Preparing for Storm Surge and Inundation Effects - Building in Resiliency

1
"Preparing for Storm Surge and Inundation Effects
- Building in Resiliency"
Roy K. Dokka Dept. Civil Environmental
Engineering and Center for GeoInformatics Louisi
ana State University, Baton Rouge, LA 70803
2
Relevant Observations

• Civilization exists by Natures consent, subject
  to change without notice.
• Nature knows no disasters.
• A natural disaster is the residue of human
  design that failed to comply with the laws of
  nature.
• No one plans to fail, they just fail to plan.
• Resiliency-- "the process of, capacity for, or
  the outcome of successful adaptation despite
  challenging or threatening circumstances."

3
Steps to Resiliency

• Broad community recognition of the scope of the
  problem.
• Making the Level of Resiliency Decision. What
  outcomes do we want?
• Assessment of current vulnerabilities. What will
  likely happen?
• Mitigation of current vulnerabilities. Shall we
  stop doing the stupid stuff?
• Assessment and planning for future
  vulnerabilities. What could happen?
• Develop controls to stop building-in new
  vulnerabilities.
• Has Nola taken taken any of these steps?

4
Storm Surge Assessment

• Modeling for assessment of storm surge is getting
  better and more realistic.
• Key inputs for modeling, specifically topography
  and bathymetry are inaccurate.
• Topography has been updated along LA coast with
  LiDAR but accuracy assessment shows it to be
  uncertain. Subsidence has compromised our ability
  to relate measurements to the vertical datum. The
  same is probably true for central Gulf states.
  Bathymetry is out-of-date.
• Relationships between datums (water level,
  geodetic, stream) are not understood by most
  people.
• Result?Storm surge models are not as realistic as
  they could be, FEMA flood maps are inaccurate,
  and future hurricane protection may not be
  adequate.
• Why?

5
Its the topography, stupid. Subsidence has
ruined Louisianas vertical control system!
NOAA told the U.S. Congress that Louisiana in
2001 that the system used to measure elevations
was,
inaccurate and obsolete and unable to support
public safety.
Bad vertical control affects levees, evacuation
roads, FEMA flood maps, and georeferencing of
LiDAR.
6
Subsidence in the last 50 years
MS
AL
TX
Baton Rouge
New Orleans
7
Build a levee in the delta plain and the land
subsides below sea level. This has already
happened in lower Plaquemines Parish.
Flood protection levee
Mississippi River
NOAA LIDAR
Hot colors high cold colors low. Notice that
medium blue marks sea level. Areas in dark blue
within the flood protection levees are below sea
level.
Flood protection levee
South
8
What is Causing Subsidence?
Deep Processes ?Sediment and water load induced
flexure of the lithosphere 0 to
-7mm/yr ?Faulting 0 to -15mm/yr ?Salt
evacuation 0 to -?? mm/yr Shallow Processes
?Natural consolidation and compaction ?Desiccat
ion by urbanization (behind levees)
?Organic soil oxidation ?Oil gas extraction
0 to -3 mm/yr ?Water pumping 0 to -65mm/yr
9
Center for GeoInformatics was created at LSU and
funded by Congress to help NOAA fix the vertical
control problem
This required

• Creation of State-wide Geospatial Reference
  Network A robust, spatially and temporally
  precise, and ubiquitous reference frame that can
  be accessed by users. Without access to a datum,
  all measurements are uncertain. GULFNet
• Harnessing Technology for Faster, Better, Cheaper
  Measurements Real-time kinematic GPS network to
  allow users to obtain accurate and precise 3-D
  positions anywhere along the coast.
• Geological understanding? to understand
  underlying causes of subsidence.

10
GULFNet

• The basis for National Spatial Reference System
  of the USA in Louisiana.
• Surveyors use our 2M LSU GULFNet GPS Reference
  network to determine accurate
• elevations and positions.
• Data provided to NOAA
• who then serve it to the
• public.
• Basis for high precision
• GIS data bases.
• Scientific uses such as
• measuring land
• subsidence and accuracy
• assessment of LiDAR.

11
Covington-Mandeville Sample area
Slidell Sample area
Maurepas-Laplace Sample area
Rigolets Sample area
West-East Bank Sample area
Ngt100,000
Plaquemines Sample area
12
Accuracy Assessment of DEMS -- SE Louisiana
Difference between NAVD88 and NGVD29
LiDAR test points are blue. --Red line is
acceptable error envelope for LiDAR 30m USGS
DEM test points are magenta
13
Accuracy Assessment Statistics
Sampling Surveys by Name.

• Slidell A - 55 pass _at_ 0.8 ft (n1220)
• Slidell B 46 pass _at_ 0.8 ft (n68)
• Slidell C - 48 pass _at_ 0.8 ft (n924)
• Port Sulphur - 83 pass _at_ 0.8 ft (n2835)
• Boothville 74 pass _at_ 0.8 ft (n489)
• St. Bernard - 84 pass _at_ 0.8 ft (n1387)
• Mississippi 22 pass _at_ 0.8 ft (n803)

Federal standards say a passing score is gt90
14
Here the LiDAR is too low, thus SSMs and FIRMS
will overestimate the risk.
Red dotLIDAR DEM pixel that is more than 0.8
foot below truth ?Associated numbers are
truth-DEM pixel value ?truth is a GPS
orthometric point referenced to NAVD88 that
is known to 0.3 ft NAVD88
48 pass at 0.8 ft
15
Red dotLIDAR DEM pixel that is more than 0.8
foot below truth ?Associated numbers are
truth-DEM pixel value ?truth is a GPS
orthometric point referenced to NAVD88 that
is known to 0.3 ft NAVD88 Green dotLIDAR DEM
pixel that is more than 0.8 foot above truth
Here the LiDAR is too high, thus SSMs and FIRMS
will underestimate the risk.
Plaquemines Sample Area
16
Test Results

• Statement of hypothesis that was tested
• 90 of the pixels of the Lidar-based DEM is
  accurate to 0.5 feet NAVD88. Because our methods
  have a resolution of 0.3 feet, this only allowed
  testing at 0.8 ft NAVD88.
• Conclusion
• The hypothesis that states that 90 of pixels
  comprising the Lidar-based DEM is accurate to
  0.8 feet NAVD88 is false. By logical extension,
  the DEM is also not accurate to 0.5 feet NAVD88.
• Recommendation
• Because the Lidar DEM is the basis for storm
  surge modeling that supports hurricane protection
  design, floodplain mapping, and emergency ops, it
  would prudent to extend the accuracy assessment
  to include the entire Gulf Coast. This should
  include an assessment of why the DEM is
  inaccurate and what might be done to improve it.

17
NOAA SLOSH was successful during Katrina because
subsidence and datum differences were accounted
for
Model was run with tide level at 2.5 ft to
account for subsidence effects since topo mapping
and the datum Shift between NGVD29 and NAVD88.
Katrina
18
What about other areas of the Gulf?
Houston. You have a problem!
19
Galveston-Baytown
20
Age of Bathymetry
21
Conclusions

• Preparations for future storm surge and in
• resiliency building for our coast will benefit
  greatly
• from realistic models that can mimic nature.
  These
• models can educate and help plan and test
• mitigation strategies to find out what works
  and
• what unintended consequences are created.
• The greatest, currently unmet need for storm
  surge
• modeling is accurate, high resolution
  topographic
• and bathymetric data. Today, such topographic
  and
• bathymetric data are not generally available.
• All coastal players, i.e., planners, engineers,
  GIS,
• emergency managers, floodplain managers, and
• ecologists, must be on the same geospatial
  page.

22
Thank You!
Any questions about this vulnerable land?