1.3. Uncertainty Ranges Associated with EPA's
Estimates of the Area of Land Close to Sea Level
Authors: James G. Titus, U.S. Environmental Protection Agency
Dave Cacela, Stratus Consulting Inc.
This section should be cited as:
Titus, J.G., and D. Cacela. 2008. Uncertainty Ranges Associated with EPA's Estimates of the
Area of Land Close to Sea Level. Section 1.3 in: Background Documents Supporting Climate
Change Science Program Synthesis and Assessment Product 4.1: Coastal Elevations and
Sensitivity to Sea Level Rise, J.G. Titus and E.M. Strange (eds.). EPA 430R07004. U.S.
EPA, Washington, DC.
-------
Section 1.3.1. Approach Author: James G. Titus
Introduction
Digital Elevation Model output allows one to
easily generate a point estimate ("best guess") of
the amount of land below a particular elevation
X by simply tabulating the number of points
below X and multiplying by the cell size that
each point represents. The accuracy of available
elevation data varies, however, so the accuracy
of these point estimates of the area estimates will
vary as well. For some purposes, it may be
sufficient to have a "best guess" estimate. But for
other purposes, one needs some sort of
uncertainty range. Fortunately, most elevation
data come with a precision estimate, which
makes it possible to develop an uncertainty
range.
Section 1.3 explains how Dave Cacela and this
author generated an uncertainty range for the
estimates of the amount of land close to sea level
within different shore protection categories and
different elevations, which form the basis of this
report. Section 1.3.1 explains the assumptions
and the basic approach for estimating
uncertainty; Section 1.3.2 explains how the
approach was implemented. Section 1.3.3
provides the results. The final results constitute
the three appendices to this section.
Like Section 1.2, by Jones and Wang, the
starting point is the elevation data set developed
in Section 1.1 by Titus and Wang. The approach
for specifying uncertainty is based on the most
important sources of error in that analysis. The
actual implementation, however, uses the output
from Section 1.2, in which Jones and Wang
overlay the elevation study by Titus and Wang
with the eight state-specific shore protection
studies that Titus and Hudgens developed in
their unpublished analysis mentioned in Section
1.2. Section 1.1 provided cumulative elevation
distributions for dry land and nontidal wetlands;
Section 1.2 subdivided the dry land into the
various shore protection categories. Our
exposition of the approach taken focuses on the
elevation distribution of dry land. But not only
did we apply the procedure to the totals for dry
land, we also applied it to all the other shore
protection categories and nontidal wetlands.
We warn the reader at the outset that this section
switches between metric (standard international)
and English (imperial) units of measurement.
The final results are in metric units—but most of
the underlying elevation data were based on
topographic maps with contour intervals
measured in feet. The point of measurements
provided in this section is generally to explain
the relationship between input data and
assumptions, not to inform the reader about the
magnitude of any particular effect. Therefore, the
reader unfamiliar with one or the other system of
measurements need not attempt to make
conversions. In the few cases where that actual
magnitude may matter, our convention is metric.
Background
Previous assessments of the land vulnerable to
sea level rise have provided an uncertainty range;
but the uncertainty range did not include
uncertainty associated with topographic
information. EPA's 1989 Report to Congress
provided an uncertainty range about the area of
land lost for a rise in sea level of 50, 100, or 200
cm. In Appendix B to that Report to Congress,
Titus and Greene (1989) developed the
uncertainty range, based on a study by Park et al.
(1989), who used a sample of study area sites,
and calculated a point estimate of land loss of
each site. The published uncertainty range used a
simple sampling error approach, treating the
study sites as a random sample from the entire
population of USGS quads. Because Park et al.
did not report an uncertainty range for their
-------
[70 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
sample sites, Titus and Greene made no attempt
to include that uncertainty. In effect, Titus and
Greene assumed that Park et al. accurately
estimated the amount of land at particular
elevations in those areas they assessed. The true
uncertainty associated with their estimates
included both sampling and measurement error;
but the published uncertainty range considered
only the sampling error.
This study uses the elevation data from Section
1.1, as formatted by the analysis explained in
Section 1.2. That data set estimated the
elevations of all land above spring high water.
That is, it estimated elevations for dry land and
nontidal wetlands, but did not estimate
elevations for tidal wetlands. (Knowing that land
is tidal wetland tells us that the land elevation is
below spring high water and above mean low
water, which provides a narrower uncertainty
range about the elevation than if we know only
that the land is below, for example, the 10-ft
contour on a topographic map.) Because they
obtained data for the entirety of the study area,
there is no sampling error. The source of error
stems entirely from the limitations in precision
of the Section 1.1 results.
The overall approach is to make an assumption
about the potential vertical error of the elevation
data and the extent to which that error is random
versus systematic. The magnitude of the error
varies by data source: because we assume that
error is a function of contour interval, which in
turn varies by topographic quad, we calculate
error separately for each topographic quad. Let
us first explain our basis for focusing on vertical
error of the elevation data, and then explain how
low and high estimates for areas were calculated
where the input data were USGS contour maps
and other data with relatively coarse contour
intervals (1 meter or worse), as well as our
procedure for when the data had higher quality
(2 feet or better).
Horizontal and Vertical Precision
Figure 1.3.1 depicts the various sources of data
used to estimate elevations and the areas of land
at particular elevations. In most locations, Titus
and Wang relied on USGS 1:24,000 scale maps
with various contour intervals. The second most
common source of data was LIDAR provided by
Maryland or North Carolina, which give
elevations at various points in a grid.
USGS maps follow the national mapping
standards for vertical and horizontal precision.
The vertical standard is that 90 percent of the
well-defined points along a contour must be
within one-half the contour interval above or
below the stated elevation of the contour. The
horizontal standard is that 90 percent of the
points should be within one fiftieth of an inch
(about half a millimeter). On a 1:24,000 scale
map, the allowable horizontal accuracy would be
12 meters. The LIDAR data sources generally
have vertical precision on the order of 10-30 cm
and horizontal error of less than 1 meter.
To keep the analysis reasonably manageable, this
study ignores the horizontal error and focuses
entirely on the vertical errors. Inspection of the
USGS maps and the maps produced by Titus and
Wang shows that most lowland is in an area
where the contours are hundreds—and often
thousands—of meters apart. Random error on
the order of 12 meters is very small by
comparison and not likely to substantially
change an estimated error range. The horizontal
error of LIDAR seemed even less likely to
matter. In an assessment of the impacts of rising
sea level, what matters is that most of the input
data had contour intervals of 5 feet (150 cm) or
worse, and we are interested in the implications
of a 50-cm rise.
-------
[ SECTION 1 .3 71 ]
PA
MD
fi
/ ^
J c
DC
VA
ATLANTIC
OCEAN
Contour Intervals
| Spot Elevation
Lidar
2 Feet
1 Meter
5 Feet
10 Feet
10 Feet, State Data
20 Feet
Figure 1.3.1. Input Elevation Data used in Section 1.1 to Estimate Area of Land Close to Sea Level.
Quadrangles with a 10-ft contour interval and a 5-ft supplemental contour are shown as 5 feet. The
Maryland data included 5-ft contours drawn from spot elevation with RMS error of 5 feet; hence the legend
calls the data "10 feet, State Data"; USGS 5-ft contours have an RMS error of 2,5 feet.
-------
[72 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
Areas with USGS Maps as the Input Data
This analysis assumes that the standard deviation
of error within a neighborhood is one-half the
contour interval, based on National Map
Accuracy standards. For reasons discussed
below, the calculations also assume that half the
error is random and half is systematic, so that the
standard deviation of the uncertainty is one-
quarter the contour interval for areas the size of a
county or larger. These assumptions are adjusted
to address possible error in the estimate of spring
high water (SHW).
Our Initial Model of Vertical Error
Based on a comparison of their model results
with LIDAR from Maryland and North Carolina
(see Section 1.1, Jones 2007, and Jones et al.
2008), Titus and Wang report that the root mean
square (RMS) error1 of their elevation data sets
tended to be approximately one-half the contour
interval of the input contour. (Strictly speaking,
their comparison measured the root mean square
of the difference between the DEM and the
LIDAR, which overestimates the error of the
DEM.2) That finding seems roughly consistent
with the National Map Accuracy Standard that
90 percent of the well-defined points should be
within one-half contour interval of the stated
elevation (Bureau of the Budget 1947)—
"roughly" because they are not identical: If mean
error is zero, a 90 percent confidence limit will
almost always be a wider interval than the range
defined by an estimate plus or minus the RMS
1 RMS error is calculated by taking the difference between
the estimated and actual values for each point, squaring
that difference, taking the sum of squares, dividing that
sum by the total number of data points, and taking the
square root. If the mean error is zero, RMS error is equal
to the standard deviation of the error. If the mean error is
not zero, then RMS error is equal to the square root of the
sum of (a) the square of the mean error plus (b) the square
of the standard deviation of the error.
2 In general, whenever one has two independent
measurements Mi and M2, with random error ei and e2,
variance(Mi-M2) = variance (ei) + variance (e2).
Thus, the variance of one error is equal to the variance of
the difference minus the variance of the other error.
error. In a normal distribution, the 90 percent
interval would encompass a range ±1.64 times
the RMS error (generally called standard
deviation or o in this case).3 But one would
expect the error across all elevations to be
greater than the error at those elevations where
we have a contour. For example, if a USGS map
says that one contour is 5 feet above the vertical
datum and that another contour is 10 feet above
the vertical datum, and then one estimates an 8-ft
contour through interpolation, we would expect
the USGS contours to be somewhat more
accurate than the 8-ft contour derived from the
two USGS contours. So the assumption that 90
percent of the points along the contour are within
one-half the contour interval of the stated
elevation would be roughly consistent with the
assumption that the standard deviation of error
for all elevations is one-half the contour
interval.4 Because Titus and Wang did not know
whether their estimates have a mean error or not,
the more general term "RMS error" better
describes the uncertainty. The contour intervals
vary from place to place—but we know the
contour interval at all locations. Therefore, this
study assumes that RMS error equals one-half
the contour interval for all locations where
contour maps were the underlying source of the
data.
Given that the availability of an estimate of the
RMS error, this author's first thought was that
the low and high estimates could be derived by
simply (a) adding and subtracting the RMS error
from the DEM5 data set developed by Titus and
Wang, cell by cell, and then (b) retabulating the
data. In effect, this approach would add and
3The RMS error band includes about 68 percent of all data
points.
4In the case of normally distributed error, we are saying, in
effect, that 90 percent of the points along the contour are
within 0.5 contour interval, while 90 percent of all points
are within 0.82 (1.64/2) the contour interval of the stated
elevation.
5DEM is an abbreviation for digital elevation model.
Literally, that means the model used to calculate
elevations. People in the business of making elevation
maps, however, often use this term when referring to the
actual set of elevation data points calculated by their
model. The Titus and Wang data set we used has data
points on a 30-m grid.
-------
[ SECTION 1 .3 73 ]
15 ft
C
o
-4—'
03
>
Q)
LU
10 ft
SHW
Area below
1st contour
subtract the RMS
error from the
cumulative
distribution of
elevations. However,
as those authors
discuss in Section
1.1, their DEM
contained plateaus
along the input
contours, which were
artifacts of the
interpolation
algorithm, with no
physical basis.6
Therefore, they
concluded that a
linear interpolation
of elevations
between the contours
would give a better
estimate of the area
of land below a
particular elevation
than the cumulative distribution of their cell-by-
cell DEM output. Therefore, their elevation
density distribution
assumed that elevations were uniformly
distributed between contours. If the input data
said that there are 100 ha of land between the 5-
and 10-ft contours, for example, then there are
20 ha between the 5- and 6-ft contours, they
assumed. Thus, their cumulative elevation
distribution function was a series of line
segments connecting a few points that represent
actual observations based on the contour interval
and the area of land above spring high water land
below specific contours.7 (See the green line in
Figure 1.3.2, discussed below.)
This study assumes that the same logic that
applies for the "point estimates" would apply to
Green dots represent observation from the maps; Green Line represents interpolated central estim^
Red dots represent positive contour error or negative SHW error
Black dots represent negative contour error or positive SHW error
Solid lines: No SHW error or all three SHW error estimates coincide
Dashed lines: SHW error compounds contour error (relative to SHW)
Dotted lines: SHW error offsets contour error (relative to SHW)
Area below
2nd contour
Area below
3rd contour
Figure 1.3.2. Interpolated Elevation Estimates Relative to NGVD29. Central
estimate and high contour error (with and without SHW error, relative to NGVD,
ignoring model error). This case assumes a 5-ft contour interval, a 1 -ft error in
estimating the elevation of spring high water, and contour error of 2.5 feet. Red
dots represent positive contour error and negative SHW error, both of which cause
a positive error in our estimates of elevation relative to SHW.
EPA's effort to estimate an uncertainty range.
Choosing instead to add or subtract one-half
contour interval from the DEM, would (for
example) create data sets with plateaus at 2.5,
7.5, 12.5, and 17.5 feet in those areas where the
USGS data had a contour interval of 5 feet, just
as the Titus and Wang output had plateaus at
SHW, 5, 10, 15, and 20 feet.8
Let us go back to the source information. For
each quad, Titus and Wang provide
• the areas of land that lie below specific
elevation contours from the input data set
(e.g., the area between the 5- and 10-ft
contours in a given quad), and
• their estimate of the elevation of spring high
water relative to NGVD29 (derived from
NOAA tidal datum).
6See Section 1.1.3 at Step 4, and especially Table 1.1.3 in
Section 1.1.4. The large horizontal error but small vertical
error in replicating contours is indicative of large plateaus.
7In an area with a 5-ft contour interval, those points would
be (SHW, 0), (5, A(5)), (10, A(10)), (15, A(15)), (20,
A(20)) ... etc., where A(x) is the area of land between
spring high water and elevation x.
8Their data set also created plateaus just above their spring
high water supplemental contour. Thus, if spring high
water is 2 feet (NGVD29), then the high-elevation estimate
would have a plateau at 4.5 feet; the low-elevation
estimate would have a plateau at 2.5 feet below spring high
water, that is, -0.5 feet (NGVD29).
-------
[74 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
Area below Area below Area below
1st contour 2*contour 3*contour
Figure 1.3.3. Interpolated Elevation Estimates Relative to Spring High Water. Central estimate and
high contour error (with and without SHW error, relative to SHW, ignoring model error). This case assumes
a 5-ft contour interval, a 1 -ft error in estimating the elevation of spring high water, and contour error of 2.5
feet.
The estimates of the land below various
elevations were based on simple linear
interpolation of this information.9 Figures 1.3.2
through 1.3.4 illustrate a proposed approach to
generating high and low elevation estimates,
respectively. But before discussing that
approach, let us examine a depiction of the Titus
and Wang analysis (see Section 1.1) used as
input to this study. In Figure 1.3.2 (as well as
Figures 1.3.3 and 1.3.4), the four green dots
represent the values of the input data. This
example quad has a 5-ft contour interval, and
spring high water is estimated to be 3 feet above
NGVD29. The first green dot shows the
estimated elevation of spring high water; this dot
9In some cases, the 5-ft contour was seaward of the
wetland boundary and the Titus and Wang interpolation
disregarded the 5-ft contour on the assumption that it was
obsolete. In those cases, the interpolation created—in
effect—a new 5-ft contour farther inland, which was used
in quantifying the land below 5 feet in a given quad.
appears along the vertical axis because all the
dry land and nontidal wetlands are above spring
high water (by definition). The other three points
show the amount of land (other than tidal
wetlands) below the 5-, 10-, and 15-ft contours.
The green line is the cumulative elevation
distributions that Titus and Wang derived
through interpolation—but transposed so that the
cumulative elevation is on the horizontal axis
and elevation on the vertical axis. The figures are
transposed from the traditional way of depicting
cumulative distribution functions, because the
transposed version gives us the actual profile of a
typical transect or cross section of the land.
Now let us consider a possible way to think
about high and low error. In Figure 1.3.2, the
three red dots with elevations of 7.5, 12.5, and
17.5 feet represent high estimates of the
elevation of the contours. That is, given the RMS
error of one-half the contour interval (2.5 feet),
the 5-ft contour could actually be as high as 7.5
feet. Along the vertical axis, we see three dots.
-------
[ SECTION 1 .3 75 ]
C
o
+J
(0
>
0
LU
I
Area below
1st contour
i
Area below
2nd contour
Area below
3rd contour
Figure 1.3.4. Interpolated Elevation Estimates Relative to NGVD29. Central estimate and low contour
error (with and without SHW error, relative to NGVD, ignoring model error). This case assumes a 5-ft
contour interval, a 1 -ft error in estimating the elevation of SHW, and contour error of 2.5 feet
As previously mentioned, the green dot is the
estimate of spring high water (3 feet). The red
and black dots at 2 and 4 feet, respectively,
represent the possibility that Titus and Wang
over- or underestimated SHW, respectively. The
three red lines represent the alternative high-
elevation cumulative elevation distributions (and
average profile) implied by the three different
estimates of the elevation of SHW. In all these
cases, the profile is steeper than the profile
implied by the input data. The dashed line—
where spring high water is less than estimated—
provides the steepest profile and hence the
greatest error. Put another way, the dashed line
assumes that SHW is lower—and the contour is
higher—than assumed by Titus and Wang; i.e.,
the errors compound. Figure 1.3.3 shows the
same four cases, but with elevations relative to
spring high water instead of NGVD29.
Comparing Figures 1.3.2 and 1.3.3 may help one
visualize the impact of SHW error on the land
profile (cumulative elevation distribution)
assumed in the calculations. Each of the four
profiles has the same shape in Figure 1.3.3 as it
has in Figure 1.3.2. When measured against
NGVD (Figure 1.3.2), the three high-contour
error profiles start at different elevations
(reflecting uncertainty about the elevation of the
lowest spot of dry land, SHW) but coincide after
the first contour (because SHW error has no
impact on the topographic contours). When
measured against SHW (Figure 1.3.3), the
profiles all start out at zero, because error in
estimating SHW has no impact on the
definitional assumption that dry land extends
down to SHW. But the profiles diverge because
errors in SHW have a 1:1 impact on elevations
measured relative to SHW. Whatever the true
elevation of the 5-ft contour relative to
NGVD29, overestimating SHW by 1 foot lowers
the estimated elevation relative to SHW by 1
foot.10
"The error of elevations relative to spring high water
would be 1 foot greater if the red dot (in Figure 1.3.2) was
Solid lines: No SHW error or all three SHW error estimates coincide
Dashed lines: SHW error compounds contour error (relative to SHW)
Dotted lines: SHW error offsets contour error (relative to SHW)
5ft
SHW
Green dots represent observation from the maps; Green Line represents interpolated central estimate
Red dots represent positive contour error or negative SHW error
Black dots represent negative contour error or positive SHW error
-------
[76 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
All the figures show the implications of errors in
spring high water and elevation estimates. There
is no reason to think that these errors are
correlated and every reason to assume that they
are independent: two different federal agencies
(USGS and NOAA) compiled the underlying
data.11 Therefore, when calculating uncertainty,
we should assume that these errors are
independent. It follows that the total elevation
error is calculated as the square root of the sum
of squares. Thus, in areas where the contour
error is significant, the error in spring high water
makes very little difference. But in areas with
precise elevation data, error in spring high water
can account for about one-half the total error.
Figure 1.3.4 presents a story similar to Figure
1.3.2 but for the low elevation case. The story is
not completely symmetrical because of the first
contour. The contour interval of the USGS maps
at this location is 5 feet; but it is almost
impossible for the USGS contour to have
overestimated the actual elevations by 2.5 feet.
Substantial dry land ("area below 1st contour") is
above SHW (approximately 3 feet NGVD) and
below the first contour. If the low elevation
estimate were to assume that the lowest contour
is at 2.5 feet, there would be an impossible
result: the land above SHW (3 feet) cannot also
be below 2.5 feet. This analysis avoids such an
anomaly by assuming that RMS error is one-half
the actual contour interval used. Thus, if SHW is
between 2 and 4 feet, the lowest contour interval
is 1 to 3 feet; so the low case assumes that the
lowest contour is between 3.5 and 4.5 feet above
NGVD (depending on the error in estimating
SHW) rather than at 2.5 feet.
Although map accuracy standards provide a
basis for the contour-error assumption, the
literature does not provide a good estimate of
uncertainty for SHW. This exposition has looked
at the case where the error in SHW is 1 foot,
because whole numbers can help simplify
numerical illustrations. Our final results,
however, assume that uncertainty for spring high
water is approximately 15 cm (6 inches). Section
1.1 suggests that error is likely to be less than 6
inches, pointing out that the estimates are based
on interpolation of spring tide ranges from more
than 750 sites, and that the variation from site to
site tends to be about 5 cm (2 to 3 inches), or
less. Within a given quad—the unit of analysis
for this study—those errors should cancel to
some extent, causing the error to be less.
Using an Error Function to Represent Low
and High Cumulative Distributions
The previous discussion explains the low and
high estimates as alternative possibilities for the
average shore profile, given the points along the
profile for which observations are available. That
is, the discussion compared the "best guess"
profile estimated by Titus and Wang, with
proposed high and low profiles. Recall, however,
that although one usually displays y = f(x), in
this case, the argument of the function is shown
on the vertical axis. That is, in Section 1.1, Titus
and Wang estimated the area as a function of
elevation. Similarly, this study needs to estimate
the low and high estimates as a function of
elevation.
For computational purposes, it may be useful to
think of error as a function of the best-guess
central estimate. Viewed together, Sections 1.1
and Section 1.2 estimate the area of land within
each shore protection category within each quad
by 0.1-ft elevation increments. Thus, if one can
express low = f(central estimate) and high =
g(central estimate), then one need merely assign
low and high elevations to each area. That is:
A_loWik.low.f(E) = Aik,E
A_highik.high.g(E) — A;k_E
the actual value, and 1 foot less if the black dot was the
actual value.
11 The Section 1.1 estimates of spring high water are based
entirely on NOAA tidal observations and NOAA analysis
relating mean sea level to the fixed reference elevations
used by topographic data (i.e., NAVD88 and NGVD29).
-------
[ SECTION 1 .3 77 ]
Note: H(A1_low) = contour - k*(contour-SHW),
where k represents the ratio of RMS error to contour H(A1_lowest) = contour - k*(contour-SHWH) H(A1_bitlow) = contour - k*(contour-SHWL)
Interval, typically 1/2 SHWH=SHW-SHW_error SHWL=SHW+SHW_error
Figure 1.3.5. High and Low Estimates as a Function of the Best Guess. The difference between the
red line and the green is the high vertical error; the difference between the black line and the green is the
low vertical error. High error is constant beyond the first contour; low error is constant beyond the second
contour. The vertical scale of this drawing is exaggerated below one contour to better display the
relationships at low elevations.
where Ay,E represents the area of land in the ith
shore protection category in the kthUSGS quad
at elevation E, as estimated in Section 1.212; f
and g are the error functions that express low and
high elevation estimates as a function of the
central estimate of elevation, and A low and
A high represent the areas of land at elevation E
in the low and high elevation cases. Figure 1.3.5
shows the low and high elevations as a function
of the central estimate of elevation, i.e., functions
f and g.
Refinements
Our initial model has two important flaws: it
assumes that precision in modeling a single point
is the same as our precision in estimating the
total, and it ignores the model error of our linear
"Jones and Wang overlaid the elevation data from Titus
and Wang with the shore protection likelihood maps from
an unpublished analysis to create cumulative elevation
distribution functions for each of the shore protection
categories. In effect, they subdivided the cumulative
elevation distribution functions estimated by Titus and
Wang, into the separate cumulative distribution functions
for the different categories of likelihood of shore
protection. Thus, all the uncertainties we analyze here
result from the Titus and Wang analysis; but the actual
input data came from Jones and Wang.
interpolation. Let us examine each of these
issues.
Systematic and random error. Intuitively, one
might assume that the precision with which one
can reasonably estimate the area of vulnerable
land is the same as the precision of the input
data. But that is true only if all errors are
perfectly correlated. If we think that all
elevations are likely to have been over- or
underestimated by the same amount, then the
ability to estimate the total is no more precise
than the ability to estimate the elevation of a
particular location. In such a case, there is no
random error; all error is systematic. But that
should rarely be the case.
Most elevation estimates include both a random
and a systematic component. Along the contour,
random errors would be expected as a human
being attempts to trace a contour while viewing
aerial photographs through a stereoplotter;
systematic error might occur through biases
caused by settings in the instrumentation or by
subsiding benchmark elevations. Between the
contours, systematic errors are likely because the
actual "lay of the land" often departs from what
one would expect from a linear interpolation. In
developed areas, people have often filled and
-------
[78 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
bulkheaded the shore, increasing the amount of
land 50-100 cm above the tides at the expense of
land 0-50 cm above the tides; in undeveloped
areas bluffs occur in some areas, and the land
follows a more gentle slope in other areas.
A sophisticated treatment of this question is
beyond our time and budget constraints.
Therefore, we need a simple parameterization.
Figures 1.3.6 compares the cumulative elevation
distributions of LIDAR collected by the state of
Maryland (see Section 1.1, Jones 2007, and
Jones et al. 2008 for additional details) to the
interpolated results for the area on the Eastern
Shore of Maryland where LIDAR was available
(see Figure 1.3.1), subdivided into four subareas
with varying data quality. The vertical axes omit
magnitudes, which are unimportant for the
purposes here.
The four figures all suggest that systematic error
is well less than one-half the contour interval. In
the areas with a 5-ft contour interval (Figure
1.3.6a), the DEM interpolation is about 1 foot
lower (to the left) than the LIDAR below 3 feet;
but above 4 feet the interpolation and LIDAR are
less than 0.5 feet (15 cm) apart. In the areas with
a 1-m contour (Figure 1.3.6b), the DEM
interpolation and LIDAR are less than 10 cm (4
inches) apart below 1 meter. Above that point,
the DEM interpolation increases to 50 cm greater
than the LIDAR, but the difference is generally
25 cm. In the area that used the Maryland DNR
data—which have an RMS error of 5 feet—the
difference is less than 1 foot (30 cm) below the
10-ft contour (Figure 1.3.6c). It increases to 2.5
feet at the 15-ft contour before declining. In
those areas that rely on USGS 20-ft contours
(Figure 1.3.6d), the DEM underestimates the
elevation by 2 to 3 feet, on average.
These comparisons (as well as the comparison
with North Carolina LIDAR reported by Jones
[2007] and Section 1.1.) lead to two insights
worth applying in this error assessment. First, in
areas the size of a county or two, the cumulative
elevation distribution is within one-half the
nominal RMS error of the data most of the time;
and it almost never exceeds the reported RMS
error. Therefore, one would expect that when
there are many counties (e.g. results for entire
states), the cumulative elevation distribution
would continue to converge and almost never
exceed one-half the nominal RMS error of the
data set. That is, it seems safe to assume that the
systematic error over a large area is no more
than one-half the reported RMS error of the
data. Therefore, this error assessment assumes
that when USGS maps are the input data set, the
low and high estimates are one-quarter the
contour below and above the central estimates
derived by interpolating between those contours
in Section 1.2. that the high error may be greater
than the low error, as displayed in Figures 1.3.2
and 1.3.4.
Model error from linear interpolation. The
potential for linear interpolation to understate
elevations appears to be particularly pronounced
at very low elevations. The approach described
so far assumes, in effect, that below the first
contour, error is proportional to elevation
(relative to SHW). But there is no reason to
assume that precision increases at low
elevations; that was simply an artifact of linear
interpolation in a scheme designed to prevent
assuming the impossible, such as dry land being
below spring high water. These assumptions
seem more defensible on the low end than on the
high end. That is, assuming that the area of land
below elevation X is proportional to X below the
first contour is more unreasonable for the high-
elevation uncertainty than the low-elevation
uncertainty:
-------
[ SECTION 1 .3 79 ]
Elevation (cm)
Elevation (ft) NGVD
Figure 1.3.6. Cumulative Area of Land Close to Sea Level according to USGS National Elevation Data
(NED), interpolation of the Titus and Wang DEM, and State of Maryland's LIDAR in the area where LIDAR
was available (see Figure 1.3.1). The data are divided according to the best available data other than
LIDAR: (a) USGS maps with 5- ft contours; (b) USGS maps with 1 meter contours, (c) 5-foot contours
created from MD-DNR data in areas where USGS maps had 20-ft contours; and (d) USGS 20-ft contours.
See Section 1.1 and accompanying metadata for more details.
-------
[80 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
MD DNR Data (nominal 5-ft)
(/) 25000
0)
¦NED
¦DEM Interpolate
¦Lidar
0 2 4 6 8 10 12 14 16 18 20 22 24
Elevation (ft) NGVD
2500
2000
(/>
0)
USGS 20-ft Contour
re 1500
o
0)
CO
0)
1000
500
NED
DEM Interpolate
Lidar
4 6 8 10 12 14 16 18 20 22 24
Elevation (NGVD ft)
Figure 1.3.6. Cumulative Area of Land Close to Sea Level according to USGS National Elevation Data
(NED), interpolation of the Titus and Wang DEM, and State of Maryland's LIDAR in the area where LIDAR
was available (see Figure 1.3.1). The data are divided according to the best available data other than
LIDAR: (a) USGS maps with 5- ft contours; (b) USGS maps with 1 meter contours, (c) 5-foot contours
created from MD-DNR data in areas where USGS maps had 20-ft contours; and (d) USGS 20-ft contours.
See Section 1.1 and accompanying metadata for more details.
-------
[ SECTION 1 .3 81 ]
Second, the tendency for the DEM interpolation
to underestimate elevations appears to be
somewhat more pronounced than any tendency
to overestimate elevations. In Maryland this is
clearly the case. (Titus and Wang, and Jones,
found that in North Carolina, the interpolation
overestimated elevations of very low land; but
they concluded that the unique situation of North
Carolina was probably to blame in that case.13)
That tends to reinforce our inclination to assume
• The wetlands boundary is at the kink of the
most common concave-up profile. So the use
of wetlands data means that interpolation
already accounts for cases where the profile
is below a linear trend.
• The accuracy assessment shows the Section
1.1 DEM to underestimate elevations close to
spring high water (see Figure 1.3.6):
- In Maryland, they generally found that
more than half of the land between spring
high water and the first contour was
above the midpoint between spring high
water and the elevation of the first
contour.
- The error was particularly great when the
contour interval was large.
• USGS contour selection also creates a
downward bias: Consider an area with a 10-ft
contour. If there is much land below the 5-ft
contour, USGS is likely to reduce the contour
interval to 5 feet or at least collect a 5-ft
supplemental contour. This does not always
occur, but the tendency is enough for a high-
elevation scenario to assume that there is no
land below the 2.5-ft contour.
13Much of North Carolinas coastal wetlands are truly are
classified as nontidal wetlands, and hence the
interpolations in Section 1.1 treated them as uniformly
distributed between SHW and the 5-ft contour, which is
generally more than 1 meter above SHW. (The final results
used LIDAR and hence are not affected directly by this
problem.) Much of those wetlands are at sea level, and
classified as nontidal because the rivers and sounds along
which they are found have an astronomical tide so small
that, for most practical purposes, it is nontidal. When
considering the impact of sea level rise, it would be more
accurate to consider these areas to be "nanotidal wetlands."
• The mathematics limits downside
uncertainty: Because elevations must be
above spring high water, they can only be a
little bit less than the very low elevations
under consideration, while they could be
much higher.
Thus, if the point estimate assumes 100 hectares
within 0.5 feet above spring high water, it is
desirable that the low estimate does not assume
100 hectares to be 2 feet below spring high
water. That does not mean, however, that the
high estimate ought to rule out the possibility
that this land is actually 3 feet above spring high
water. Low bluffs really are common along the
coast—so a high scenario that assumes a low
bluff with an elevation of contour/4 is actually
quite realistic. (By contrast, a high scenario that
assumes an unmapped dike protecting low land
that it contour/4 below spring high water is not
realistic.) Put another way, there is good reason
to not think that there is a large amount of dry
land below high tide—but there is no reason to
think that there is a significant amount of land
just above spring high water. Therefore, the high
scenario should allow for the possibility that
there is no significant amount land barely above
the tides.
Figure 1.3.6 supports this concern. In Figures
1.3.6a and 1.3.6c, the interpolation understates
elevations by about 1 foot below 4 feet in
elevation, and then declines. In Figure 1.3.6d,
where the underlying USGS maps have a 20-ft
contour interval, the interpolation finds as much
land below 3 feet as LIDAR finds below 5 feet,
and as much land below 17 feet as the LIDAR
finds below 20 feet. Thus, at an elevation of one-
quarter the contour interval, the error is about
two-thirds the error seen at the contour. (In
Figure 1.3.6b, the error is fairly minor at all
elevations.)
-------
[82 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
Area below Area below Area below
1st contour 2*contour 3*contour
Green dots represent observation from the maps,
central estimate
Red dots: positive contour error
Black dots: negative contour error
Solid lines: No SHW error
Green Line represents interpolated
Figure 1.3.7. High Elevation Estimates Relative to Spring High Water, including Possible Model
Error (with and without SHW error, relative to NGVD, ignoring model error). This case assumes a 5-ft
contour interval, a 1 -ft error in estimating the elevation of SHW, a contour error of 2.5 feet, and a high-end
error that is always at least one-quarter the contour interval
There is no completely satisfactory way to model
this possibility. The simplest approach would
have been to simply add and subtract one-quarter
the contour interval to the entire distribution, but
this analysis employs a more complicated
approach in part to avoid impossible results in
the low case (e.g., dry land up to one-quarter the
contour interval below SHW). But this is not a
problem with the high scenario. Therefore, the
high scenario assumes that all land is at least
one-quarter times the contour interval above
SHW. In effect, the high estimate assumes that
one can not rule out a bluff with an elevation at
one-quarter the lowest contour interval.
Comparing Figure 1.3.7 to Figure 1.3.3 shows
that this assumption has no impact on elevations
above the first contour.
Areas with Higher Precision Data
In areas with higher precision data, these
considerations are less important. They mostly
apply to problems between contours; and EPA
does not need elevations in increments finer than
50 cm. What is important is that no matter how
precise the elevation data, we will report some
uncertainty because LIDAR measures elevations
relative to a fixed reference plane, while we
report elevations relative to spring high water,
which we estimate imprecisely. As mentioned
above, this analysis assumes that the estimates of
spring high water have an error of 15 cm (6
inches).
In Section 1.1, Titus and Wang used the LIDAR,
spot elevation, and actual DEM results where
contour intervals were 2 feet (60 cm) or less.
Therefore, the interpolation model did not apply
and it would be reasonable to simply add or
subtract the systematic error. We saved some
time, however, by applying the algorithm
developed for USGS data to these results as well
rather than rewriting a separate algorithm.
-------
Section 1.3.2. Implementing the Approach using
Geographically Specific Error Functions Approach
Author: Dave Cacela, Stratus Consulting Inc.
The objective of elevation uncertainty analyses is
to acknowledge uncertainty about the actual
elevation of any particular geographic region and
to quantify it so that the elevation in a particular
region can be expressed as a range of plausible
values. Consequently, estimates of flooded areas
under any particular scenario of sea level rise can
also be expressed as a range of plausible values.
This section reports the actual methods used to
calculate ranges of plausible elevation that
reflect the reasoning about landscapes,
interpretation of map accuracy, and between-
contour interpolation methods described in
Section 1.3.1. It is intended to describe the
essential features of methodology introduced in
Section 1.3.1 that were actually applied in the
uncertainty analysis in a manner that includes
specific mathematical definitions that allow for
reproducibility.
The reasoning in Section 1.3.1 about uncertainty
is described in terms of two generalized error
functions. One of the functions defines the lower
limit of plausible elevation and the other defines
the upper limit. Considered jointly, the error
functions define the amount of uncertainty about
elevation (vertical error) associated with any
geographic point. To quantify uncertainty in a
particular geographic location, the generalized
error functions are used with parameters that are
specific to that particular location to define
plausible ranges of elevation for that location.
Plausible ranges of elevation determine in this
manner are subsequently translated into plausible
ranges of area that may be inundated by various
sea level rise scenarios.
Magnitude of Uncertainty in the
Data Sources
Uncertainty analyses consider two main sources
of uncertainty. The analyses consider both types
of uncertainty jointly to generate an estimate of
total uncertainty that is specific to each
geographic area in the study.
One source of uncertainty derives from
imprecision in elevation values in the source
data. Each location in the study area is
represented by one of several types of source
data with differing amounts of inherent
precision. As described in Section 1.3.1, the
inherent precision of each type of source data is a
known value that is expressed as the root mean
square error (RMSE) and in the same units of
measure as the vertical units provided (Table
1.3.1). Data with greater inherent precision have
less uncertainty with regard to the true elevation
of a particular geographic point and, conversely,
source data with lesser inherent precision have
more uncertainty with regard to the true
elevation of a particular point. (See Figure 1.3.8
and Table 1.3.1; and Section 1.1 and Section 1.2
for additional details concerning the precision of
the source data used in the study area.)
The second source of uncertainty derives from
the estimated elevation of SHW relative to the
NGVD29 for any particular section of coastline
as derived from local tide gage data. The
elevation of SHW is relevant because the
elevations provided by the source data are
expressed relative to the NGVD29 datum, but
the estimation of inundation is expressed relative
to SHW (see Section 1.1 for a description of how
the elevations relative to SHW were derived).
-------
[84 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
Aggregate Uncertainty
All NGVD29 elevations from the source data are
converted to elevations relative to SHW by:
Ejk = Engvdjk - SHWk (1)
where:
Ejk is the derived nominal elevation of point j
in region k relative to SHW
Engvdjk is the nominal elevation of point j in
region k relative to NGVD29, as provided in
the source data
SHWk is the estimated (NGVD29) elevation
of SHW for region k.
SHWk is not known with absolute certainty; thus
the precision of Ejk is a function of two sources
of uncertainty: (1) the magnitude of uncertainty
inherent in Engvd,jk and (2) the magnitude of
uncertainty in SHWk. In principle, the magnitude
of uncertainty in SHWk could vary by region k,
but in this study SHWk is defined as a constant
value of 0.5 feet. These two sources of
uncertainty were assumed to be statistically
independent; thus, the magnitude of total
uncertainty is estimated with the basic equation:
jk '
P' --l/Pmshw,k +Pngviik (^)
where:
Pmshw.k is the magnitude of uncertainty in
SHWk expressed as RMSE, defined as a
constant value of 0.5 feet
Pngvd.jk ^Cjk Or
Pngvd.jk is a specified the magnitude of
uncertainty in Engvd.jk expressed as RMSE
(feet)14
Pjk is the magnitude of total effective
uncertainty in Ejk (feet)
14For areas described by some types of source data, e.g.,
USGS topographic maps, Pngv4jk is defined as a certain
fraction of the contour intervals used in the base maps, but
for other types of source elevation data not based on
contour intervals, e.g., elevations derived from LIDAR
data, Pngvdjk is a constant (Table 1.3.1). For USGS maps,
P i ig\ ¦- i.j i kC.
Cjk is the magnitude of contour intervals
represented in the relevant source data for
point j,k15
A- is a scalar that varies by source data (e.g.,
0.5; see Table 1.3.1).
(1)
The basic definition of Pjk was not applied
universally to all points in region k. In some
subregions within region k, Pjkis associated with
points j,k, but in other subregions, particularly
regions of low elevation, Pjkis redefined by an ad
hoc function of Ejk that is described below.
Estimating Elevation Uncertainty
The magnitude of uncertainty about Ejk was
defined as Pjk at all relatively high elevations. In
such regions, upper and lower bounds on Ejk
were defined simply as:
Ejki = Ejk-Pjk (3)
Ejku = Ejk + Pjk (4)
where:
Ejk is the nominal elevation of point j,k 16
Ejki and Ejk.u represent the lower and upper
bounds on Ejk, respectively.
However, the simple formulations in Eqult&bns 3
and 4 were considered inadequate for providing
realistic bounds for Ejk in locations with low
elevation, where "low elevations" are defined to
be lower than selected reference elevations. For
estimating Ejku, a reference elevation was taken
to be E'jk , the elevation of "first contour," which
is Ejk corresponding toE„gvd,jk equal to the lowest
nonzero elevation contour in the source data for
region k. For estimating Ejk.i, an additional
reference elevation was taken to beE"jk, the
elevation of "second contour," which is Ejk
corresponding to Engvd.jk equal to the second-
lowest nonzero elevation contour in the source
data for region k.
"For source data not based on a contour interval, such as
SPOT and LIDAR, contour interval was derived from the
RMSE of the source data.
" Nominal elevations were determined from the source
data using interpolation methods described in Section 1.1.
-------
[ SECTION 1 .3 85 ]
The general uncertainty modeling procedure can
be succinctly described as two complex error
functions. One such function describes the error
in a positive direction, i.e., the amount by which
the "true" elevation, E*jk, could exceed the
nominal elevation Ejk. The other such function
describes the error in a negative direction, i.e.,
the amount by which the "true" elevation, E*jk,
E'jk =(Cjk -MSWHk)
P
]k,u
'jk
gE
jk
jk ¦
J'k
jk
E =E + P
Ljk,u Ljk jk,u
could lie below the nominal elevation Ejk. The
functions are asymmetrical because of the
assumption that the magnitude of errors in the
negative direction will tend to be relatively
dampened if Ejk is lower than E'jk or E"jk
(defined below; see Section 1.3.1 for the
justification of this assumption).
The error function for determining an upper
bound on Ejkis a set of line segments defined as:
(5)
(6)
(?)
IfE]kE]k
where:
SHW is the elevation of mean spring high water for point j,k
gis a constant (e.g., 0.25)
E'jk is the elevation (relative to SHW) of "first contour"
Pjk.u is the magnitude of error in a positive direction
Ejk,u is the upper bound on Ejk-
The error function for determining a lower bound on Ejkis a set of line segments defined by:
E"jk=(2Cjk-MSWHk)
P' =
r Jk
P/A-j -
P' F
r jk jk
P'
mshwr+(*E'Jky
EV-
jk
((E]k-Wlk)(?lk-V'lk))/ (E"lk-Wlk)
jk'' ' jk
jk;
jk
' jk '
jk
If v., >Qand\:... H' ; andY., .. Y.",
//ejA->ev
(8)
(9)
(10)
E jk j = max (0, (Ejk - Pjk j)) (11)
where:
P'jk is a measure of uncertainty analogous to Pjk
E"jk is Ejk corresponding to E"ngvdjk, the elevation of the second-lowest non-zero elevation
contour in the base map for region k
Pjkj is the magnitude of error in a negative direction
Ejk,i is the lower bound on Ejk-
The typical shape of the error functions defined by Equations 1 through 11 are depicted in Figure
1.3.8.
-------
[86 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
(a)
Nominal Elevation (above MSHW)
(b)
Nominal Elevation (above MSHW)
(c)
0 E'jk E"jk
Nominal Elevation (above MSHW)
Figure 1.3.8. Generalized Error Functions Used to Estimate Uncertainty Bounds on Elevation.
Panel (a) depicts magnitude of uncertainty in a positive direction; panel (b) depicts magnitude of
uncertainty in a negative direction; and panel (c) describes the net effect of the functions depicted in
panels (a) and (b), expressed as positive and negative uncertainty bounds relative to the nominal
elevation.
-------
[ SECTION 1 .3 87 ]
Estimating Ranges of Plausible
Elevation
Before the uncertainty analyses, acreages for a
particular region and protection scenario were
compiled into bins corresponding to elevations
above SHW 0.1-ft increments.17 For example, for
scenarios,
Ak.s.o.i = area between Ejk = 0 feet andEjk = 0.1
foot (hectares)
Ak.s.o.2= area between Ejk= 0.1 feet and Ejk=
0.2 foot (hectares), etc.
Thus, collectively the Ak.s values can be
considered as a density18 with each element
associated with a particular Ejk- Considering the
meaning of Ejk.i and Ejk.u, each Ak,s can be
associated with all three values: Ejk, Ejk.i, and
Ejk.u. By extension, each Ejk elevation can be
associated with three alternative values of Ak,s by
aligning with cases where Ejk = Ejk,i and Ejk =
Ejk,u. In this manner, two additional "densities"
are generated such that for each Ejk there are
three alternative corresponding Ak.s. The
alternative densities have little implicit meaning,
but converting each of the alternative densities to
cumulative distributions provides alternative
elevation profiles that are meaningful for
generating a range of estimates of total flooded
area under various amounts of sea level rise.
Procedural Notes
Data processing and calculations related to the
elevation uncertainty analyses were conducted
with S-Plus software (Professional Developer
version 7; Insightful Corporation, Seattle, WA).
In addition to quality control procedures used
during development of the S-Plus algorithms
used to solve for the uncertainty endpoints,
quality control procedures were conducted
independently from the S-Plus algorithms using
MS-Excel spreadsheets for selected test cases.
17The data used as the basis for the uncertainty analyses
were expressed with a resolution of 0.1 feet (see footnote
14), and the general processing of those data to develop
uncertainty limits were conducted with a resolution of 0.1
feet. Prior to comparisons with elevations of interest (e.g.,
a selected amount of sea level rise), the basic results with
0.1 foot resolution were further subdivided into 10 bins of
equal size to provide a quasi-resolution of 0.01 feet.
18Not strictly a probability density because the sum of all
Hk,s equal a total area in region k for scenarios, not one.
-------
[88 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
Table 1.3.1. Features of distinct base map data sources related to estimation of elevation uncertainty
-------
[ SECTION 1 .3 89 ]
Table 1.3.1. Features of distinct base map data sources related to estimation of elevation uncertainty
-------
[90 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
Table 1.3.1. Features of distinct base map data sources related to estimation of elevation uncertainty
-------
[ SECTION 1 .3 91 ]
Table 1.3.1. Features of distinct base map data sources related to estimation of elevation uncertainty
-------
[92 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
Table 1.3.1. Features of distinct base map data sources related to estimation of elevation uncertainty
-------
[ SECTION 1 .3 93 ]
Table 1.3.1. Features of distinct base map data sources related to estimation of elevation uncertainty
-------
[94 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
Table 1.3.1. Features of distinct base map data sources related to estimation of elevation uncertainty
NJ Perth Amboy Middlesex 20 ft 20 0.5 0.25 304.8 4.07
-------
[ SECTION 1 .3 95 ]
Table 1.3.1. Features of distinct base map data sources related to estimation of elevation uncertainty
NY New London Suffolk 10 ft 10 0.5 0.25 152.4 2.15
-------
[96 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
Table 1.3.1. Features of distinct base map data sources related to estimation of elevation uncertainty
VA Chincoteague East Accomack 5 ft 5 0.5 0.25 76.2 2.81
OeS
VA Chincoteague West Accomack 5 ft 5 0.5 0.25 76.2 1.71
VA Cobb Island Northampton 5 ft 5 0.5 0.25 76.2 2.84
VA Courtland Southampton 5 ft 5 0.5 0.25 76.2 1.7
-------
[ SECTION 1 .3 97 ]
Table 1.3.1. Features of distinct base map data sources related to estimation of elevation uncertainty
-------
[98 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
Table 1.3.1. Features of distinct base map data sources related to estimation of elevation uncertainty
VA King and Queen Court King and Queen 10 ft 10 0.5 0.25 152.4 2.98
House
VA Machodoc Westmoreland 10 ft 10 0.5 0.25 152.4 1.61
-------
[ SECTION 1 .3 99 ]
Table 1.3.1. Features of distinct base map data sources related to estimation of elevation uncertainty
VA Rappahannock Caroline 10 ft 10 0.5 0.25 152.4 2.56
Academy
-------
[ 100 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
Table 1.3.1. Features of distinct base map data sources related to estimation of elevation uncertainty
Contour
RMS
interval
k
cm
SHW
State
Quadrangle
County
Source
(ft)
(base)3 g
(base)
(ft)b
VA Zuni Isle of Wight 10 ft 10 0.5 0.25 152.4 1.76
a. The values of k listed here are the "base" value of k that relates contour interval to RMS as RMS = k(base) H
contour interval. The procedures for conducting uncertainty analyses allow for universal rescaling of k. These
values were scaled by a factor of 0.5 in the analysis; i.e., we assume that error = 0.25 times the contour
interval in most quads.
b. For these locations, the values of 1 for contour interval and 0 for SHW were provided to trick the algorithm
into calculating "contour error" as RMSE/2. This was necessary because of the format in which Jones and
Wang had saved the central estimate results for those areas with high precision data. The value of g doe not
matter because g had no effect above the contour interval, which is less than the 50 cm increment with
which our results are reported.
c. For these locations, the values of 2 for contour interval and 0 for SHW were provided to trick the algorithm
into calculating "contour error" as RMSE/2. This was necessary because of the format in which Jones and
Wang had saved the central estimate results for those areas with high precision data.
-------
Section 1.3.3. Results Author: James G. Titus
The results from this section are displayed in
Appendices A, B, and C (along with regional
summaries). What we call ''low" and
"high "(elevation) as we explain oar approach in
this section are reversed in the tables, because
the high elevation means less vulnerability and a
lower area close to sea level, and vice versa.
We encourage the reader to examine these tables
and think about both the ratio of the high to the
low estimate and the vertical error implied by a
given line in the table. If the high estimate at 50
cm is greater than the low estimate for 100 cm,
then the vertical error is greater than 25 cm. If
the high estimate at 50 cm is greater than the low
estimate for 150 cm, then the vertical error is
greater than 50 cm.
If the ratio of high to low at 50 cm is great, that
may mean that the uncertainty is great; but it
may also mean that there is an inflection point
nearby. For example, if the data (e.g., LIDAR)
show several times as much land between 50 and
65 cm as between 0 and 50 cm, then even if the
error is only 15 cm, the ratio of high to low could
be very large.19 This happens in some areas with
LIDAR. As a result, if one considers only the
ratio of high to low, one might be surprised that
the areas with LIDAR do not always seem much
more precise than the areas that relied on USGS
5-ft contours. (A second reason that the LIDAR
does not always appear more precise than areas
with 5-ft contours is that the first contour interval
is only 2-3 feet in areas where spring high water
is 2-3 feet above NGVD29. Although the
subsequent contour intervals are greater, the
ratios of high to low get closer to 1 as elevations
19For example, if the LIDAR shows 10 ha between 0 and
50, and 100 ha between 50 and 65, if error is 15 cm, the
high estimate would be 110 ha, and the low estimate would
be less than 10. The ratio of high to low would this be
more than 11.
increase.) Nevertheless, variations in precision
are palpable when one looks at areas with a 10-
or 20-ft contour interval. See the Pennsylvania
tables in Appendix A, where Bucks County has
mostly 20-ft contours but Philadelphia has 2-ft
contours.
Overall, we estimate between 2,374 and 3,221
square kilometers of land within 50 cm above the
tides, and 3,351 to 3,940 square kilometers
within 1 meter above the tides in the middle
Atlantic (see Appendix C). Our input data and
assumptions are based on RMS error; but at the
state and regionwide level, much of the errors
should cancel. The true amount of land close to
sea level is very likely to fall within the ranges
we have estimated.
One final warning: The available output
provided by Jones and Wang (explained in
Section 1.2), which this effort used as input,
extended only to an elevation of 20 feet above
SHW. Therefore, we cannot literally apply our
formula for the high-elevation (low-area) case
for elevations above 20 feet minus "error." In
cases with a 20-ft contour interval, error is 5 feet;
so we cannot apply the low-area formula above
15 feet. The algorithm explained in Section 1.3.2
treats no data as zero, assuming in effect that
there is no land above 20-ft SHW. We
considered suppressing all calculations above 4.5
meters in such cases, but opted instead to
provide the results with an asterisk. That
approach seems more reasonable: In these cases,
assuming that there is no land above 20-ft SHW
is clearly an extreme lower bound. But we doubt
that it seriously distorts the statewide results.
Typically, a state has only a few quads with a 20-
ft contour interval—generally in areas that have
very little low land. So even if we had been able
to correctly apply our formula (i.e., if Jones and
Wang in Section 1.2 had interpolated above 20-ft
SHW) the calculated area would not be much
-------
[ 102 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
greater than zero when considered at the
statewide level. Thus, instead of suppressing our
"low area" estimate, we provide an estimate that
is slightly lower than a rigorous application of
our approach.
References
Bureau of the Budget. 1947. National Map
Accuracy Standards. Government Printing
Office, Washington D.C. Available from:
http://rockyweb.cr.usgs.gov/nmpstds/nmas.html.
Accessed October 1, 2006.
Environmental Protection Agency. 1989. The
Potential Effects of Global Climate Change on
the United States. Report to Congress. EPA-230-
05-89-050. U.S. EPA, Washington, D.C.
Jones, R. 2007. Accuracy Assessment of EPA
Digital Elevation Model Results. Memorandum
and attached spreadsheets prepared for the U.S.
EPA under Work Assignment 409 of EPA
Contract #68-W-02-027. Distributed with the
elevation data.
Jones, R., J. Titus, and J. Wang. 2008. Metadata
for Elevations of Lands Close to Sea Level in the
Middle Atlantic Region of the United States.
Metadata accompanying Digital Elevation Model
data set. Distributed with the elevation data.
Park, R.A., M.S. Treehan, P.W. Mausel, and
R.C. Howe. 1989. The effects of sea level rise on
U.S. coastal wetlands. In The Potential Effects of
Global Climate Change on the United States.
EPA-230-05-89-050. U.S. EPA, Washington,
DC.
Titus, J.G., AND M.S. Greene. 1989. An
overview of the nationwide impacts of sea level
rise. In The Potential Effects of Global Climate
Change on the United States. Report to
Congress. Appendix B: Sea Level Rise. EPA-
230-05-89-052.U.S. EPA, Washington, DC.
-------
Appendix A
Low and High Estimates of the Area of Land Close to Sea Level, by State3
(square kilometers)
al_ow and high are an uncertainty range based on the contour interval and/or stated root mean square error (RMSE)
of the input elevation data. Calculations assume that half of the RMSE is random error and
half is systematic error. For a discussion of these calculations, see Section 1.3 of this report.
-------
[ 104 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
Table A.1 Low and High Estimates of the Area of Land Close to Sea Level in New York
This value is probably too low because of a data limitation. See Section 1.3 of this report.
Note: A peer reviewer noticed that the draft maps showed Gardiners Island as "likely" even though the text said that it had been changed to "unlikely". The effect of that error was to
overstate the area of land below one meter where shore protection is likely, and understate the area where shore protection is unlikely, by 0.7, 0.9, and 1.1 square miles for the land
within 50, 100, and 200 cm above spring high water. We corrected the maps, but not the quantitative results in this report.
-------
[ SECTION 1.3 105 ]
Table A.2 Low and High Estimates of the Area of Land Close to Sea Level in New Jersey
Meters above Spring High Water
low high low high low high low high low high low high low high low high low high low high
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
probably too low because of a data limitation. See Section 1.3 of this report.
-------
[ 106 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
Table A.2 Low and High Estimates of the Area of Land Close to Sea Level in New Jersey (continued)
This value is probably too low because of a data limitation. See Section 1.3 of this report.
-------
[ SECTION 1.3 107 ]
Table A.3 Low and High Estimates of the Area of Land Close to Sea Level in Pennsylvania
This value is probably too low because of a data limitation. See Section 1.3 of this report
-------
[ 108 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
Table A.4 Low and High Estimates of the Area of Land Close to Sea Level in Delaware
-------
[ SECTION 1.3 109 ]
Table A.5 Low and High Estimates of the Area of Land Close to Sea Level in Maryland
Meters
above Spring High Water
low
high
low
high
low
high
low
high
low
high
low
high
low
high
low
high
low
high
low
high
County
0
.5
1
0
1
5
2
0
2.5
3.0
3.5
4.G
4.
5
5.
0
--Cumulative
(total) amount
of Dry Land below a given elevation-
Anne Arundel
1.7
7.2
6.7
15
12
26
20
39
32
50
44
59
54
68
63
77
72
86
81
94
Baltimore County
2.3
6.6
7.3
13
14
20
21
27
28
36
37
46
47
56
57
65
66
73
75
81
Baltimore City
0.2
2.1
0.9
3.9
1.7
5.7
2.7
7.5
4.2
9.7
5.7
12
7.4
14
9.6
17
12
19
14
21
Calvert
0.4
3.9
1.7
5.8
3.1
7.6
4.6
10
6.1
14
7.6
17
10.0
21
14
26
17
31
21
36
Caroline
0.7
3.2
2.2
6.1
4.1
9.2
6.9
13
9.9
16
13
20
16
23
19
27
23
30*
26
33*
Cecil
0.2
2.5
1.0
5.2
1.8
7.9
3.7
12
5.7
16
7.8
20
11
25
16
29
20
34
24
38
Charles
0.7
12
4.8
21
9.0
30
15
40
22
53
30
67
40
77
53
85
66
93
77
99
Dorchester
30
120
150
215
231
269
282
313
322
348
358
386
396
416
423
439
445
457
462
474
Harford
0.7
17
7.6
25
15
33
22
40
28
49
34
57
42
64
50
69
59
74
65
78
Howard
0
0.01
0.01
0.03
0.01
0.05
0.02
0.07
0.04
0.1
0.05
0.14
0.07
0.2
0.1
0.2
0.1
0.3
0.2
0.3
Kent
0.2
8.4
4.8
16
10
23
16
33
23
45
29
56
37
68
48
80
59
93
71
105
Prince George's
0.2
2.2
0.9
3.9
1.6
5.6
2.9
7.2
4.3
8.9
5.6
11
7.1
13
8.9
16
11
19
13
21
Queen Anne's
0.6
4.1
5.3
12
14
22
24
35
37
50
52
68
69
88
89
107
107
126
125
143
Somerset
17
58
70
101
113
153
168
193
198
210
215
233
240
260
268
289
297
318
327
345
St. Mary's
2.4
16
8.0
28
14
41
24
58
35
79
46
101
62
118
83
129
104
139
120
148
Talbot
2.2
7.8
11
24
30
54
64
99
110
139
149
175
184
210
218
239
245
260
266
279
Wicomico
5.0
15
18
29
32
43
47
58
62
72
76
86
90
101
105
115
119
129
133
142
Worcester
4.4
21
25
48
53
83
88
119
124
153
158
183
187
209
213
235
239
261
265
288
Statewide
69
307
326
570
560
832
812
1104
1053
1350
1267
1596
1500
1833
1737
2045
1960
2243*
2165
2425*
This value is probably too low because of a data limitation. See Section 1.3 of this report
-------
[ 110 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
Table A.5 Low and High Estimates of the Area of Land Close to Sea Level in Maryland (continued)
low
high
low
high
low
high
low
high
low
high
low
high
low
high
low
high
low
high
low
high I
County
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
Wetlands
Tidal
Cumulative (total) amount of Nontidal Wetlands below a given elevation
Anne Arundel
12
0.2
0.7
0.6
1.6
1.1
4.8
3.1
8.1
6.3
11
9.5
12
12
14
13
15
14
16
15
17
Baltimore County
10
0.1
0.3
0.3
0.7
0.7
1.0
1.0
1.3
1.3
1.5
1.5
1.7
1.7
1.8
1.8
2.0
2.0
2.2
2.2
2.3
Baltimore City
0.2
<0.01
0.03
0.01
0.04
0.02
0.05
0.03
0.1
0.04
0.1
0.05
0.1
0.06
0.1
0.06
0.1
0.07
0.1
0.08
0.1
Calvert
15
0.1
0.9
0.4
1.3
0.7
1.7
1.1
2.2
1.4
3.0
1.7
3.8
2.2
4.7
3.0
5.7
3.8
6.6
4.7
7.5
Caroline
14
0.3
1.4
0.7
2.6
1.3
4.0
2.5
5.3
3.5
6.4
4.4
7.5
5.3
8.6
6.2
9.8
7.1
11*
8.0
12*
Cecil
13
0.01
0.2
0.04
0.7
0.1
1.2
0.4
1.7
0.8
2.3
1.2
2.8
1.7
3.5
2.2
4.2
2.8
4.9
3.5
5.5
Charles
24
0.1
3.8
1.5
6.5
2.9
9.2
4.8
12
7.0
14
9.3
16
12
18
14
20
16
21
18
23
Dorchester
425
15
46
53
70
76
90
94
104
107
112
114
121
124
129
131
136
137
139
140
143
Harford
29
0.2
2.5
1.2
3.8
2.3
5.0
3.3
6.2
4.3
7.6
5.2
9.0
6.4
10
7.8
11
9.1
11
10
12
Howard
0
0
0.03
0.01
0.04
0.02
0.04
0.03
0.05
0.04
0.06
0.04
0.06
0.05
0.07
0.06
0.08
0.06
0.09
0.07
0.10
Kent
18
0.1
1.1
0.9
2.6
2.0
4.1
3.3
5.4
4.3
6.8
5.2
7.9
6.1
9.3
7.2
11
8.3
13
9.7
14
Prince George's
14
0.1
0.8
0.3
1.4
0.6
2.0
1.0
2.5
1.5
3.2
2.0
3.8
2.5
4.7
3.2
5.6
3.8
6.5
4.6
7.2
Queen Anne's
21
0.2
1.1
1.5
3.0
3.2
4.8
4.9
6.5
6.5
8.1
7.9
9.6
9.5
12
11
14
13
16
15
18
Somerset
265
6.6
16
17
21
23
31
35
40
41
43
45
52
54
60
62
69
71
78
81
90
St. Mary's
19
0.5
2.8
1.7
5.3
2.8
7.8
4.6
11
6.7
15
8.8
19
12
22
16
25
20
28
23
31
Talbot
26
0.1
0.3
0.5
1.0
1.3
2.1
2.5
4.2
4.8
6.2
6.8
8.5
9.1
12
13
15
16
17
18
20
Wicomico
67
5.4
9.9
11
13
16
22
24
29
30
35
37
44
47
54
56
60
62
66
67
70
Worcester
142
0.7
5.2
6.0
10
11
16
17
22
23
29
30
36
37
42
43
48
49
54
54
58
Statewide
1116
29
93
97
146
145
207
203
261
249
304
289
355
341
406
390
451
435
490*
474
531*
Cumulative (total) amount of land below a given elevation
Dry Land
69
307
326
570
560
832
812
1104
1053
1350
1267
1596
1500
1833
1737
2045
1960
2243*
2165
2425*
Nontidal Wetlands
29
93
97
146
145
207
203
261
249
304
289
355
341
406
390
451
435
490*
474
531*
All Land
1116
1214
1516
1539
1832
1820
2155
2130
2481
2418
2769
2672
3067
2957
3354
3243
3612
3510
3849*
3754
4071*
This value is probably too low because of a data limitation. See Section 1.3 of this report
-------
[ SECTION 1.3 111 ]
Table A.6 Low and High Estimates of the Area of Land Close to Sea Level in Washington, D.C.
-------
416
257
159
49
5
4
18
8.8
49
14
0.5
43
0.8
5.2
378
112
57
141
68
392
53
33
36
39
111
ES ASSOCIATED WITH EPA'S ESTIMATES ]
mates of the Area of Land Close to Sea Level in Virginia
low high low high low high
0.5 1.0 1.5
Meters above Spring High Water
low high low high low high low high
2.0 2.5 3.0 3.5
56 111
37 78
20
2.8
33
10
0.1 0.5
0.3 0.9
1.1 3.9
0.5 2
1.8 6.5
0.8 2.7
0.04 0.1
1.5 5.4
0.05 0.2
0.3 0.9
11 43
2.4 9.3
2.4 8.9
2.8 11
3.6
33
2
14
89
7.3
1.7 5.7
0.9 3.2
2 6.8
13 33
-Cumulative (total) amount of Dry Land below a given elevation-
93 159
65 115
29
6.3
44
15
0.3 0.7
0.6 1.3
2.5 5.9
1.2 3
4.1 9.9
1.7 4.2
0.1 0.2
3.3 8.1
0.1 0.3
0.6 1.3
27
5.7
5.5
6.9
8.5
66
14
13
17
21
66 139
4.6 11
3.7 8.6
2 4.8
4.4
26
11
50
137 204
98 149
39
9.7
55
20
0.5 1.3
0.9 1.7
3.8 7.6
1.9 3.9
6.4 14
2.7 5.4
0.1 0.2
5.2 11
0.2 0.3
0.9 1.8
42 92
9
8.7
11
14
21
18
24
28
108 190
7.1 15
5.5
3.1
7
41
12
8.4
14
67
180
131
49
13
0.6
1.2
5.2
2.6
8.7
3.6
0.2
7.1
0.2
1.2
58
12
12
15
19
149
9.7
7.5
4.2
10
55
243
172
71
25
1.9
2.1
9.2
4.7
20
6.8
0.3
16.7
0.4
2.3
141
37
25
44
35
230
22
15
13
19
76
221
160
61
17
0.8
1.5
6.6
3.3
11
4.6
0.2
9
0.3
1.5
74
16
15
19
24
186
12
9.6
5.4
13
67
279
192
87
29
2.6
2.5
11
5.5
26
8.1
0.3
22
0.5
2.8
190
53
32
64
42
268
28
19
18
23
84
258
180
78
21
1.4
1.8
8
4
15
5.7
0.3
12
0.3
1.9
100
24
20
27
29
220
17
13
9.6
16
75
315
211
104
34
3.3
2.9
12
6.3
31
9.4
0.4
27
0.5
3.3
239
69
38
84
48
307
34
22
22
27
93
294
200
94
25
2.1
2.2
9.5
4.8
20
6.9
0.3
17
0.4
2.4
147
39
26
46
36
258
22
16
14
20
84
-------
[ SECTION 1.3 113 ]
Table A.7 Low and High Estimates of the Area of Land Close to Sea Level in Virginia (continued)
low
C
high
1.5
low
1
high
.0
low high
1.5
low
high
2.0
low high
2.5
low high
3.0
low high
3.5
low high
4.0
low high
4.5
low high
5.0
Mathews
4.7
15
13
33
26
54
44
73
62
85
79
97
90
108
101
113
111
117
115
121
Hampton Roads
24
91
78
200
154
333
264
469
381
650
519
848
711
1045
907
1192
1089
1307
1215
1424
James City
0.1
3.8
2
7.2
4.7
11
7
14
9.4
18
12
22
15
26
19
30
23
34
27
39
York
1.4
6
5
13
9.9
21
16
28
23
33
28
37
33
42
38
45
42
48
44
51
Newport News
2.2
6.9
6
11
9.7
15
13
18
16
21
19
25
23
28
26
33
30
38
35
42
Poquoson
1.4
4.5
4
8.8
7.4
13
11
16
15
16
16
17
17
17
17
17
17
17
17
17
Hampton
1.9
5.9
5
18
13
32
25
45
38
60
51
74
65
88
80
93
90
98
95
102
Surry
0
1.4
1
2.7
1.7
4.1
2.7
5.3
3.6
6.2
4.6
7.1
5.5
8
6.4
9
7.2
9.9
8.1
11
Isle of Wight
0.2
3.4
2
6.2
4.2
9.1
6
12.8
8
17
10
22
14
26
18
31
22
35
27
42
Norfolk
1.9
5.8
5
17
13
30
24
42
35
67
52
91
77
115
101
120
118
124
122
128
Virginia Beach
9.3
33
30
69
55
117
94
163
138
219
185
273
241
327
295
368
347
393
378
418
Suffolk
0.7
4.3
3.1
7.1
5.4
10
7.5
15
10
23
13
31
21
39
28
50
37
60
47
73
Portsmouth
1.2
3.9
3.5
9.6
7.6
15
13
22
18
33
27
45
38
56
50
61
58
65
63
70
Chesapeake
3.5
12
11
31
22
57
45
87
69
137
100
205
162
272
229
337
298
385
353
430
Other Jurisdictions
0
9.9
5.7
19
12
29
19
40
26
54
32
67
44
80
56
93
68
106
81
122
Charles City
0
3.2
1.8
6.3
4
9.6
6.2
13
8.4
18
11
23
15
28
19
32
23
37
28
43
Chesterfield
0
1.3
0.8
2.6
1.7
3.9
2.5
4.8
3.4
5.5
4.3
6.2
5
7
5.7
7.7
6.3
8.4
7
8.9
Colonial Heights
0
0.04
0.02
0.1
0.05
0.1
0.07
0.12
0.09
0.14
0.12
0.15
0.1
0.2
0.1
0.2
0.15
0.19
0.16
0.24
Hanover
0
0.02
0.02
0.05
0.03
0.1
0.05
0.2
0.1
0.3
0.1
0.4
0.2
0.5
0.3
0.6
0.4
0.7
0.5
0.7
Henrico
0
0.8
0.5
1.5
1
2.3
1.5
2.8
2
3.2
2.5
3.7
2.9
4.1
3.3
4.6
3.8
5.1
4.2
6.3
Hopewell
0
0.4
0.2
0.8
0.5
1.1
0.7
1.3
1
1.4
1.2
1.6
1.4
1.7
1.5
1.8
1.6
1.9
1.7
2.2
New Kent
0
2.1
1.2
4.1
2.6
6.2
4
9.4
5.4
13
6.9
17
10
21
14
25
18
29
22
34
Petersburg
0
0
0
0
0
0
0
<0.01
0
0.01
<0.01
0.01
<0.01
0.01
0.01
0.02
0.01
0.02
0.01
0.03
Prince George
0
1.9
1.1
3.8
2.4
5.7
3.7
8.1
5
11
6.3
14
8.8
17
12
20
15
23
17
26
Williamsburg
0
0.05
0.03
0.1
0.06
0.1
0.1
0.2
0.1
0.3
0.2
0.3
0.2
0.4
0.3
0.4
0.3
0.5
0.4
0.6
Statewide
54
236
189
479
362
751
585
1029
816
1362
1060
1707
1368
2051
1708
2332
2028
2582
2283
2830
-------
[ 114 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
Table A.7 Low and High Estimates of the Area of Land Close to Sea Level in Virginia (continued)
Meters above Spring High Water
Jurisdiction
low
high
low
high
low
high
low
high
low
high
low
high
low
high
low
high
low
high
low
high
0.
5
1.
0
1.
5
2.
0
2.
5
3.
0
3.
5
4.
0
4.
5
5.
0
Tidal
Cumulative (total) amount of Nontidal Wetlands below a given elevation
Eastern Shore
946
7
22
20
48
39
76
63
101
87
114
107
126
119
137
131
146
141
153
149
161
Accomack
484
7
21
19
45
36
70
58
92
80
104
98
114
108
124
119
132
128
138
134
145
Northampton
462
0.4
1.2
1
3.4
2.5
5.9
4.7
8.1
7
9.7
8.8
11
10
13
12
14
14
15
15
16
Northern Virginia
17
0
1
0
2
1
3
2
3
2
4
3
4
3
5
4
5
4
6
5
6
Stafford
6.8
0
0.5
0.3
1
0.6
1.5
1
1.9
1.3
2.3
1.7
2.6
2
2.9
2.3
3.3
2.6
3.6
3
3.9
Alexandria
0.2
0
0.03
0.02
0.07
0.04
0.1
0.06
0.11
0.09
0.11
0.11
0.11
0.11
0.11
0.11
0.11
0.11
0.11
0.11
0.12
Fairfax
4.9
0
0.2
0.1
0.4
0.2
0.6
0.4
0.7
0.5
0.8
0.6
0.9
0.7
1.1
0.9
1.2
1
1.3
1.1
1.4
Prince William
5.1
0
0.2
0.1
0.3
0.2
0.5
0.3
0.6
0.4
0.6
0.5
0.7
0.6
0.8
0.7
0.8
0.7
0.9
0.8
0.9
Rappahannock Area
20
0
0.6
0.3
1.2
0.7
1.7
1.1
2.4
1.5
3
1.9
3.6
2.5
4.2
3.1
4.9
3.7
5.5
4.3
6.2
Fredericksburg
0
0
<0.01
<0.01
0.01
<0.01
0.01
<0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
King George
13
0
0.5
0.3
1
0.6
1.5
1
2
1.3
2.4
1.7
2.8
2.1
3.3
2.5
3.7
2.9
4.1
3.3
4.6
Spotsylvania
0.1
0
0.02
0.01
0.03
0.02
0.05
0.03
0.06
0.04
0.06
0.05
0.07
0.06
0.08
0.06
0.08
0.07
0.09
0.08
0.12
Caroline
6.3
0
0.1
0.03
0.1
0.1
0.2
0.1
0.3
0.2
0.5
0.2
0.7
0.3
0.9
0.5
1.1
0.7
1.3
0.9
1.5
Northern Neck
57
0
2.5
1.2
4.8
2.9
7.3
4.7
9.8
6.4
14
8.1
18
10
22
14
26
18
30
22
34
Westmoreland
14
0
0.5
0.3
1
0.6
1.5
1
2.2
1.3
3.9
1.7
5.6
2.5
7.2
4.1
8.9
5.7
10.6
7.3
12
Richmond
22
0
0.9
0.4
1.7
1
2.5
1.6
3.3
2.2
3.9
2.8
4.5
3.4
5.1
4
5.7
4.5
6.3
5.1
6.9
Northumberland
11
0
0.5
0.3
1.1
0.6
1.6
1
2.2
1.4
3.7
1.8
5.1
2.4
6.6
3.8
8
5.2
9.6
6.6
11
Lancaster
9.8
<0.01
0.5
0.3
1.1
0.7
1.6
1.1
2.1
1.4
2.5
1.8
2.8
2.2
3.2
2.5
3.5
2.8
3.8
3.2
4.2
Middle Peninsula
165
2.6
12
9.5
26
19
40
31
54
44
66
55
78
67
90
79
98
90
106
98
113
Essex
28
0
0.8
0.4
1.5
0.9
2.3
1.5
2.9
2
3.4
2.5
3.9
3
4.4
3.5
4.8
3.9
5.3
4.4
5.9
King and Queen
22
0
0.9
0.5
1.7
1.1
2.5
1.6
3.1
2.2
3.5
2.8
4
3.2
4.4
3.6
4.8
4
5.3
4.4
5.8
King Wlliam
36
0
0.4
0.2
0.7
0.5
1.1
0.7
1.4
0.9
1.7
1.2
2
1.5
2.3
1.8
2.6
2
2.9
2.3
3.3
Middlesex
9.7
<0.01
0.7
0.4
1.4
0.8
2.1
1.4
2.8
1.9
3.1
2.4
3.5
2.8
3.8
3.2
4.1
3.5
4.5
3.8
4.8
Gloucester
44
1.4
5.5
4.5
12
9.1
19
15
25
20
28
25
31
27
34
30
36
33
37
34
38
-------
[ SECTION
1.3 115 ]
Table A.7 Low and High Estimates of the Area of Land Close to Sea Level in Virginia (continued)
Meters above Spring High Water
Jurisdiction
low
high
low
high
low
high
low
high
low
high
low
high
low
high
low
high
low
high
low
high
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
Mathews
27
1.2
3.8
3.5
8.6
6.7
14
11
19
16
26
22
34
29
41
37
46
44
51
48
55
Hampton Roads
329
12
42
38
74
64
96
84
127
104
167
127
205
164
245
202
285
242
326
279
391
James City
33
<0.01
0.8
0.4
1.5
0.9
2.2
1.4
2.8
1.9
3.3
2.5
3.7
2.9
4.2
3.3
4.6
3.8
5.1
4.2
5.6
York
17
0.19
0.9
0.7
2.7
1.9
4.9
3.7
6.7
5.6
7.4
6.9
8
7.6
8.7
8.2
9.1
8.8
9.5
9.2
9.9
Newport News
15
0.1
0.3
0.3
0.7
0.5
1
0.9
1.3
1.2
1.4
1.35
1.42
1.4
1.5
1.4
1.5
1.5
1.6
1.6
1.7
Poquoson
24
0.02
0.1
0.1
0.4
0.3
0.8
0.6
1.1
0.9
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
Hampton
14
0.06
0.2
0.2
0.4
0.3
0.6
0.5
0.9
0.7
1.5
1.1
2.2
1.8
2.9
2.5
4
3.3
5.1
4.4
6.2
Surry
11
0
0.6
0.3
1.3
0.8
1.9
1.2
2.4
1.7
2.5
2.1
2.7
2.4
2.9
2.6
3
2.7
3.2
2.9
3.4
Isle of Wight
29
<0.01
0.3
0.2
0.6
0.4
0.9
0.6
1.4
0.8
2.2
1
3.1
1.5
4
2.4
4.8
3.2
5.7
4
7.3
Norfolk
4.7
0.1
0.3
0.2
0.5
0.4
0.8
0.7
1.1
0.9
1.3
1.1
1.5
1.3
1.7
1.5
1.7
1.7
1.7
1.7
1.7
Virginia Beach
112
4.2
14
13
25
22
33
29
41
37
46
43
50
48
53
51
56
54
57
56
59
Suffolk
26
0.03
0.2
0.1
0.3
0.2
0.4
0.3
0.8
0.4
1.3
0.5
1.8
1
2.3
1.4
3.1
2.1
6.8
2.9
33
Portsmouth
3.7
2.4
7.7
6.8
8.9
8.9
9.2
9.1
9.5
9.3
9.9
9.6
10
10
11
10
11
10.7
11
10.9
11
Chesapeake
40
4.5
17
15
32
28
40
36
58
44
89
56
120
86
152
116
186
149
217
180
251
Other Jurisdictions
85
0
5.5
3.2
11
6.9
16
10
20
14
22
18
24
20
26
22
28
24
30
26
33
Charles City
22
0
1.9
1.1
3.7
2.4
5.6
3.6
6.8
4.9
7.4
6.2
8
6.9
8.6
7.5
9.2
8.1
9.8
8.6
11
Chesterfield
11
0
0.4
0.2
0.7
0.4
1.1
0.7
1.2
0.9
1.2
1.1
1.2
1.17
1.24
1.2
1.3
1.2
1.3
1.2
1.3
Henrico
4.2
0
0.04
0.02
0.08
0.05
0.12
0.1
0.2
0.1
0.2
0.1
0.2
0.2
0.3
0.2
0.3
0.2
0.4
0.3
0.4
Hopewell
0.7
0
0.1
0.1
0.2
0.1
0.3
0.2
0.3
0.3
0.4
0.3
0.4
0.3
0.4
0.36
0.4
0.37
0.41
0.38
0.42
New Kent
34
0
2.3
1.3
4.5
2.9
6.8
4.4
8.1
6
8.7
7.6
9.3
8.2
9.8
8.8
10.4
9.3
11
9.9
12
Prince George
11
0
0.8
0.5
1.5
1
2.3
1.5
3.1
2
3.9
2.6
4.7
3.3
5.5
4
6.3
4.8
7.1
5.5
7.5
Williamsburg
0.4
0
0.02
0.01
0.03
0.02
0.05
0.03
0.06
0.04
0.07
0.05
0.08
0.06
0.1
0.07
0.11
0.09
0.12
0.1
0.14
Statewide
1619
21
86
72
167
134
240
197
317
260
389
320
459
387
529
455
594
523
657
583
745
Cumulative (total) amount of land below a given elevation
Dry Land
54
236
189
479
362
751
585
1029
816
1362
1060
1707
1368 2051
1708
2332
2028
2582
2283
2830
Nontidal Wetlands
21
86
72
167
134
240
197
317
260
389
320
459
387
529
455
594
523
657
583
745
All Land
1619
1694
1941
1881
2265
2115
2611
2401
2965
2694
3370
2999
3785
3374 4199
3782
4545
4170
4858
4486
5193
-------
744
65
16
149
336
496
188
1.9
267
326
155
1.5
0.03
130
0.02
0.06
55
707
41
2.8
38
90
8.0
166
325
460
149
432
65
0.9
380
556
6349
SOCIATED WITH EPA'S ESTIMATES ]
the Area of Land Close to Sea Level in North Carolina
Meters above Spring High Water
' high low high low high
2.0 2.5 3.0
-Cumulative (total) amount of dry land
109
4.7
0
24
26
127
6.5
0.02
20
50
71
<0.01
0
11
0
0
7.4
433
3.0
0
2.6
15
0.2
35
64
40
12
12
2.4
0
269
22
1368
156
6.8
<0.01
31
46
179
9.2
0.04
32
71
86
<0.01
0
16
0
0
11
482
4.0
0
5.6
20
0.3
43
95
65
17
18
3.7
0
321
38
1757
177
8.2
<0.01
36
59
220
11
0.05
37
87
91
<0.01
0
17
0
0
12
496
4.4
0
7.0
22
0.3
46
116
83
19
24
4.7
0
331
49
1957
235
10
0.01
43
100
287
15
0.1
54
119
102
<0.01
0
22
0
0
17
533
5.6
0
11
28
0.4
55
150
112
25
39
6.5
0
351
68
2388
257
12
0.02
48
115
326
17
0.1
60
143
106
<0.01
0
22
0
0
17
548
6.1
0
13
30
0.4
58
170
131
28
52
7.8
0
358
81
2609
317
15
0.06
55
147
379
22
0.2
78
178
117
0.01
<0.01
27
0
0
21
586
7.7
0.01
18
35
0.8
68
194
161
36
79
10
0
369
106
3030
341
17
0.1
60
157
402
27
0.3
85
201
121
0.01
<0.01
28
0
0
22
600
8.4
0.02
19
37
0.9
71
209
178
40
97
12
0
371
128
3232
401
20
0.2
68
189
421
35
0.4
104
234
131
0.02
<0.01
35
0
0
26
632
11
0.05
23
43
1.5
81
230
202
51
124
15
0.02
374
165
3615
422
22
0.2
74
201
427
42
0.5
111
252
133
0.03
<0.01
36
0
0
26
641
11
0.06
24
45
1.6
85
243
221
55
145
17
0.03
375
192
3803
below a
482
26
0.4
83
232
437
55
0.6
132
273
140
0.07
<0.01
50
<0.01
0
31
660
14
0.1
27
52
2.6
96
263
259
69
189
21
0.06
378
238
4208
given elevation
505 576
28
0.5
89
241
443
65
0.7
140
285
143
0.1
32
1.1
98
256
452
85
0.8
165
300
147
0.2
<0.01 <0.01
52 69
<0.01 <0.01
0
31
666
15
0.2
28
55
2.8
100
274
290
74
227
24
0.07
378
272
4429
0
36
682
19
0.3
30
61
3.8
111
289
350
89
296
30
0.1
379
340
4899
600
35
1.5
105
262
459
100
0.9
175
306
148
0.2
<0.01
72
<0.01
0
37
686
20
0.4
30
64
4.0
116
296
382
94
335
34
0.15
380
387
5131
-------
[ SECTION 1.3 117 ]
Table A.8 Low and High Estimates of the Area of Land Close to Sea Level in North Carolina (continued)
Meters above Spring High Water
County |ow high low high low high low high low high low high low high low high low high low high
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Wetlands
Tidal
—
Cumulative (total) amount of Nontidal Wetlands below
a given
elevation-
Beaufort
35
65
95
105
131
139
162
171
202
215
244
252
272
278
290
294
306
310
320
323
330
Bertie
0.3
110
123
127
132
136
142
147
153
159
167
171
177
181
186
191
200
207
219
225
234
Bladen
0
<0.01
0.1
0.2
0.6
0.9
1.8
2.1
3.3
4.1
6.3
7.3
10
11
15
16
21
23
29
31
36
Brunswick
109
38
44
47
52
55
58
61
65
67
71
73
77
79
82
85
88
90
93
95
98
Camden
7.1
137
146
149
155
157
165
168
175
177
184
187
194
197
201
203
210
214
233
243
258
Carteret
334
34
67
87
117
136
164
180
202
216
231
237
243
247
254
258
267
273
281
286
293
Chowan
0
29
32
34
37
38
40
42
44
46
49
51
56
59
64
70
79
84
91
96
104
Columbus
0
0.2
0.5
0.8
1.3
1.9
2.7
3.2
3.9
4.4
5.1
5.5
6.1
6.4
6.7
7
7.3
7.5
8.0
8.9
11
Craven
12
59
74
80
94
100
115
121
137
142
154
159
170
173
184
188
198
202
213
217
227
Currituck
125
129
144
150
159
164
172
178
184
188
194
196
199
201
203
204
206
209
215
219
221
Dare
168
376
525
553
604
619
651
659
664
664
665
666
666
666
666
666
666
666
666
666
666
Duplin
0
0
0
0
0
0
0
0
0.01
0.03
0.1
0.2
0.5
0.7
1.4
1.8
2.9
3.4
4.7
5.3
6.7
Edgecombe
0
0
0
0
0
0
0
0
0
0
0
0
<0.01
<0.01
<0.01
<0.01
0.01
0.01
0.03
0.05
0.09
Gates
0
78
89
89
93
94
98
99
102
103
107
108
114
115
121
122
126
126
129
129
132
-------
0.2
1.6
81
689
33
17
150
60
14
72
232
124
239
180
70
6.8
623
197
5405
6349
5405
13026
ASSOCIATED WITH EPA'S ESTIMATES ]
ie Area of Land Close to Sea Level in North Carolina (continued)
Meters above Spring High Water
low high low high low high low high low high low
0
0
54
488
11
0
73
36
2.0
31
73
62
113
47
27
0
523
86
3048
0
0
58
538
13
0.07
88
39
2.6
35
81
68
128
52
30
0
554
92
3354
0
0
58
549
14
0.13
93
40
2.7
36
86
71
132
55
32
0
559
96
3465
0
0
61
571
16
0.38
103
42
3.5
40
97
75
145
61
35
0
569
101
3694
0
0
62
578
16
0.5
106
43
3.7
41
106
79
150
66
36
0
0
65
592
18
1.1
114
45
5.9
45
123
84
161
74
39
0 <0.01
571 579
106 112
3794 3992
0
0
66
598
19
1.5
117
46
6.0
46
131
88
165
79
41
0.02
582
118
0
0
69
614
21
2.8
124
48
7.3
49
142
93
175
86
44
0.4
591
128
0 <0.01
0
69
619
21
3.3
126
49
7.6
51
148
96
179
90
46
0.6
593
134
4347
0
71
634
23
4.9
130
51
9.6
54
161
102
189
98
49
1.4
601
145
4509
1368 1757
3048 3354
5687 6384
1957 2388
3465 3694
6694 7354
4087 4269
Cumulative (total) amount of land below
2609 3030 3232 3615 3803 4208
3794 3992
7676 8293
4087 4269
8591 9157
4347 4509
9422 9989
<0.01
0
71
638
24
5.6
132
52
9.9
55
171
106
192
103
51
1.6
606
152
4583
a given
4429
0.01
0
74
653
26
7.6
136
53
11
59
186
113
202
113
54
2.3
614
162
4741
elevation
4899 5131
0.01
0
74
660
26
8.4
137
54
11
60
192
116
206
124
57
2.6
616
168
4818
4583 4741
10284 10912
0.02
<0.01
77
672
28
11
140
56
12
64
201
119
216
137
60
3.6
620
175
4969
5529
4818 4969
11221 11770
-------
Appendix B
Low and High Estimates for the Area of Dry and Wet Land Close to Sea Level,
by Subregion3 (square kilometers)
a The low and high estimates are based on the on the contour interval and/or stated root mean square error (RMSE) of the data used to calculate
elevations and an assumed standard error of 30 cm in the estimation of spring high water. For details, see main text of this Section 1.3.
-------
[ 120 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
Table B.1 Low and High Estimates for the Area of Dry and Wet Land Close to Sea Level - Long Island Sound, New York
Elevations above spring high water
Locality
50 cm
1 meter
2 meters
3 meters
5 meters
Low
High
Low
High
Low
High
Low
High
Low
High
Cumulative
total) amount of dry land below a given elevation
Westchester
0.2
1.5
1.1
3.0
2.8
5.8
5.1
8.6
10.0
12.4
Bronx
0.4
2.6
1.8
5.1
4.8
9.8
8.7
14.6
16.9
19.6
Queens
6.2
17.0
14.6
28.1
31.7
48.6
50.7
66.6
76.5
80.8
Brooklyn
3.1
9.1
8.0
15.6
18.8
30.5
34.0
47.4
58.9
62.8
Nassau
2.2
19.2
12.9
44.5
50.9
85.4
85.4
104.1
119.3
132.1
Suffolk
13.7
51.5
43.1
96.8
114.9
181.3
188.6
251.3
318.8
371.4
Total
25.8
100.9
81.4
193.1
223.9
361.4
372.4
492.6
600.4
679.1
Tidal
Cumulative (total) amount of wetlands below a given elevation
Westchester
1.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.1
0.1
Bronx
1.2
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.1
0.1
Queens
11.9
0.0
0.2
0.1
0.3
0.4
0.5
0.5
0.6
0.7
0.7
Brooklyn
10.1
0.0
0.1
0.1
0.1
0.1
0.1
0.1
0.2
0.2
0.2
Nassau
43.7
0.1
0.4
0.3
0.7
0.8
1.5
1.4
2.1
2.6
3.2
Suffolk
72.1
1.5
5.7
4.9
9.8
10.8
15.2
15.1
18.3
20.8
23.8
Total
140.0
1.7
6.4
5.4
11.0
12.1
17.4
17.2
21.3
24.3
28.1
Dry and nontidal wetland
27
107
87
204
236
379
390
514
625
707
All land
140
167
247
227
344
376
519
530
654
765
847
-------
[ SECTION 1.3 121 ]
Table B.2 Low and High Estimates for the Area of Dry and Wet Land Close to Sea Level in New York Harbor
Elevations above spring high water
50 cm
1 meter
2 meters
3 meters
5 meters
Low
High
Low
High
Low
High
Low
High
Low
High
Locality
State
Cumulative
total) amount of dry land below a given elevation
Monmouth
NJ
2.0
5.4
5.9
10.5
15.8
18.7
22.4
24.7
31.2
32.5
Middlesex
NJ
0.4
8.8
4.3
17.4
14.7
31.2
25.4
43.5
45.6
62.0
Somerset
NJ
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.2
Union
NJ
0.4
6.9
4.2
13.7
12.6
22.7
20.2
29.3
31.7
40.9
Hudson
NJ
0.6
16.2
10.4
32.2
30.6
49.0
46.4
56.9
60.4
67.5
Essex
NJ
0.4
6.1
3.9
12.0
11.3
19.6
17.8
25.3
27.8
32.2
Bergen
NJ
0.9
15.6
10.2
31.0
29.4
44.2
42.5
49.0
51.1
58.2
Passaic
NJ
0.0
0.2
0.1
0.3
0.3
0.7
0.6
1.1
1.3
1.9
Ellis Island
NJ
0.0
0.0
0.0
0.1
0.1
0.1
0.1
0.1
0.1
0.1
Staten Island
NY
0.3
7.8
5.1
15.5
14.9
24.9
23.3
30.8
33.9
39.0
Brooklyn
NY
0.0
0.8
0.5
1.6
1.6
3.1
2.7
4.5
5.3
6.4
Manhattan
NY
0.0
2.2
1.4
4.3
4.2
8.3
7.2
12.1
14.1
17.5
Bronx
NY
0.0
0.6
0.4
1.2
1.2
2.7
2.2
4.4
5.3
6.9
Westchester
NY
0.0
1.3
0.7
2.6
2.3
4.7
4.1
6.1
6.4
8.3
Total
5.1
71.9
47.1
142.6
138.9
230.0
214.9
288.0
314.1
373.7
Tidal
Cumulative (total) amount of wetlands below a given elevation
Monmouth
NJ
7.7
0.1
0.3
0.4
0.6
0.8
0.9
1.1
1.2
1.7
1.8
Middlesex
NJ
21.7
0.1
1.2
0.7
2.3
2.1
3.9
3.5
5.3
5.7
7.8
Union
NJ
2.3
0.0
0.2
0.1
0.3
0.3
0.5
0.4
0.6
0.6
0.8
Hudson
NJ
12.0
0.0
0.2
0.1
0.3
0.3
0.4
0.4
0.5
0.5
0.5
Essex
NJ
0.3
0.0
0.0
0.0
0.1
0.1
0.1
0.1
0.1
0.1
0.1
Bergen
NJ
15.0
0.0
0.6
0.4
1.2
1.1
1.5
1.5
1.5
1.6
2.1
Passaic
NJ
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.1
Staten Island
NY
4.0
0.0
0.5
0.3
0.9
0.9
1.4
1.3
1.6
1.7
1.9
Bronx
NY
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.1
0.1
0.1
Westchester
NY
0.7
0.0
0.0
0.0
0.1
0.1
0.1
0.1
0.1
0.1
0.1
Rockland
NY
2.3
0.0
0.0
0.0
0.0
0.0
0.1
0.1
0.1
0.1
0.2
Orange
NY
0.2
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Putnam
NY
1.3
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Dutchess
NY
0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Total
67.6
0.2
3.0
2.0
5.8
5.6
9.0
8.6
11.1
12.2
15.5
Dry and nontidal wetland
5
75
49
148
145
239
223
299
326
389
| All land
68
73
142
117
216
212
307
291
367
394
457
-------
[ 122 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
Table B.3 Low and High Estimates for the Area of Dry and Wet Land Close to Sea Level in New Jersey Shore
Elevations above spring high water:
County
50 cm
1 meter
2 meters
3 meters
5 meters
Low
High
Low
High
Low
High
Low
High
Low
High
Cumulative (total) amount of Dry Land below a given elevation
Cape May
7.6
21.8
23.8
42.0
56.1
73.5
78.4
102.2
124.2
144.1
Atlantic
4.0
13.5
14.0
29.0
40.8
53.9
57.3
71.0
88.5
105.8
Burlington
0.0
2.1
1.3
4.1
4.0
8.9
7.0
15.1
18.4
27.1
Ocean
4.6
18.7
21.8
44.0
67.3
80.6
93.2
106.8
136.6
149.1
Monmouth
2.1
4.9
5.5
9.4
15.3
19.9
26.4
31.8
50.4
54.9
Total
18.3
61.1
66.5
128.5
183.5
236.9
262.3
326.9
418.1
481.0
Tidal
Cumulative (total) amount of wetlands below a given elevation
Cape May
153.2
2.9
12.0
10.2
20.4
22.2
33.1
32.2
42.7
47.6
55.2
Atlantic
204.0
4.8
17.9
14.7
29.2
31.9
50.1
48.3
68.2
82.0
102.9
Burlington
37.3
0.2
9.7
6.2
19.1
18.7
32.7
30.0
41.3
45.8
57.2
Ocean
124.8
2.3
11.6
10.0
21.7
25.8
38.3
39.0
49.4
56.5
65.8
Monmouth
4.4
0.5
0.9
1.0
1.4
1.9
2.3
2.9
3.2
4.8
5.1
Total
523.6
10.7
52.1
42.1
91.9
100.5
156.5
152.4
204.9
236.5
286.3
Dry and nontidal wetland
29
113
109
220
284
393
415
532
655
767
All land
524
553
637
632
744
808
917
938
1055
1178
1291
-------
[ SECTION 1.3 123 ]
Table B.4 Low and High Estimates for the Area of Dry and Wet Land Close to Sea Level in Delaware Estuary
Elevations above spring high water:
50 cm
1 meter
2 meters
3 meters
5 meters
Low
High
Low
High
Low
High
Low
High
Low
High
Locality
State
Cumulative (total) amount of dry land below a given elevation
Sussex
DE
6.4
18.2
15.8
30.8
37.3
55.2
60.0
78.6
103.3
119.7
Kent
DE
8.8
24.8
21.9
40.6
47.9
77.6
86.1
119.2
177.8
209.9
New Castle
DE
7.1
19.0
16.8
29.9
34.4
52.2
54.2
75.0
99.0
119.0
Delaware
PA
0.4
6.1
4.0
12.1
11.5
18.0
17.2
20.7
22.2
25.9
Philadelphia3
PA
3.6
6.1
6.8
12.4
20.0
24.8
31.6
36.8
51.5
54.8
Bucks
PA
0.0
4.4
0.2
8.5
5.3
18.0
11.9
27.4
25.3
42.1
Mercer
NJ
0.0
0.1
0.0
0.1
0.1
0.2
0.2
0.4
0.3
0.4
Burlington
NJ
0.1
4.3
0.4
8.4
5.3
16.4
11.0
24.5
22.5
42.2
Camden
NJ
0.0
3.8
0.1
7.3
4.3
14.8
9.5
22.4
20.4
34.5
Gloucester
NJ
0.2
9.2
6.1
18.4
17.7
33.3
29.6
46.5
53.5
69.3
Salem
NJ
5.9
26.9
21.3
48.7
53.8
84.4
83.9
114.0
135.5
160.3
Cumberland
NJ
3.0
15.8
12.1
28.9
30.3
53.2
49.5
76.9
90.8
114.3
Cape May
NJ
0.4
3.5
2.5
7.5
8.6
19.9
20.9
36.9
55.5
68.0
Total
35.9
142.0
108.0
253.7
276.5
468.0
465.7
679.2
857.7
1060.4
Tidal
Cumulative (total) amount of wetlands below a given elevation
Sussex
DE
67.4
2.1
4.8
4.6
6.2
6.8
8.6
9.0
10.6
12.3
13.3
Kent
DE
168.7
4.9
11.4
10.4
16.6
19.0
24.6
25.9
30.9
38.8
43.5
New Castle
DE
73.5
1.8
3.8
3.5
4.8
5.1
6.7
6.7
8.4
9.7
11.1
Delaware
PA
3.6
0.1
0.8
0.6
1.7
1.6
2.2
2.2
2.3
2.3
2.3
Philadelphia
PA
0.6
0.5
0.6
0.6
0.9
1.2
1.4
1.6
1.7
1.9
1.9
Bucks
PA
1.9
0.0
0.9
0.1
1.9
1.2
4.1
2.9
6.3
6.2
8.2
Mercer
NJ
1.8
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Burlington
NJ
5.4
0.0
0.6
0.0
1.2
0.7
2.3
1.5
3.4
3.1
5.8
Camden
NJ
1.5
0.0
0.3
0.1
0.7
0.5
1.3
0.9
1.9
1.8
2.7
Gloucester
NJ
18.0
0.2
8.8
5.9
17.4
16.8
25.9
25.0
28.8
30.4
33.5
Salem
NJ
110.1
9.6
25.1
22.3
35.8
38.2
49.0
48.9
55.4
60.3
67.6
Cumberland
NJ
212.6
4.7
23.6
18.1
42.1
43.6
65.5
63.5
80.6
89.8
103.2
Cape May
NJ
48.3
4.3
14.7
12.2
25.1
28.2
40.3
41.5
51.2
58.6
63.7
Total
713.5
28.3
95.5
78.5
154.2
163.0
231.8
229.7
281.6
315.1
356.8
Dry and nontidal wetland
64
237
187
408
440
700
695
961
1173
1417
All land
713
778
951
900
1121
1153
1413
1409
1674
1886
2131
a This number includes Philadelphia's 2.4 square kilometers of dry land below spring high water, of which 0.87, 0.26, 0.054, and 0.005 are at least 0.5, 1, 2, and 3
meters below spring high water, respectively. Most of this land is near Philadelphia International Airport.
-------
[ 124 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
Table B.5 Low and High Estimates for the Area of Dry and Wet Land Close to Sea Level in DelMarVa Atlantic Coast
Elevations above spring high water:
50 cm
1 meter
2 meters
3 meters
5 meters
Low
High
Low
High
Low
High
Low
High
Low
High
Locality
State
Cumulative (total) amount of Dry Land below a given elevation
Northampton
VA
5.1
14.5
13.0
16.8
17.9
20.6
21.4
24.6
30.5
35.0
Accomack
VA
7.5
22.6
20.1
37.7
44.5
61.7
65.8
81.2
103.7
118.9
Worcester
MD
3.7
18.6
21.7
42.4
77.5
102.8
134.0
154.6
219.1
234.6
Sussex
DE
11.1
32.4
27.6
53.5
64.5
94.9
104.2
139.5
196.5
234.2
Total
27.4
88.1
82.5
150.3
204.4
280.0
325.4
399.9
549.9
622.7
Tidal
Cumulative (total) amount of wetlands below a given elevation
Northampton
VA
436.4
0.3
0.8
0.7
2.1
2.8
4.4
4.6
5.2
5.8
6.1
Accomack
VA
327.3
1.3
4.1
3.5
10.4
13.5
20.7
21.9
26.2
31.2
33.7
Worcester
MD
118.5
0.4
4.3
5.0
8.8
14.1
18.1
23.4
27.0
36.0
37.6
Sussex
DE
41.0
1.7
4.9
4.2
7.5
8.8
12.2
12.9
15.7
18.9
20.7
Total
923.3a
3.7
14.1
13.4
28.7
39.2
55.4
62.7
74.1
91.9
98.1
Dry and Nontidal wetland
31
102
96
179
244
335
388
474
642
721
All Land
923
954
1025
1019
1102
1167
1259
1311
1397
1565
1644
a Includes 375 square kilometers of tidal mudflats in Northampton and Accomack counties.
-------
[ SECTION 1.3 125 ]
Table B.6 Low and High Estimates for the Area of Dry and Wet Land Close to Sea Level in Hampton Roads, Virginia
Elevations above spring high water
Locality
50 cm
1 meter
2 meters
3 meters
5 meters
Low
High
Low
High
Low
High
Low
High
Low
High
Cumulative (total) amount of Dry Land below a given elevation
Virginia Beach
9.3
33.0
30.3
68.7
93.6
163.2
184.7
272.9
378.1
418.2
Chesapeake
3.5
11.9
10.8
30.6
44.6
86.6
100.4
204.5
353.0
429.7
Norfolk
1.9
5.8
5.2
17.1
24.0
42.4
52.4
91.2
121.7
128.2
Portsmouth
1.2
3.9
3.5
9.6
12.8
22.0
26.7
45.0
62.6
69.9
Suffolk
0.7
4.3
3.1
7.1
7.5
15.2
13.0
31.0
47.3
73.3
Isle of Wight
0.2
3.4
2.1
6.2
6.0
12.8
10.1
21.6
26.8
42.0
Surry
0.0
1.4
0.7
2.7
2.7
5.3
4.6
7.1
8.1
11.2
James City
0.1
3.8
2.2
7.2
7.0
14.2
11.8
22.1
26.7
38.7
York
1.4
6.0
4.8
13.1
16.3
27.7
28.3
37.3
44.3
51.3
Newport News
2.2
6.9
6.1
11.0
12.9
17.9
19.3
24.8
34.9
42.3
Poquoson
1.4
4.5
4.1
8.8
10.9
16.3
16.4
16.6
16.7
16.7
Hampton
1.9
5.9
5.3
18.1
25.4
45.3
51.2
73.8
94.7
102.4
Total
23.8
90.8
78.2
200.2
263.6
468.9
519.0
847.9
1214.9
1423.8
Tidal
Cumulative (total) amount of wetlands below a given elevation
Virginia Beach
111.9
4.2
14.5
13.3
24.9
29.1
40.9
43.5
49.6
56.5
59.3
Chesapeake
39.7
4.5
16.6
15.4
32.1
36.4
58.3
55.7
120.2
180.3
250.8
Norfolk
4.7
0.1
0.3
0.2
0.5
0.7
1.1
1.1
1.5
1.7
1.7
Portsmouth
3.7
2.4
7.7
6.8
8.9
9.1
9.5
9.6
10.3
10.9
11.2
Suffolk
26.4
0.0
0.2
0.1
0.3
0.3
0.8
0.5
1.8
2.9
33.1
Isle of Wight
28.6
0.0
0.3
0.2
0.6
0.6
1.4
1.0
3.1
4.0
7.3
Surry
11.5
0.0
0.6
0.3
1.3
1.2
2.4
2.1
2.7
2.9
3.4
James City
32.8
0.0
0.8
0.4
1.5
1.4
2.8
2.5
3.7
4.2
5.6
York
17.0
0.2
0.9
0.7
2.7
3.7
6.7
6.9
8.0
9.2
9.9
Newport News
15.1
0.1
0.3
0.3
0.7
0.9
1.3
1.4
1.4
1.6
1.7
Poquoson
23.7
0.0
0.1
0.1
0.4
0.6
1.1
1.1
1.1
1.1
1.1
Hampton
14.3
0.1
0.2
0.2
0.4
0.5
0.9
1.1
2.2
4.4
6.2
Total
329.4
11.7
42.4
38.0
74.2
84.5
127.1
126.5
205.4
279.5
391.1
Dry and Nontidal wetland
35
133
116
274
348
596
645
1053
1494
1815
All Land
329
365
463
446
604
677
925
975
1383
1824
2144
-------
[ 126 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
Table B.7 Low and High Estimates for the Area of Dry and Wet Land Close to Sea Level in Middle Peninsula and Northern Neck Areas, Virginia
Elevations above spring high water
Locality
50 cm
1 meter
2 meters
3 meters
5 meters
Low
High
Low
High
Low
High
Low
High
Low
High
Cumulative (total) amount of Dry Land below a given elevation
Gloucester
4.1
16.0
13.2
32.9
40.5
66.9
66.9
84.2
96.4
110.8
Mathews
4.7
14.8
13.4
33.1
43.9
73.1
78.6
96.8
114.7
120.7
Middlesex
0.2
3.4
2.0
6.8
7.3
14.4
13.1
22.8
28.1
38.9
King William
0.0
1.6
0.9
3.2
3.1
8.4
5.4
17.7
22.7
36.1
King and Queen
0.0
2.9
1.7
5.7
5.5
11.9
9.6
19.0
22.7
32.9
Essex
0.0
3.8
2.0
7.3
7.1
15.5
12.3
27.9
34.2
52.8
Lancaster
0.1
7.0
3.6
13.8
13.8
28.0
24.0
41.5
48.4
67.9
Northumberland
0.0
5.9
2.8
11.5
11.0
24.1
19.2
63.8
84.5
140.9
Richmond
0.0
4.6
2.4
8.9
8.7
18.5
15.0
31.6
38.2
56.5
Caroline
0.0
0.4
0.3
0.9
0.9
1.8
1.5
2.8
3.4
5.2
Spotsylvania
0.0
0.1
0.1
0.2
0.2
0.3
0.3
0.5
0.5
0.8
Fredericksburg
0.0
0.1
0.0
0.1
0.1
0.2
0.2
0.3
0.4
0.5
Total
9.2
60.5
42.4
124.2
142.1
263.2
246.0
409.0
494.2
664.0
Tidal
Cumulative (total) amount of wetlands below a given elevation
Gloucester
43.5
1.4
5.5
4.5
11.9
14.7
24.8
24.6
30.8
34.4
38.5
Mathews
27.0
1.2
3.8
3.5
8.6
11.4
19.0
21.6
33.6
48.1
55.1
Middlesex
9.7
0.0
0.7
0.4
1.4
1.4
2.8
2.4
3.5
3.8
4.8
King Wlliam
35.6
0.0
0.4
0.2
0.7
0.7
1.4
1.2
2.0
2.3
3.3
King and Queen
21.6
0.0
0.9
0.5
1.7
1.6
3.1
2.8
4.0
4.4
5.8
Essex
27.5
0.0
0.8
0.4
1.5
1.5
2.9
2.5
3.9
4.4
5.9
Lancaster
9.8
0.0
0.5
0.3
1.1
1.1
2.1
1.8
2.8
3.2
4.2
Northumberland
11.4
0.0
0.5
0.3
1.1
1.0
2.2
1.8
5.1
6.6
10.8
Richmond
21.7
0.0
0.9
0.4
1.7
1.6
3.3
2.8
4.5
5.1
6.9
Caroline
6.3
0.0
0.1
0.0
0.1
0.1
0.3
0.2
0.7
0.9
1.5
Spotsylvania
0.1
0.0
0.0
0.0
0.0
0.0
0.1
0.1
0.1
0.1
0.1
Fredericksburg
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Total
214.3
2.6
14.1
10.5
29.7
35.1
62.0
61.7
90.9
113.5
136.9
Dry and Nontidal wetland
12
75
53
154
177
325
308
500
608
801
All Land
214
226
289
267
368
392
539
522
714
822
1015
-------
[ SECTION 1.3 127 ]
Table B.8 Low and High Estimates for the Area of Dry and Wet Land Close to Sea Level in Potomac River
Elevations above spring high water
50 cm
1 meter
2 meters
3 meters
5 meters
Low
High
Low
High
Low
High
Low
High
Low
High
Locality
State
Cumulative (total) amount of Dry Land below a given elevation
Westmoreland
VA
0.0
4.7
2.4
9.3
9.0
21.2
15.5
53.0
69.2
112.3
King George
VA
0.0
2.7
1.5
5.4
5.2
11.4
9.0
21.9
27.3
42.8
Stafford
VA
0.0
1.4
0.8
2.7
2.7
5.4
4.6
8.1
9.5
13.5
Prince William
VA
0.0
1.0
0.5
2.0
1.9
3.9
3.3
5.5
6.4
8.8
Fairfax
VA
0.0
2.0
1.1
3.9
3.8
7.6
6.6
10.7
12.4
18.1
Alexandria
VA
0.0
0.4
0.3
0.9
0.9
1.7
1.5
2.5
2.9
4.0
Arlington
VA
0.0
0.2
0.1
0.5
0.5
1.3
0.8
2.6
3.4
5.0
DC
1.6
3.0
2.8
4.4
5.5
7.4
8.9
11.1
15.9
17.7
Prince George's
MD
0.1
1.1
0.5
2.2
1.6
4.0
3.2
5.4
6.6
9.9
Charles
MD
0.7
10.9
4.6
19.4
14.1
38.4
28.3
64.0
74.2
96.0
St. Mary's
MD
1.6
12.0
5.6
19.8
14.9
39.2
27.9
70.1
81.2
99.8
Total
4.1
39.5
20.1
70.4
60.0
141.5
109.5
255.1
308.9
428.1
Tidal
Cumulative (total) amount of wetlands below a given elevation
Westmoreland
VA
14.4
0.0
0.5
0.3
1.0
1.0
2.2
1.7
5.6
7.3
12.0
King George
VA
13.5
0.0
0.5
0.3
1.0
1.0
2.0
1.7
2.8
3.3
4.6
Stafford
VA
6.8
0.0
0.5
0.3
1.0
1.0
1.9
1.7
2.6
3.0
3.9
Prince Wlliam
VA
5.1
0.0
0.2
0.1
0.3
0.3
0.6
0.5
0.7
0.8
0.9
Fairfax
VA
4.9
0.0
0.2
0.1
0.4
0.4
0.7
0.6
0.9
1.1
1.4
Alexandria
VA
0.2
0.0
0.0
0.0
0.1
0.1
0.1
0.1
0.1
0.1
0.1
Arlington
VA
0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
DC
0.5
0.0
0.0
0.0
0.1
0.1
0.1
0.1
0.2
0.3
0.3
Prince George's
MD
1.6
0.0
0.3
0.1
0.5
0.4
0.8
0.7
0.9
1.2
2.1
Charles
MD
22.9
0.1
3.6
1.4
6.2
4.6
11.3
9.0
15.9
17.8
22.2
St. Mary's
MD
11.7
0.3
1.8
0.8
3.3
2.4
7.1
4.9
12.9
15.4
22.5
Total
81.5
0.5
7.6
3.5
13.9
11.1
26.8
21.0
42.7
50.1
70.1
Dry and Nontidal wetland
5
47
24
84
71
168
130
298
359
498
All Land
82
86
129
105
166
153
250
212
379
441
580
-------
[ 128 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
Table B.9 Low and High Estimates for the Area of Dry and Wet Land Close to Sea Level - Maryland Western Shore
Elevations above spring high water
50 cm
1 meter
2 meters
3 meters
5 meters
Low
High
Low
High
Low
High
Low
High
Low
High
Locality
Cumulative (total) amount of Dry Land below a given elevation
Prince George's
0.0
1.1
0.4
1.7
1.3
3.2
2.3
5.3
6.5
10.8
Charles
0.0
0.7
0.3
1.2
0.9
2.0
1.7
2.5
2.7
3.3
St. Mary's
0.8
3.8
2.5
8.0
8.8
18.8
18.2
30.6
38.5
48.4
Calvert
0.4
3.9
1.7
5.8
4.6
10.1
7.6
17.3
21.2
35.7
Anne Arundel
1.7
7.2
6.7
14.6
20.2
38.7
43.5
59.1
80.5
94.3
Howard
0.0
0.0
0.0
0.0
0.0
0.1
0.1
0.1
0.2
0.3
Baltimore City
0.2
2.1
0.9
3.9
2.7
7.5
5.7
11.9
14.1
21.0
Baltimore
2.3
6.6
7.3
13.0
20.8
27.0
37.0
45.8
74.5
80.7
Harford
0.7
17.3
7.6
25.1
21.7
40.3
34.2
57.1
65.5
78.2
Total
6.1
42.7
27.5
73.4
81.1
147.8
150.3
229.7
303.7
372.7
Tidal
Cumulative (total) amount of wetlands below a given elevation
Prince George's
12.3
0.0
0.5
0.2
0.9
0.7
1.8
1.3
2.9
3.5
5.1
Charles
1.3
0.0
0.2
0.1
0.2
0.2
0.4
0.3
0.4
0.5
0.6
St. Mary's
7.0
0.3
1.0
0.8
2.0
2.2
3.9
3.9
5.9
7.5
8.8
Calvert
14.6
0.1
0.9
0.4
1.3
1.1
2.2
1.7
3.8
4.7
7.5
Anne Arundel
12.1
0.2
0.7
0.6
1.6
3.1
8.1
9.5
12.4
15.3
17.1
Howard
0.0
0.0
0.0
0.0
0.0
0.0
0.1
0.0
0.1
0.1
0.1
Baltimore City
0.2
0.0
0.0
0.0
0.0
0.0
0.1
0.0
0.1
0.1
0.1
Baltimore
10.5
0.1
0.3
0.3
0.7
1.0
1.3
1.5
1.7
2.2
2.3
Harford
29.4
0.2
2.5
1.2
3.8
3.3
6.2
5.2
9.0
10.2
12.0
Total
87.3
0.8
6.2
3.7
10.5
11.6
24.0
23.5
36.4
43.9
53.6
Dry and Nontidal wetland
7
49
31
84
93
172
174
266
348
426
All Land
87
94
136
119
171
180
259
261
353
435
514
-------
[ SECTION 1.3 129 ]
Table B.10 Low and High Estimates for the Area of Dry and Wet Land Close to Sea Level - Chesapeake Bay Eastern Shore
Elevations above spring high water
50 cm
1 meter
2 meters
3 meters
5 meters
Low
High
Low
High
Low
High
Low
High
Low
High
Locality
State
Cumulative (total) amount of Dry Land below a given elevation
Cecil
MD
0.2
2.5
1.0
5.2
3.7
11.6
7.8
20.0
24.3
37.9
Kent
MD
0.2
8.4
4.8
15.9
16.3
32.9
28.8
56.1
71.4
105.2
Queen Anne's
MD
0.6
4.1
5.3
11.9
24.2
35.0
51.6
68.2
125.2
142.6
Caroline
MD
0.7
3.2
2.2
6.1
6.9
12.5
13.2
19.7
25.9
32.9
Talbot
MD
2.2
7.8
11.1
23.7
64.0
98.7
148.7
175.1
265.6
279.4
Sussex
DE
0.5
1.6
1.4
3.3
4.3
7.1
8.5
13.8
26.0
36.3
Dorchester
MD
30.1
120.0
150.4
214.9
281.9
312.9
358.4
386.2
461.6
474.0
Wicomico
MD
5.0
14.9
18.3
28.6
47.1
58.5
76.0
86.2
133.2
141.6
Somerset
MD
17.1
58.4
70.5
100.7
167.8
193.4
215.1
232.5
326.5
344.6
Worcester
MD
0.7
2.7
3.1
5.8
10.6
16.5
23.6
28.4
46.1
53.4
Accomack
VA
5.8
18.4
16.8
40.4
53.3
87.5
94.2
110.4
129.5
138.1
Northampton
VA
2.3
7.2
6.5
15.8
20.8
34.5
39.9
62.8
98.7
123.7
Total
65.3
249.1
291.4
472.4
701.0
901.2
1065.8
1259.5
1734.0
1909.7
Tidal
Cumulative (total) amount of wetlands below a given elevation
Cecil
MD
12.6
0.0
0.2
0.0
0.7
0.4
1.7
1.2
2.8
3.5
5.5
Kent
MD
18.3
0.1
1.1
0.9
2.6
3.3
5.4
5.2
7.9
9.7
14.4
Queen Anne's
MD
21.4
0.2
1.1
1.5
3.0
4.9
6.5
7.9
9.6
14.6
17.9
Caroline
MD
14.4
0.3
1.4
0.7
2.6
2.5
5.3
4.4
7.5
8.0
11.7
Talbot
MD
26.1
0.1
0.3
0.5
1.0
2.5
4.2
6.8
8.5
17.9
19.6
Sussex
DE
6.7
0.6
1.8
1.6
2.7
3.1
4.4
4.8
6.4
10.1
13.1
Dorchester
MD
424.8
14.9
45.8
53.4
70.1
94.4
104.0
113.8
120.6
140.1
142.5
Wcomico
MD
67.0
5.4
9.9
10.7
13.5
24.2
29.2
37.0
44.4
67.0
70.2
Somerset
MD
265.4
6.6
15.7
17.3
21.3
34.8
39.8
45.1
51.5
80.6
90.1
Worcester
MD
23.7
0.3
0.9
1.0
1.6
2.7
4.0
6.3
8.8
18.2
20.8
Accomack
VA
156.4
5.3
16.7
15.3
34.6
44.8
71.8
76.5
88.2
103.2
111.1
Northampton
VA
25.5
0.1
0.4
0.4
1.2
1.9
3.7
4.2
6.2
8.8
10.1
Total
1062.4
33.8
95.3
103.3
155.0
219.5
279.9
313.0
362.4
481.7
526.9
Dry and Nontidal wetland
99
344
395
627
921
1181
1379
1622
2216
2437
All Land
1062
1162
1407
1457
1690
1983
2244
2441
2684
3278
3499
-------
Appendix C
Low and High Estimates of the Area of Land Close to Sea Level,
by Region: Mid-Atlantica (square kilometers)
a The low and high estimates are based on the on the contour interval and/or stated root mean square error (RMSE) of the data used to calculate
elevations and an assumed standard error of 30 cm in the estimation of spring high water. For details, see main text of this Section 1.3.
-------
[ SECTION 1.3 131 ]
Table C.1 Low and High Estimates of the Area of Land Close to Sea Level by Region
Meters above Spring High Water
Jurisdiction low high low high low high low high low high low high low high low high low high low high
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
¦Cumulative (total) amount of Dry Land below a given elevation
L.I. Sound and Peconic
6
31
22
59
42
86
63
111
85
135
106
158
127
181
149
200
170
216
190
229
South Shore Long Island
19
70
59
134
108
198
161
250
216
293
266
335
309
369
347
400
380
429
410
450
NY Harbor/
Raritan Bay Total
5
72
47
143
93
200
139
230
185
260
215
288
240
316
265
343
290
360
314
374
New York
0
13
8 25
16
37
24
44
32
51
40
58
46
65
52
72
59
76
65
78
New Jersey
5
59
39
117
77
163
115
186
153
209
175
230
194
251
213
271
231
284
249
295
New Jersey Shore
18
61
66
129
131
186
184
237
223
283
262
327
304
369
344
409
382
445
418
481
Delaware Bay Total
19
62
52
108
88
154
124
206
166
259
217
312
268
366
321
421
374
470
427
512
New Jersey
3
19
15
36
27
53
39
73
52
94
70
114
90
134
109
154
127
170
146
182
Delaware
15
43
38
71
61
101
85
133
114
165
146
198
178
232
212
267
247
300
281
330
Delaware River Total
17
80
56
146
103
210
152
262
201
315
249
368
296
417
342
467
386
512
430
549
Delaware: fresh
2
6
5
10
8 14
11
19
15
24
19
28
24
32
28
36
32
39
35
42
Delaware: saline
5
13
12
20
17
27
23
33
29
40
35
47
41
54
49
62
56
70
64
77
New Jersey: fresh
0
18
7
35
17
52
28
67
39
83
52
98
65
114
77
130
90
144
102
154
New Jersey: saline
6
27
21
48
37
68
53
82
68
96
82
109
95
121
108
133
119
143
130
152
Pennsylvania
4
17
11
33
24
49
37
61
50
73
61
85
71
96
81
106
90
115
99
123
Atlantic Coast of
Del-Mar-Va Total
27
87
81
148
140
212
200
275
259
334
318
390
373
443
425
495
477
548
529
599
Delaware
11
32
28
53
46
74
64
95
82
117
104
139
126
163
149
187
172
210
196
234
Maryland
3
17
20
40
44
69
74
97
101
123
126
145
148
163
165
180
182
196
199
211
Virginia
13
37
33
55
49
69
62
82
75
94
87
106
99
117
111
129
122
141
134
154
Chesapeake Bay Total
102
466
441
906
791
1357
1193
1827
1587
2334
1973
2859
2448
3378
2962
3818
3446
4234
3865
4633
Delaware
1
2
1
3
3
5
4
7
6
10
9
14
12
18
15
24
20
29
26
36
Maryland
66
290
306
530
515
763
738
1007
952
1227
1141
1451
1352
1670
1572
1865
1778
2047
1966
2213
fresh
9
35
33
70
63
115
106
167
152
212
192
263
243
325
307
394
377
466
449
533
vulnerable
49
187
234
344
379
477
515
605
633
704
731
804
830
892
911
958
974
1011
1024
1058
saline
00
CD
00
39
117
74
171
118
235
167
311
218
385
280
454
354
513
427
570
492
623
-------
[ 132 UNCERTAINTY RANGES ASSOCIATED WITH EPA'S ESTIMATES ]
Table C.1 Low and High Estimates of the Area of Land Close to Sea Level by Region (continued)
Meters above Spring High Water
Jurisdiction
low
high
low
high
low
high
low
high
low
high
low
high
low
high
low
high
low
high
low
high
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
District of Columbia
2
3
3
4
4
6
5
7
7
9
9
11
11
13
13
15
14
16
16
18
Virginia
34
172
131
369
268
583
445
805
622
1088
815
1383
1073
1677
1362
1915
1634
2141
1857
2366
fresh
1
26
15
50
33
75
50
106
67
152
89
198
125
244
169
292
214
340
260
394
vulnerable
3 8
7
17
14
26
22
35
30
40
37
44
42
48
46
51
50
53
52
55
saline
30
138
108
302
222
482
373
665
525
896
689
1140
906
1385
1147
1573
1370
1748
1545
1916
Virginia Beach
Atlantic Coast
7
27
25
56
45
99
78
142
118
180
158
219
196
257
235
288
272
299
293
310
Pamlico Albemarle Sounds
621
1028
1186
1519
1684
2052
2239
2601
2774
3108
3274
3629
3827
4244
4449
4789
4932
5173
5269
5441
Atlantic Coast of NC
103
151
182
238
273
336
370
429
458
507
529
579
603
655
682
740
768
829
855
908
Total NY to NC
945
2136
2218
3585
3498
5089
4903
6569
6272
8008
7567
9463
8991
10994
1052012370
1187613515
13001
14486
Wetlands
Tidal
Cumulative amoiint of MnntiHal \A/c»tlanHc hc»lr\\A/ c\ ni\/£»n o I ox/at inn
v_/u 111 u i c4 u v c ^ lulcj i j ciiiiuuiii ui i \Ui i uucii V V c u cj 11 u o uciuvv cj y i vei i l^icvciliuii
L.I. Sound and Peconic
36
1
2
2
4
3
6
4
7
6
8
7
9
8
10
9
11
10
12
11
13
South Shore Long Island
104
1
4
4
7
6
9
8
10
9
11
11
12
11
13
12
13
13
14
14
15
NY Harbor/
Raritan Bay Total
68
0
3
2
6
4
8
6
9
7
10
9
11
9
12
10
13
11
14
12
16
New York
9
0
1
0
1
1
1
1
2
1
2
2
2
2
2
2
2
2
2
2
2
New Jersey
59
0
2
2
5
3
7
5
7
00
CD
7
9
8 10
9
11
9
12
10
13
New Jersey Shore
524
11
52
42
92
72
129
101
157
128
181
152
205
174
227
196
249
216
269
237
286
Delaware Bay Total
497
16
54
45
90
72
121
98
139
121
156
140
173
157
188
172
202
186
214
199
224
New Jersey
261
9
38
30
67
51
92
72
106
91
119
105
132
118
142
129
153
139
161
148
167
Delaware
236
7
16
15
23
20
29
26
33
31
37
35
41
39
46
43
49
47
53
51
57
Delaware River Total
216
12
41
33
64
49
85
65
93
80
101
90
108
97
115
103
122
109
127
116
133
Delaware: fresh
5
0
1
1
1
1
2
2
2
2
2
2
3
2
3
3
3
3
3
3
3
Delaware: saline
69
1
3
3
3
3
4
4
5
4
5
5
6
5
6
6
7
6
7
00
New Jersey: fresh
29
0
10
6
20
12
29
19
31
25
34
29
37
32
40
34
43
37
46
39
48
New Jersey: saline
108
10
25
22
35
30
44
37
47
44
50
47
52
50
55
52
57
54
59
56
62
Pennsylvania
6
1
2
1
4
3
6
00
6
9
7
10
8 11
9
12
9
12
10
12
Atlantic Coast of
Del-Mar-Va Total
757
3
13
13
26
24
38
36
49
47
57
55
64
62
70
68
74
73
78
77
82
-------
133 ]
21
34
27
1132
13
497
142
259
95
0
622
118
45
458
96
4695
710
7401
4486
7401
; Area of Land Close to Sea Level by Region (continued)
Meters above Spring High Water
low high low high low high low high low high low high low high low high
2
0
1
44
1
29
2
26
1
0
14
1
2
12
5
4
5
151
2
88
9
69
10
0
60
12
5
43
6 21
2083 2625
197 255
2374 3221
945 2136
2374 3221
8819 10857
4
5
4
143
2
92
7
79
6
0
49
8
4
37
10
259
3
137
18
101
18
0
119
21
11
87
20 37
2772 3039
275
3351
315
3940
2218 3585
3351 3940
11069 13025
7
9
9
231
2
136
14
110
12
0
92
14
9
69
10
13
16
375
4
194
28
137
29
0
178
30
18
129
33 47
3130 3320
335 374
3959 4512
9
14
13
334
3
189
23
147
19
0
141
21
15
106
12
17
19
489
4
244
38
166
40
0
240
37
25
178
42 57
3401 3562
393 429
4487 5001
11
18
18
425
4
231
32
170
28
0
190
27
21
142
14
22
21
590
5
282
48
182
51
0
302
44
28
230
52 66
3640 3789
448 481
4963 5449
13
22
20
510
5
267
42
188
36
0
239
34
26
179
16
26
23
699
6
329
62
206
61
0
363
52
31
280
61 73
3852 3984
495 525
5381 5864
15
26
22
618
6
315
57
213
45
0
296
40
30
227
17
28
24
809
8
377
79
228
70
0
424
59
35
330
69 81
4045 4173
538 568
5788 6266
16
29
23
724
7
361
74
232
55
0
356
47
33
276
18
31
25
911
9
420
99
242
79
0
481
70
38
373
76 88
4235 4352
583 616
6189 6652
Cumulative (total) amount of land below a given elevation
3498 5089
3959 4512
1295715101
4903 6569
4487 5001
1489017070
6272 8008
4963 5449
1673418957
7567 9463
5381 5864
18448 20826
8991 10994
5788 6266
20279 22760
10520 12370
6189 6652
22208 24521
------- |