Geotimes
Feature
LIDAR: Mapping a
Shoreline by Laser Light
James C. Gibeaut
Sidebar: Policy-makers
go on a virtual field trip
One evening in November 1997, at the end of another long field day along the Texas
coast, our team of geoscientists compiled data from topographic transects and
made quick checks for errors. It’s a good day if we measure 15 transects,
called beach profiles, over the dunes and across the beaches, spaced out along
30 kilometers of shoreline.
That day, the data were free of errors, so we
had some satisfaction until we thought about what was missed. Fifteen discreet
lines spaced 2 kilometers apart to describe the shape, height and volume of beaches
and dunes is rather pitiful. We only measured about 0.1 percent of the topography.
And how about the locations of the transects? Were they representative of the
beach in that area? Through time, did they become unrepresentative because of
a dune blowout or construction?
Pictured here is Surfside, Texas, following
Tropical Storm Frances in 1998. This storm renewed Texas’ efforts in understanding
and mitigating shoreline retreat. All images courtesy of James Gibeaut.
Back then, the process worked, but we were missing a lot. When we would arrive
on a profile site (either by truck, boat, foot or helicopter), we’d first
try to locate the survey mark from which we started the profile last time. Of
course, more times than we liked, the marker was lost to erosion, burial or construction,
and we had to take the time to install and survey a new one in order to acquire
repeatable data.
After finding the old or surveying a new starting point, we would set up our survey
instrument and send the person with the reflecting prism on the end of a survey
rod along the profile line oriented perpendicular to the shoreline. First, we
would have to make the decision of where to put the profile along the beach. Next,
the rod person would decide which points were significant enough to take the time
to measure along the profile line before we would move to the next profile. And
the process continued the same way for each profile.
For coastal geologists studying how beaches and dunes change in time and space,
something new was needed. Airborne topographic LIDAR mapping is now making a huge
contribution to solving this field-sampling dilemma. LIDAR — for “light
detection and ranging” — systems send out pulses of laser energy to
measure the distance to a surface that reflects the energy. By combining information
on the exact position and pointing direction of the laser source, as measured
onboard an aircraft, we can determine the 3-D position of the reflecting surface.
We can now acquire data with 10 centimeters accuracy and submeter data-point spacing
along hundreds of kilometers of shoreline in a day.
Historically, coastal geologists and engineers conducted regional studies using
sparse data or local studies using detailed data. Subtle variations in topography
and height above sea level are a result of active processes, which control the
distribution of habitats, such as beaches, dunes and wetlands. Detailed and accurate
topographic data, therefore, is fundamental to making significant advances in
understanding the dynamic coastal zone. A rise or fall of just 10 centimeters
in sea level relative to the land surface can cause habitats to shift tens to
hundreds of meters horizontally. Current U.S. Geological Survey topographic maps
are usually more than 10 years old and typically have contour intervals of 5 feet,
which is far too coarse for adequately describing the topography of a barrier
island, for example.
In Texas and elsewhere, LIDAR is becoming an invaluable tool for researchers in
understanding the wide array of habitats, geomorphologic features and processes
active in the dynamic coastal zone. One specific application of LIDAR is mapping
the shoreline for changes over time.
Change along the Texas shoreline
The Texas Gulf of Mexico shoreline is 600 kilometers long and is generally
undergoing long-term (tens of years) retreat. Rates of change in specific areas,
however, vary greatly from seaward advance of 1 meter per year to retreat of
more than 3 meters per year. Bay shorelines, which total more than 9,000 kilometers
in length, are also generally retreating.
In 1999, largely in response to ongoing shoreline retreat intersecting new and
old development along the southeast Texas Gulf of Mexico coast, the Texas Legislature
passed the Coastal Erosion Planning and Response Act — authorizing the
Texas General Land Office to implement a statewide coastal erosion response
program.
The Bureau of Economic Geology (Bureau) has been helping to identify critical
coastal erosion areas in support of implementing the act. For decades, the Bureau,
which is part of the John A. and Katherine G. Jackson School of Geosciences
at the University of Texas at Austin, has had a strong program in coastal research,
studying coastal processes, evaluating the distribution of different habitats
along the coast, and mapping and analyzing the changing shoreline.
Because accurate
topographic information is vital to our understanding of coastal environments,
in 2000, the Bureau purchased a LIDAR system manufactured by Optech Inc., and
embarked on the Texas Shoreline Change Project (TSCP). The goal of the project
is to establish a state-of-the-art shoreline monitoring and shoreline change
analysis program that will help guide coastal erosion and storm hazard mitigation
projects along bay and Gulf of Mexico shorelines.
Pictured here is a perspective view of the digital elevation model for a portion
of Matagorda Island in Texas, produced with LIDAR.
Comparing shoreline positions through time yields the rate of shoreline change.
To arrive at an annual rate of change, it is best to compare multiple shoreline
positions over a long period. The central concept is that past long-term shoreline
movement is the best indicator of what is likely to happen in the long-term
future. This approach works provided there is no fundamental change in the future
system, such as construction of large engineering structures, including seawalls
or jetties, which can change the movement of sand along the coast. The construction
of jetties extending well seaward of the surf zone at bay entrances, for example,
occurred during the early 1900s and has greatly affected shoreline change patterns
along the Gulf coast.
Shorelines mapped in the 19th century by the former U.S. Coast Survey are the
earliest source of shoreline position accurate enough for shoreline change analysis,
but the bulk of our shoreline data in Texas comes from vertical aerial photography
taken since the 1930s. We scan the photos and digitally match them to 1990s
digital orthophotos produced by the Texas Strategic Mapping Program. Orthophotos
are photographs that were scanned and processed to remove distortion. They are
geo-referenced using surveyed locations of features or targets visible in the
photographs.
Unfortunately, the old photographs were rarely taken at any specific time with
regard to the tides, so documentation regarding the time of day is usually not
available. Therefore, the waterline is not at a consistent or known elevation
in the photographs, and even if it were, the level of the waterline is not the
best shoreline indicator to map for change analysis. Therefore, a shoreline
feature must be interpreted from 2-D photography and then digitized. This boundary
between wet and dry sand on the beach is displayed as a tonal contrast on the
photographs.
The position of the wet/dry line is more consistent in time than the waterline
and approximates the landward extent of the waves rushing up the beach during
the last high tide. It is also high enough on the beach so that short-term erosion
and deposition that occur in the swash zone on a daily or monthly basis do not
skew the long-term analysis.
Although not affected as greatly as the waterline, the wet/dry line is still
shifted by short-term changes in water level, wave activity and recent rainfall,
which have nothing to do with the shoreline changing. Furthermore, the wet/dry
line is not always a distinct and easy line to draw on the photographs. Error
is also involved when attempting to take the distortion out of a photograph
by digitally warping it to match an existing map (or orthophoto), or by going
into the field and surveying positions of features or targets visible in the
photograph and then warping the image to match those features. This procedure
can be difficult in remote areas with few distinct features to use as reference
points or where access is difficult. These problems have been overcome by applying
LIDAR to map shorelines; however, LIDAR also poses its own unique set of challenges.
Surveying mechanics
Airborne LIDAR mapping requires the integration of three basic measurement
sources: laser ranges and associated scan angle information; information on
the roll, pitch and yaw of the aircraft from an Inertial Measurement Unit; and
absolute aircraft positions (trajectory), derived from a differential, geodetic
GPS network. The slant distance from the aircraft to the ground is determined
by measuring time elapsed from when the laser is fired to when energy reflected
by the target returns to the sensor, then multiplying this time by the speed
of light. We divide this distance by two to account for the two-way travel time
and combine it with data from the other measurement systems to yield the horizontal
and vertical position of the reflectors. In our Optech LIDAR system, the laser
shots are directed from side to side across the flight path by a mirror oscillating
in a full range of up to 40 degrees.
The concept of how the survey positions are obtained is rather simple; however,
considering the aircraft’s height above the ground, the aircraft’s
speed and turbulent movement, and the fact that the laser is firing upwards
of 25,000 times per second, actual implementation of the concept is quite a
feat! We must take extreme care in conducting the surveys to get the best results.
Many things can go wrong.
Successful airborne surveys begin with preparation on the ground. The Bureau
uses a network of tidal stations, survey monuments and temporary reference points
along the Gulf of Mexico coast as GPS base stations during LIDAR surveys. The
farther the aircraft is from the ground reference station, the more potential
error there is in computing the aircraft trajectory. Therefore, we choose base
stations, which are occupied during the survey flight, along the flight path
no farther than 100 kilometers apart so that the airplane is always within 50
kilometers. We also conduct a high-accuracy ground survey of a stretch of road,
runway or some other stable surface with no vegetation on it in the survey area.
The ground survey is used to calibrate the LIDAR system, and we pass over it
during each flight.
Other factors involved in planning a coastal LIDAR survey include determining
the period when the GPS satellites form the best constellation for the most
accurate results, when the tides, waves and weather conditions are most favorable,
and when survey operations won’t interfere with other aircraft in the area.
Our LIDAR system uses a near-infrared frequency for the laser, which is optimum
for terrain but cannot penetrate water, so the survey target must be above the
waterline.
Another timing restriction is daylight. Surveys are conducted in daylight for
safety reasons if using a single-engine aircraft or when navigating the shoreline
by sight, which we have found to be most efficient for relatively straight,
open-ocean shorelines. The laser energy cannot penetrate fog, rain or clouds,
so it either has to be clear or we need to fly below the clouds. We can’t
fly too low, however; the laser is too strong and can damage people’s eyes
on the ground. The swath width also becomes narrower the lower we fly. The lowest
we prefer to fly is 450 meters above the ground, but for beach surveys, the
ideal flight altitude is about 800 meters with an aircraft speed of about 100
knots. This height and speed combination gives us a 550-meter wide swath, 1-meter
data-point spacing and potential accuracy of better than 10 centimeters vertically.
We fly the shoreline using a video camera that has the same field of view as
the laser. The pilot maneuvers the aircraft to keep the waterline in the middle
of the view. This gives us data to about 275 meters landward, or one-half the
swath width, more than enough to cover the first two rows of dunes and structures.
We make at least two passes along the shoreline to increase the density of the
data and to fill in any holes from prior cloud cover.
Extracting a shoreline
Having collected these LIDAR data, we must couple them to the older data to
see the shorelines’ history. We must extract a shoreline from the LIDAR
data that is comparable to shorelines mapped from historical photography. We
therefore need to determine the typical elevation of the wet/dry line. Going
back to our ground-surveyed beach profiles, we noted the location of the wet/dry
line, and we determined that it averages around +0.6 meters above mean sea level
along the Texas Gulf coast. This also generally coincides with being just below
the major berm crest on the beach.
To get the shoreline from the LIDAR data, we process the millions of LIDAR data
points into a 1-meter grid, contour the grid, and extract the +0.6-meter contour
line. After “cleaning” the contour line to eliminate effects of cars,
people and small closed contours, we have a shoreline that is comparable to
our historic shorelines but one that is much more accurate and rigorously selected.
In addition to measuring the two-way travel time of the laser pulse, our LIDAR
system also measures the intensity of each returning pulse. Wet sand causes
a lower-intensity reflection than dry sand, which allows us to map the wet/dry
line in the LIDAR data. We can then measure the horizontal and vertical variation
of the wet/dry line. As we collect more LIDAR data we can determine the variability
of the wet/dry line alongshore and through time, giving us a better idea of
the reliability of the shorelines mapped from photography.
While the legacy database of historical photography is invaluable in understanding
the process of shoreline change, LIDAR is the future for addressing this problem.
Annual LIDAR surveys can be used for short- and long-term shoreline change studies.
The data can provide a baseline for studying the effects of storms that may
occur during the year and for delineating areas prone to damage caused by washover
or episodic retreat during subsequent storms.
There is much to be done with regard to developing and applying LIDAR technology,
and with our research and development partners, we are applying new instrument
components and developing surveying, processing and analysis techniques for
coastal surveys. LIDAR will allow major advances in our understanding of coastal
environments, just as the advent of vertical aerial photography did in the 1930s.
 Policy-makers
go on a virtual field trip
Policy-makers, who have varied and generally nonscientific backgrounds,
often find themselves trying to digest the tables and graphs of data technical
staff may present to them as background for making decisions. And it can
be difficult for an environmental policy-making panel, such as the 12-member
Texas Coastal Coordination Council, to make field trips together to observe
the issues they are facing.
So the Bureau of Economic Geology, in conjunction with the Texas General
Land Office in March 2002, used LIDAR data to make 3-D, stereo models
of the beaches, dunes and geotextile tube erosion control structures along
the Texas coast.
During a public meeting, we projected the model on a 10-foot high by 14-foot
wide screen. Council members went on a virtual “fly-through”
tour of 100 kilometers of shoreline to observe the effects of erosion
and geotextile tubes on the geomorphology of the beaches and dunes.
The detailed topographic models, made possible by LIDAR, allowed Council
members to make their own observations and placed other information, such
as geotextile tube integrity and beach width, in an intuitive context
for them. This experience showed the usefulness of LIDAR topographic models
and virtual-viewing technology to present information and concepts to
the public and in our schools.
JCG
Back to top
|
Gibeaut is a Research Associate
at the Bureau of Economic Geology in the John A. and Katherine G. Jackson School
of Geosciences of the University of Texas at Austin. E-mail: jim.gibeaut@beg.utexas.edu.
Much of this work is conducted in collaboration with the Center for Space Research
at the University of Texas at Austin and has support from the Texas General Land
Office, the Texas Coastal Coordination Council, NASA, the Navy, the Army Research
Office and Optech Inc.
Back to top
