Pilot tests on a North Carolina coastline prove that RTK GPS is a sound method for coastal monitoring surveys.

Each year, wind and waves carry tons of sand from one shoreline and deposit them along another. These processes of erosion and accretion can dramatically affect coastal resources, businesses and residences. If not carefully monitored, the beach you're dreaming about today may not be there tomorrow.

Case in point: Coastal North Carolina. This area has been hard hit by numerous hurricanes in recent years, including Hurricane Bertha in 1996, Hurricane Fran only two months later, Hurricane Bonnie in 1998, Hurricanes Dennis and Floyd in 1999 and, most recently, Hurricane Isabel in 2003. As a regular duty of the U.S. Army Corps of Engineers, the agency studies and monitors many of the nation's beaches to determine coastal nourishment and re-nourishment responses, especially in response to such devastating storms. To help them make these determinations, the Corps uses coastal monitoring surveys.

As part of an indefinite quantity contract with the Corps' Wilmington District, McKim & Creed, a multi-disciplined firm with locations throughout the Southeastern United States, provides coastal monitoring surveys along many beaches of the North Carolina coast. In the spring of 2003, in a monitoring survey for barrier island beaches in Brunswick County that included Ocean Isle Beach, Holden Beach and the Shallotte Inlet area, the Corps required that Real-Time Kinematic Global Positioning System (RTK GPS) be used for monitoring water levels. The scope of work stated that Differential GPS (DGPS) data be augmented by heave/pitch/roll information to smooth the vertical position in the post processing of the data. McKim & Creed went one step further and proved that, with their systems, the RTK GPS data could be used in real time for heave corrections, eliminating many of the post-processing requirements. This, in turn, can reduce the cost of the coastal monitoring surveys and provide the accuracy needed for better planning and design documents that can lower construction costs, and ultimately save taxpayers money. But working on the coast is no easy job.

In performing beach profiles, McKim & Creed surveyors use conventional total stations to survey up to and over the primary dunes (shown here) to below -4' NGVD29, which is approximately chest high water at low tide. These surf zone areas are surveyed at low tide so the data will extend seaward enough to be overlapped with the hydrographic data.

Coastal Surveys: Challenging Surfaces

Coastal monitoring surveys, which include beach profiles, ebb shoal surveys and inlet surveys, help the U.S. Army Corps of Engineers to keep beaches viable by ensuring that erosion doesn't destroy the beaches. In performing coastal monitoring surveys, surveyors and hydrographers measure from the landward toe of the primary dune to several thousand feet off the coast. But it's no easy task. One of the greatest difficulties in performing coastal hydrographic surveys is measuring the surface of the water, or tides, correctly.

"Tide" refers to the elevation of the surface of the water, disregarding waves and swells. Tide values vary greatly with geographical location and are affected by up to 37 different forces, including the gravitational forces of the sun and the moon, winds, currents, and the topography and bathymetry of the area. The accurate measurement of tides is extremely important in coastal hydrographic surveys because the origin of all measurements is a point attached to the surface of the water.

Typically, tides have been measured by a tide gauge or a staff attached to a fixed structure or anchored to the bottom, somewhere in the general vicinity of the project. The tide (water level) at the gauge is monitored and recorded at a pre-determined interval, either automatically or by manual readings. These tide readings are entered into a post-processing software program that interpolates between readings and applies the corrected tide to each individual sounding. McKim & Creed uses HYPACK MAX by Coastal Oceanographics Inc. (Middlefield, Conn.).

But there are several problems with measuring tide this way. For one, there are very few hard structures, if any, in the ocean suitable for attaching a gauge or staff. Further, it is quite difficult to attach and maintain a gauge or staff on the structures that do exist, due to the surf and height of such structures. What's more, tide readings are not taken from the location being surveyed and any interpolation is just that; there isn't an actual tide for each sounding. These readings are measuring a relatively flat surface, or tidal curve. Swells and waves are not taken into account. An Inertial Motion Unit (IMU, or heave compensator) must be used. And finally, tides are post-processed; therefore, there aren't any real-time bottom elevations. When surveying from offshore into the surf zone, hydrographers have to overlap the offshore portion of the survey with the inshore portion done by field crews on the beach and, while wading, in the surf zone. McKim & Creed field crews survey out to somewhere between -5' and -6' NGVD29. At high tide, McKim & Creed hydrographers use a 22-foot MonArk vessel to survey the ocean bottom from several thousand feet offshore to -4' NGVD29. The data from field crews and hydrographers is then overlapped to ensure that the data has been covered by the boat and by wading. However, surveying in the surf zone can be dangerous and launching additional sessions with the boat to close the data gaps is expensive.

Today's RTK GPS: A Sound Solution

When RTK GPS technology first arrived on the scene years ago, hydrographers knew it would have a tremendous application in hydrographic surveying by providing an accurate z-positioning component of the sonar transducer at an exact place and time. In this reference, z is the vertical vector of the vessel; in this case NGVD29. The problem was, however, that nobody could precisely time-tag that z value so it could be linked with the xy values.

This was primarily due to the long and inconsistent latency times for the GPS equipment. In other words, when a GPS receiver was told to record a position, it would take one to five seconds to perform the calculations and come up with a correct z value. Neither the GPS nor the data collection software could accurately define precisely where, in that one to five seconds, that value came from. This is much less of a problem on land, where if the fact of latency is known, the surveyor can be sure to hold the antenna in a single position long enough to prevent ambiguity. At sea, the boat is moving horizontally and vertically, thus only millisecond latency is acceptable.

In a steep, wind-driven sea, a small vessel can go from a crest (the highest point) to the trough (the lowest point) of a wave in less than one second. For years, RTK GPS was used to provide tidal data of the vessel by averaging one- to three-minute RTK GPS z values. Any sudden changes in the z axis had to be measured using an IMU, a series of acceleration and motion instruments that work together to measure motion in the x, y, and z axes. Though the IMU is a highly sensitive and accurate device, it is prone to errors when course or speed suddenly change. These errors can sometimes go undetected. The IMU is also very expensive and can be very complex to install.

Today, however, improved algorithms and technological advancements have been made in RTK GPS technology. Horizontal position update rates have been increased to 20 Hz (20 times/sec), while latency time has been decreased to about 0.2 second. Both are important improvements. The whole integrated system (IMU, sonar, GPS) is only as fast as its slowest component. The other devices (IMU and sonar) update at 20 Hz or faster. In the past, RTK GPS update rates of 1 Hz were most common. Therefore, hydographers did not get a sounding at the most critical point. Latency has to be short and constant, otherwise the precise time-tagging cannot be done.

McKim & Creed uses a 24-foot SeaArk and a 22-foot MonArk to survey from several thousand feet offshore to -4' NGVD29. The RTK GPS rover receiver is mounted directly onto the vessel, so that the vessel itself essentially becomes a tide gauge, providing real-time tides at the precise location where the survey is being performed.

RTK GPS: Onshore and Offshore

For the Brunswick County barrier island monitoring project, McKim & Creed surveyors set out to use their RTK GPS systems-Trimble (Sunnyvale, Calif.) 5700/5800 units-to recover existing control, establish new or additional control on the beach, and locate the first line of vegetation. Using Topcon (Pleasanton, Calif.) GTS 223 and Leica (Norcross, Ga.) TC703 total stations, both with TDS (Tripod Data Systems, Corvallis, Ore.) software on Ranger data collectors, and self-made fixed height prism poles that were 10 feet to the center of prism, a team on the beach surveyed from inland of the primary dune, over the primary dune and in the surf zone to approximately chest high at low tide. (On other projects, McKim & Creed hydrographers have mounted the RTK GPS systems to a Honda Rancher all-terrain vehicle (ATV) to survey the areas of exposed beach between the primary dune and the low water line.)

At high tide, instead of measuring the tides in the conventional manner with a tide gauge or staff set on a fixed structure in the general vicinity of the site, McKim & Creed hydrographers mounted an RTK GPS rover receiver directly onto a 22-foot MonArk vessel to survey the ocean bottom from several thousand feet offshore to chest height. This essentially converted the survey vessel into a tide gauge, provided real-time tides at the precise location where the survey was being performed, and eliminated the need for an IMU to measure the heave of the vessel caused by waves and swells. Also, referencing the RTK GPS antenna directly to the transducer eliminated errors caused by vertical changes of draft, settlement and squat. And with the real-time display on the boat, surveyors knew if they were able to get the required overlap of the areas covered by the boat and by wading. If sea conditions prevented the required overlap, surveyors could concentrate on the offshore portion of the survey until the tide came in or until the seas subsided, or both.

When using a remote tide gauge (not onboard RTK) surveyors are measuring their depths from the sonar transducer and then referencing them to the water surface at the tide gauge location. They have to account for the physical vertical difference between the transducer and the water surface, or draft. Likewise they have to account for the variable difference as the boat changes speed or has different loads (settlement and squat). When using RTK, the water surface is out of the equation. In McKim & Creed's case, surveyors still use the IMU to measure the roll and pitch angles, which in turn are used to calculate the height of the antenna; however, a lesser piece of equipment could be used. A roll and pitch (no heave) sensor, for example, is easy to install and use, and is much less expensive than an IMU. Without roll and pitch corrections, at a depth of 30 feet (which is the maximum depth for this type of work), a 10 degree roll (which is marginal working conditions) will cause an error of less than 0.5 feet vertically and 5 feet horizontally and the magnitude of roll is far greater than that of pitch. In the critical surf zone area of 10 feet or less, this error is far less.

For this project, all data from all equipment was simultaneously collected and stored, then used in post processing as needed. If hydrographers happened to lose the radio telemetry link to the RTK GPS base unit, they could still survey using DGPS (differential GPS) and the IMU. Additionally, McKim & Creed had multiple sources of data, including the heave measurements from the IMU and the RTK, positioning from DGPS and from RTK, profile overlaps from inshore and offshore, and RTK tides and NOAA tides, that could be processed and compared for quality assurance.

And aside from the occasional rough weather, ebb shoal grounding and high waves-at times getting so high they actually broke over the boat-the project was quite a success.

"Overall, the project went very well," says Kevin Shaver, McKim & Creed hydrographer. "The weather conditions were good, with the exception of the first two weeks in May. The RTK GPS system on the boat performed beyond expectations. Based on years of experience with RTK and hydro, we really didn't expect to be able to use the RTK for heave. We thought we would end up averaging the RTK readings to provide tides only, like we've done in the past, so this surpassed our expectations."

Gauging Coastal Success

In inlets, the vertical difference in the water surface can differ significantly in just a few hundred yards of horizontal separation. This makes it practically impossible, and definitely impractical, to accurately measure tides in a dynamic coastal environment using conventional tide staffs and gauges. Even though an IMU offers accuracy, in measuring heave, roll and pitch, it cannot measure the actual elevation of the water's surface. And, it is an expensive piece of equipment.

This project proved that RTK GPS can be used to not only accurately measure the tide at the precise location and time of the soundings, but also to accurately measure heave, even in the critical surf zone where expensive IMUs have failed. For agencies and organizations needing to monitor shorelines, like the Corps of Engineers, this translates into highly accurate and affordable surveys that keep our shorelines healthy and alive.

For the residents of and visitors to Ocean Isle Beach, Holden Beach and the Shallotte Inlet area, this project provides vital information to the U.S. Army Corps of Engineers that helps track erosion and accretion-and helps ensure that these beachfront communities will continue to exist for many years to come.

Glossary of terms

Draftis the vertical distance from the water surface to the sonar transducer.

Ebb shoal is a sandbar at the mouth of an inlet that is formed from material being deposited on an ebb tide (outgoing). Ebb shoals are often exposed above the water at low tide. Surveying the ebb shoals can be particularly dangerous due to the possibility of running aground and the treacherous surf and currents caused by the shoal.

Heave is the up and down (vertical) movement caused by the waves and swells.

Pitch is the front to back (fore and aft) motion caused by waves and swells.

Roll is the side-to-side (starboard and port) motion caused by waves and swells.

Settlement and squat are the vertical movements that take place when a vessel changes its speed over ground. Squat and settlement values are unique to every vessel. They are affected by the physical properties of the vessel as well as the location of the transducer.

Sonar transducer (often called the sonar head) is the part of the sonar that transmits and receives the sonar pulse. Usually mounted in the hull of a survey boat, it is the reference point for measuring depth.

Hydrographic Surveying Using RTK GPS

These figures illustrate the benefit of using RTK GPS for the z component of a survey vessel performing hydrographic surveys in the coastal environment. The final figure shows the most desirable result. This data was collected in Shallotte Inlet in April of 2003 on McKim & Creed's survey vessel, the Hydro Philly. All three slides show the same survey data collected in one pass, but each individual slide applies water surface corrections (heave and tide) from different sources, or different combinations of sources. None of this data has been edited.

This survey line starts (left side) approximately one mile off the beach in 31 feet of water and runs through the inlet, where it terminates (right side) at the northern bank of the Atlantic Intracoastal Water Way (AICWW), which runs in an east-west direction. This survey line was run from south to north during a flood tide.

The top window is a plan view of the track line. Note that the vessel had to go around a shoal at station 30+50. The bottom window is the profile view. The red trace represents the bottom as measured by 200kHz sonar and the blue trace represents the water surface. The spikes from station 39+00 to 36+50 and from station 31+50 to 29+80 are due to turbulence in the extremely shallow water as the vessel crossed over the shoals. This spiked data is bad data and would be edited from the final soundings.

Figure 1-1. No heave corrections from either the IMU or RTK GPS have been applied, resulting in the up and down movement of the vessel showing as the bottom trace. Waves outside the shoal are 2 to 3 feet.

Figure 1-2. Heave corrections (blue trace) were applied from the IMU only. Note the bad heave data to the far right, as the boat began to slow down and turn away from the bank. Tides were not being applied here in real time. A tide gauge at a nearby channel marker was being monitored. The tide readings for the duration of this survey line remained the same at 0.8 feet.

Figure 1-3. Water surface (heave and tide) corrections were applied from RTK GPS only. Note the water surface bulge behind the first shoal. The water surface at the north end of the line (right side) was 1.2 feet below the water surface on the outside.