to the
differing types of flooding on alluvial fans. Specifically, the guidelines should
distinguish the following:
 Debris flow fans vs. fluvial fans. Debris flow fans have significantly greater flood
hazards than fans subject only to water floods.
 Channelized flow vs. sheet flooding. Fans subject to avulsive, channelized flow have
significantly greater flood hazards than fans which are dominated by sheet floods.
Recommendation #2: The methodology should distinguish high hazard portions of fans
from lower hazard portions of fans. Currently, all alluvial fan floodplains are considered
unnumbered A Zones, which may give the impression that the hazard level is equal
throughout the entire 100-year floodplain. We now know that the degree of hazard is a
function of watershed characteristics (drainage area, flow rate, flow frequency, sediment
production, wild fire risk, debris flow potential, etc.), fan surface characteristics (slope,
sediment size, fan area, topographic relief, etc.), and the hydraulic characteristics of flow
(depth, velocity, etc.). The following types of hazard levels should be distinguished on
active alluvial fans:
 Debris flow hazard areas (See Recommendation #1)
 Avulsion risk areas
 High flood depth & velocity areas
 Shallow sheet flooding areas
Recommendation #3: The terminology used in Appendix G and the NFIP should be
clarified and made internally consistent. An example of contradictory or confusing
language is in the definition and descriptions of active alluvial fans. According to the
language in Appendix G, active alluvial fans can either be areas in which “elevation on
fill would not reliably mitigate the flood risk” (high threshold), or areas of sheet flooding
(low threshold), or areas that would have been inundated in the past 100 or 1,000 years
(low threshold). Areas of shallow sheet flooding on fluvial, non-debris flow fans,
experience low flow depths and velocities. In such areas, it is likely that elevation on fill
would reliably mitigate the risk, even though the fan may be “active” in geologic time
(i.e., aggrading, albeit slowly). In addition to clarifying such potential contradictions, the
text should be edited to remove, to the extent that is feasible, non-quantitative adjectives
from definitions, such as “high,” “abrupt,” “so severe that,” etc.
Recommendation #4: Improve the technical guidance for delineation of active alluvial
fan flood hazards. Specifically, better guidance on what constitutes a composite method
(Table G-1) is needed, as well as how use of a composite method fits with the apparent
(and contradictory) prohibition of using “hydraulic analytical methods” on active alluvial
fans. In addition, more detailed description of the types of analyses required to support a
LOMR request for an active alluvial fan floodplain. Many LOMR/CLOMR submittals
are woefully deficient because of the lack of clarity regarding what is required, resulting
in multiple reviews and frustrated applicants.
Recommendation #5: The methodology should formally recognize specific key
processes that are known to occur on active alluvial fans. These processes include the
following:
 Flow attenuation between the hydrographic apex and the toe of the fan.
Because of this attenuation use of the full apex discharge is overly
conservative. Also, because of the degree of flood storage that occurs on
active alluvial fans, the impact of development in such storage areas
drastically increases peak discharges, adversely impacting downstream areas.
 Sheet flooding is the dominant flood processes on the mid- and distal-fan
areas of many active (fluvial) fans. Use of floodplain delineation models that
do not recognize the presence of sheet flooding will inaccurately depict the
true flood hazard. These sheet flooding areas are subject to flow path and
flow rate uncertainty, but generally do not present the same level of risk as
channelized flooding and/or debris hazards.
Recommendation #6: Conduct regular training to increase awareness of alluvial fan
floodplain hazards and keep abreast of advances in technology. This training should be
targeted at FEMA and local community floodplain reviewers, floodplain management
agencies, and development engineers. The target audience should be taught to recognize
alluvial fan landforms, as well as how to delineate flood hazards on active alluvial fans.
In addition, training in the use of new tools needed to quantify active alluvial fan flood
hazards is needed, such as two-dimensional modeling, geomorphic mapping, and
recognition of debris flow hazards.
Recommendation #7. Initiate a consistent review process to assure that the presence of
alluvial fans is not missed in the review and floodplain delineation process. This could
be applied only in communities known to have active alluvial fans or could be applied on
a broader scale. Currently, there is a high potential for alluvial fan flood hazards to be
missed in the delineation and review process.
Recommendation #8. Investigate the frequency of alluvial fan avulsions. FEMA and
NFIP member communities should specifically investigate the physical mechanisms that
lead to alluvial fan avulsion, and quantify the frequency with which avulsions occur on
debris flow and fluvial fans. This investigation will determine whether avulsion risk is
sufficient to warrant consideration in floodplain delineation and engineering mitigation
evaluations. Accuracy and applicability of statistical/probabilistic approaches should be
assessed.
Recommendation #9. FEMA should explore linkages between improved floodplain
delineation tools for alluvial fans and insurance/management concerns. Better
understanding of the nature and type of alluvial fan flooding will result in better
assignment of risk for insurance purposes and better management of the actual hazards.
Comments from Marc McIntosh/URS Corp (via email on 6/24/2010)
Jon,
as discussed:
Recommendation #2 - Some alluvial fan floodplains are numbered A zones.
Recommendation #5 - First bullet item should clarify the use of the full apex discharge for pre and post
development.
Recommendation #7 - Maybe adding a statement that fan hazards are sometimes delineated where they
should not be may drive the point home...
let me know if you need further clarification
thanks!
Comments from Ann Moynihan (Pima County RFCD (via email 6/21/2010)
Hi, Jeanne,
I am just introducing myself to the committee’s work, and unfortunately have been unable to
attend the last 2 conference calls. You requested comments about the white paper on or before
June 25, and the usual suspects at Pima County Flood Control who would review the paper are
just not available to do it right now.
My overall impression is that the paper does not go far enough to simply state that much of
Appendix G may be outdated realistically. If we want people in the arid southwest to use the
most updated 2-D modeling, i.e. FLO 2-D, then why don’t we just say so? Yes, a field,
geomorphic and soils analysis should be performed to set parameters and locations for modeling,
but I recommend strongly stating that FAN, HEC-RAS, HEC-HMS, HEC-1 and approximate
methods are just not suitable in most cases. Maybe this will come at a later stage? That being
said, FLO-2D requires a high level of expertise, and I would want to see modelers certifying that
they have attended both an introductory and an advanced workshop with Jim O’Brien and also
references for past FLO-2D projects.
In Recommendation #2, the paper states that fan mapping is all unnumbered A, while the
examples in Appendix G show AO’s with depths. AO’s with depths seem acceptable to me, so I
am not sure if you would want to mention something about AO’s under Recommendation #2.
Recommendation #5 is vague about how to evaluate flow attenuation. Right now, most people
are sending that apex discharge through plus in many cases adding rain. Will this be a separate
committee study or paper?
Recommendation #6. Who would provide the training and how would it be funded? Most
government people may not be able to go to “regular” training. Right now, Pima County is maybe
sending a few people to workshops when the registration is free!
Recommendation #7. Review checklists or structured guidance must be prepared to document
what the “consistent review process” is.
Recommendation #8. Should be funded by FEMA.
Recommendation #9 goes away if there is a stronger statement earlier in favor of certain tools.
That’s my 2 cents.
Thank you,
Ann
Comments from Kathryn Gross/FCDMC (via email on 5/13/2010)
Wanted to let you know I did get a chance to read it. I feel it very eloquently identifies the issues
that remain without being preachy or negative to the existing framework. I think it addresses all
the issues so I have no comments. Well maybe one. The use of “so-called” with the Dawdy
method. For some reason I see the use of “so-called” as negative connotation but I know that
technically/literally the method is referred to in that fashion as well....
Should be interesting to see how this progresses.
Kathryn
Comments from Iovanka Todt/FMA (via email on 5/12/2010)
Jon,
Thank you for the opportunity to review and comment. I think it is an excellent summary of the
issues. The AFTF discussions over the past 2.5 years have mentioned a few additional points some of these were mentioned by Doug and Greg. I appreciate your earlier suggestion to start
with this framework at the FMA conference special alluvial fan session this November, and collect
additional input from a broader range of experts and practitioners. Thank you again and look
forward to staying in touch.
Iovanka
Comments from Doug Plasencia (via email on 5/12/2010)
Jon:
This is a very well thought out discussion and list of recommendations. The only advice
that I can provide at this time is that this white paper primarily focuses on the hazard
identification phase. You may want to consider in your background and
recommendations a tree top treatment on management and mitigation issues, as well as
insurance issues. These would not necessarily need to be covered in detail, but I would
urge you to build the linkages.
Based on better hazard ID methods and data, this takes us directly into the adequacy of
the management standards (including treatment of debris and erosion). On the
insurance side it also suggests more opportunity to evaluate whether the hazards are
properly rated, and deals directly with the issue of coverage for erosion or debris flow.
Doug
Comments from Greg Jones/FCDMC (via email on 5/12/2010)
Jon,
The white paper mirrors the issues that we have been discussing and investigating. Additionally,
it is concise and well written. However, should the paper also include a reference to avulsion
potential? This potential greatly affects the uncertainty and the hazard classes indicated in this
paper.
Sincerely,
Gregory L. Jones, PE, AICP
Comments from Bob Davies/AMEC (via email on 5/10/2010)
Jon,
I just had a couple of minor comments, see attached. I think you’ve a good job stating the
importance of who cares and how big is need for providing floodplain delineations on fans. Also
you provide enough history to give the reader context. Your recommendation section is specific
and your points are well crafted, although I’m not sure the implementation for each is crystal clear
(most are).
Great job Jon!
Often, for white papers when they are drafted, the draft is circulated through either select or all
the policy committee co-chairs for commentary. Then, it is circulated to the Executive
committee. I suppose then it could go to a broader distribution. Let me know how we can help!
Hope my input helps. Call me if you want to discuss.
Bob
Appendix B: FMA-ASFPM Meeting Notes
ASFPM Arid Regions: Alluvial Fan Whitepaper Discussion
ASFRPM/FMA Joint Conference 11-3-2010, Henderson, NV
Moderator: Jon Fuller
Notes by: Nate Vaughan
Questions and comments attributed at time if name known.
Fuller - Paper posted on ASFPM Arid Regions Website.
Recommendation #1
Fuller – Everyone agrees there are different kinds of fans?
Ray – In reviewing National Academy Report, there are concerns about uncertain flow
paths. This is a problem. How do we solve the problem?
- This is related to research needs.
- This is related to Recommendations #8 and #1
- There may be differences between debris flow fans vs. fluvial fans vs. fans w/
uncertain flow paths.
- Methodology must address uncertain flow path flooding
- Recognize fan behavior(s) and that they may be plural in nature
- All fans may have uncertain flow paths.
Jim O’Brien – Distinguish fans in equilibrium w/ sediment supply. Recognize
channel forming event.
- Add sub-bullet that suggests channel flow vs. sheet flooding; that event
mapped be outside channel-forming event.
Fuller – Methodology should consider equilibrium behavior of fan as expressed
by channel capacity.
Valerie Swick – How is that different from riverine flooding?
J O’Brien – Concept of channelized flow dictate by topography – magnitude vs.
capacity @ top of fan.
Fuller – Flow direction away from channel differentiates fans from riverine
flooding.
- Recognize that different types of processes occur on different types of fans
(focus on processes)
- Validity of mapping over time – approximate time scale
- Be cognizant of time scale
- Note/determine frequency of various types of processes.
Wen – May add land-use change to definition. How to evaluate change in landuse on fan.
Katy – Because alluvial fans are dynamic potential for tagging “at-risk” areas for
more frequent analysis.
Fuller – concur with analysis associated with risk
- Recurrence intervals?
-
Not necessarily. Sensitive fans may require more frequent analyses based on
change.
Fuller – condition of watershed & potential for change should be recognized.
Recommendation #2
Fuller – do we agree that there are higher and lower hazard portions of fans?
- At what time?
Fuller - @ Engineering time-scales are there high and lower hazard portions of
fans?
David Williams – “Hazard” needs to be defined
Fuller – use of probably to define magnitude and frequency of hazard.
Doug – Are there places on fans with little to no flood hazard? Maybe. Are there
distinctions between types of “high” hazard. Perhaps everywhere on fan is high
hazard.
Fuller – Places in active areas can be worse than others.
Fuller – Can we agree there are relative levels of hazards on fans?
Wen – But hazards are not very comparable between different fans.
- How do we define hazard? We need to go beyond depth and velocity.
Fuller – The assessment of hazard should go beyond depth and velocity.
J O’Brien – Distinguish between risk and hazard in bullet items in description of
#2
- Hazard delineation should include intensity and frequency of hazard.
- Revise Rec #2 to say “less hazardous” or distinguish relative levels of hazard
instead of high vs. low hazard.
B Fry – Reflect level of hazard in delineation.
E Curtis – Inherrent uncertainty still needs to be recognized.
- SFHA should be differentiated from “high hazard area.” Per NFIP, you are
either in high hazard area or out.
Fuller – Similar to floodplain vs. floodway definition.
Ray L – Maricopa County has own FW criteria, but no one else does.
- Concept to maintain floodway area to allow for change in capacity on fans.
Fuller – This is a planning versus mapping function.
Mark Wills – Degrees of hazard are conditional and not guaranteed. May be all
high hazard.
Recommendation #3
Fuller – We all agree that terminology problems exist?
Recommendation #4
– Better technical tools are available now than 15 years ago.
- Training is needed.
Fuller – “One size does not fit all” – guidance needed.
Wen – What does better description of “composite methods” mean?
Fuller – [Shows table G-1 from Appendix G] – Unclear definition
Wen – Does composite method apply only to active or inactive fans?
Fuller – It’s subject to interpretation.
Wen – Universal application of App G – a connection between active vs. in-active
must be made.
- must use composite or cannot determine where is active vs. inactive
- What is appropriate methodology?
Fuller – I had interpreted as only active.
D Williams – Perhaps use multiple methods and compare and assess results – use
judgement.
Recommendation #5
Valerie S – Does “avulsion potential” better define?
- Assessment must be made to recognize applicability of processes
- “Peak flow attenuation” should be better defined
o It’s scary because risk is mitigated by apparent peak flow attenuation
when this is not universally true on all fans. Varies with
 Degree of saturation
 Length of flow path
 Discharge
 Others
Ray L – Bank erosion process not recognized – no sediment transport in
recommendation. Bulking should be considered.
Valerie S – No 2 fans are alike.
J O’Brien - #1 control on area being delineated is peak flow volume rather than
peak magnitude.
Wen – Energy losses occur in processes and can limit extents of fan.
Ray L – Consider multiple events. Sequencing impacts flow paths
- intersection of multiple fans would also need to be studied
Ray L – Need long-term sustainability of solutions for communities
Fuller – Difference between riverine and fans in duration fo specific hazard
conditions
Hydrology consider
o volume
o sources
o sediment impacts
- Let’s discuss in greater detail
J O’Brien – Triggers are
- fires
- development
- multiple flood events
- Use multiple maps and update based on triggers
Wen – Do we have agreement on peak flow determination? Bulked vs. clear
water
- investigate and differentiate peak flow methods from riverine methods
- recognize assessment of geologic vs. 100-year risk. Geologic assessment
explicitly investigates this.
Mark Wills – Methodology should look @ potential changes in debris production
in watershed.
Fuller – We would like a methodology which allows us to evaluate these potential
impacts on fans.
Ray – Climate change, etc.
Fuller – Lots of uncertainties are present.
Recommendations #6 and #7
Fuller – We should be better trained.
David Williams – Train reviewers
Ray – Need consistent review process
- Are other regions consistent? Are we consistent w/ them?
Wen – Fan program is recognized by FEMA, but is dated. Needs improvement.
Recommendation # 8
Marty Teal – Mandate is to take into account stochastic process.
- Monte Carlo type simulations allow “bracketing” of risk
- Look @ processes – when does a channel avulse? Where does it avulse?
- There seems to be a lack of data on fans
o Depths
o sediment
o flow
o how to measure is important
o fly topo over lots of fans
Ray – Recommendation for #10
How to document and what to document.
- there is data out there but it’s scattered
- In some cases, flows demolished small desert towns
Ray – get high water marks quickly
Fuller – Get communities to pool data (recommendation for #10)
J O’Brien – IFSAR database has all of CA flown
- Maybe refly in 10-years and compare
Valerie S – Haven’t had aerial maps to do comparison. Got lucky in 1954 with
fan 36.
Jeannie – Cutting edge work in Europe. Should be reviewed and considered.
- Compile research completed and needs to be completed.
Fuller – Wrap-up.
Notes by Kathryn Gross/FCDMC
Recommendation 1
1. All agree we have different kinds of fans
2. Ray - NRC report – uncertain flow path flooding. There is a problem regarding how we
are going to figure it out. Alluvial fan uncertainty that is the big concern
3. Time and money concern from ….
4. Method has to recognize flow path uncertainty
5. Same event in watershed may result in debris flow or fluvial flow
6. Equilibrium - natural 2-5 year event huge flooding event will go everywhere
7. Process of avulsion are limited (?) for an event you are mapping if outside of the channel
forming event
8. Jon – consider equilibrium conditions on the fan as expressed by channels (?)
9. Differences on active alluvial fans differences in process.
10. Agree have different types of change in space and in time
11. Timescale of processes versus timescale of delineation
12. Wen – landuse change is an issue. How to analyze urban fans/how to handle upstream
urbanization
13. Jeanne – dynamic systems. Potential for id more fan areas that are more at risk. More
frequent analysis. Fans sensitive to change
14. Question on recurrence intervals
Recommendation 2
1.
2.
3.
4.
5.
6.
7.
Do we agree with high and low? “At what point in time”, engineering timescale?
Doug – places on alluvial fan landform, places there are hazard, all places where wet?
Relative levels of hazard
Other parameters: depth, velocity, what other parameters to depict hazard
Ricardo – remove risk-hazard
Intensity/frequency
Ray – SFHA are all called high hazard. Need for regulatory tools for open space/ a place
to keep the water (like what is used in MC)
8. Implicit assumptions- active fan has lower hazard – may be levels of hazard
Recommendation 3
No notes written down….
Recommendation 4
1. Given complexity and technical tools available who can argue with need to improve
technical guidance
2. Ensure a regard for 1 size does not fit all
3. What is meant by better
4. Assuming G1 is all active methods
Recommendation 5
1. Use of full apex flow on all parts of fan
2. Sergio – need multiple storms, geologic assessment, avulsion
3. Georgia – look at attenuation and risk goes away. Need parameters on when it could be
considered
4. Ray – flow attenuation, bank erosion is occurring on channels, ½ the q in the channel
could be sediment flow. Account for sediment flow from channel erosion
5. Jim – flow volume is controlling attenuation. Do not ignore volume
6. Ray – multiple events back to back 100 year events
7. Susan – what about multiple fans
8. Paper should list variables that need to be considered
9. Jim – triggers for need for remap – fire development
10. Wen – what type of “peak flow” calculation. Is normal hydrology methods acceptable
for determining apex flow
11. Sergio – looking at geology – extreme event. What about processes in the watershed
Recommendation 6
1. Yes
Recommendation 7
1. Ray – look at other regions and see if their reviews are consistent with ours
2. Wen – FAN is a dated program – make windows based
Recommendation 8
1. Marty – some work already done. Corp and uncertainty. Deterministic w/ monte carlo
with stochastic “answers”
2. Quantifying physical process and stochastically
3. Explore other ways of expressing uncertainty
4. Lack of data
5. How to measure what will happen
Recommendation 10
1. Ray - Document avulsions and are encouraged to document storm events
2. Iovanka – like concept of looking at uncertainty include guidance on how to interpret