Figure 1. Overview of Critical and Emerging Critical Areas along Columbia River

Transcription

1 GPS Derived Water Levels for Large Scale Hydrographic Surveys: Implementation of a Separation Model of the Columbia River Datum, A Case Study Crescent H. Moegling 1, Jon L. Dasler 2, Jason C. Creech 3, Peter Canter 4 Project Description The National Oceanic and Atmospheric Administration s (NOAA) Office of Coast Survey (OCS) assigned project OPR-N338-KR-08 in the Columbia River to David Evans and Associates, Inc. (DEA) in The task order was part of the ongoing OMNI contract which DEA (and other firms) provide hydrographic surveying in support of OCS charting mission and reduction of critical survey backlog. Due to DEA s long history and experience surveying for other federal, state and local agencies in the Columbia River, they were selected as the best firm to perform this work. The project required IHO Order 1, object detection surveys in six areas. These data will update six charts and two insets along nearly 100 nautical miles of the Columbia and Willamette Rivers (Figure 1). Figure 1. Overview of Critical and Emerging Critical Areas along Columbia River NOAA establishes hydrographic survey priorities for general and long term project scheduling purposes. The priorities are outlined in the NOAA's Hydrographic Survey Priorities (NHSP) document. NOAA continually reviews the marine community s needs for charting products and hydrographic surveys and reviews and updates this document annually. 1 NOAA Office of Coast Survey, Silver Spring, MD, USA 2 David Evans and Associates, Inc., Vancouver, WA, USA 3 David Evans and Associates, Inc., Richmond, VA, USA 4 Applanix Marine Systems, Richmond Hill, Ontario, Canada

2 Navigationally significant areas are subdivided based on the need for hydrographic surveys. The highest priority areas are called critical areas. Critical survey areas are defined as waterways with high commercial traffic volumes (cargo, fishing vessels, cruise ships, ferries, etc.), extensive petroleum or hazardous material transport, compelling requests from users, and/or transiting vessels with low under-keel clearance over the seafloor. A new category, named emerging critical area, was created in 2004 to allow for designation of additional areas that now meet the definition of critical area, but were not included in previous editions of the NHSP. The area surveyed by DEA was comprised of critical and emerging critical areas. Prior to the issuance of the task order for this project, DEA proposed conducting a survey using GPS derived water levels to OCS. Traditionally this survey would have been reduced to chart datum using discrete tidal zoning provided by NOAA s Center for Operational Oceanographic Products and Services (CO-OPS). There are 57 zones within this project area controlled by five permanent water level stations on the Columbia River and one water level gauge on the Willamette River installed for this project. DEA has considerable experience not only surveying using tidal zoning for reducing bathymetry, but also the application of GPS derived water levels and dealing with the complex gradient of the Columbia River Datum. OCS accepted DEA s proposal to survey using ellipsoidal referenced elevations to derive corrections relative to Columbia River Datum with the condition the results be compared to traditional zoning methods. NOAA s Perspective OCS continues to rely on the expertise of their contracting partners in advancing this technology for NOAA s in-house hydrographic survey platforms. OCS has a goal to incorporate ellipsoidal referenced surveys by 2010 (reference Field Procedures Workshop Presentation by LCDR EJ Van Den Ameele). In preparation for becoming an accepted operational method, OCS Coast Survey Development Laboratory (CSDL) will continue to utilize their Research to Operations approach. This approach incorporates a managed process as follows: 1. Research of Potential Projects 2. Developmental Test and Evaluation 3. Operational Test and Evaluation a. This is the current phase of the project. OCS plans to conduct operational testing aboard the NOAA Ships RAINIER, FAIRWEATHER and THOMAS JEFFERSON in 2009 field season. In addition a Tide Buoy will be tested aboard Bay Hydrographer II. 4. Transition/Implementation a. Draft transition plan to be completed by end of 2009 field season. 5. Operational Support

3 OCS feels there are advantages and disadvantages to this approach. Advantages Decouple tide measurements from survey Reduce vertical uncertainty from heave, dynamic draft, static draft, and tidal zoning Reduced horizontal uncertainty Archival advantages: data de-linked from local sea level datums Seamless data sets Disadvantages Shore Support effort not eliminated in all cases o GPS base stations or CORS Post-processing of position data Vertical datum relationships must be well-known o Alaska, Pacific Islands have poor Geoid models

4 Charting Datum for the Columbia River Large inland waterways of the United States use adopted low water reference planes for a charting datum that are defined relative to orthometric heights (geodetic vertical datums). On rivers, the adopted low water is a gradient datum that follows the adopted low water elevation. Figure 2 illustrates typical low water reference planes used on Federal projects. Mean Lower Low Water Columbia River Datum Bonneville Dam Figure 2. Typical Low Water Reference Planes used on Inland Federal Projects (source U.S. Army Corp of Engineers) Columbia River Datum (CRD) is a non-tidal, defined gradient datum that was developed from a report by R. E. Hickson of the U.S. Army Corps of Engineers, Portland District in Mr. Hickson established water level gauges along the Columbia River and established low water levels at numerous points along the river. These observations are the foundation of the present day Columbia River Datum which is an adopted low water reference plane that runs from Harrington Point at River Mile 23 to Bonneville Dam at River Mile 145. It also extends up the Willamette River at the confluence with the Columbia River to Willamette Falls in Oregon City at Willamette River Mile 27. Historically CRD was defined relative to the National Geodetic Vertical Datum of 1929 (NGVD 29) at distinct river miles with linear interpolation applied between defined locations. Now that NGVD 29 has been superseded by the North American Vertical Datum of 1988 (NAVD 88) there is a mandate to define CRD relative to NAVD 88. There are discrepancies between agencies ranging from a few centimeters to 25 centimeters in the NAVD 88 elevation of CRD at discrete points along the Columbia River. This is further complicated by current water level gauges along the Columbia River which have some reporting error in water levels relative to CRD. Some of these errors were detected by the application of GPS water levels for the Columbia River hydrographic project. These discrepancies are due in large part to the level runs along either shoreline of the Columbia River without adequate ties across the river in the original level lines. GPS is playing a major role in defining this relationship and bridging the level runs across the Columbia River. The U. S. Army Corps of Engineers, Portland District is the steward

5 of Columbia River Datum and is in the final process of establishing the relationship between CRD and NAVD 88. Skamokawa Gauge Wauna Gauge Longview Gauge St. Helens Gauge Vancouver Gauge Bonneville Dam Figure 3. Columbia River Datum Relative to NAVD 88 and Location of NOAA Gauges. Conventional NOAA hydrographic surveys along the Columbia River Datum use discrete tidal zoning for providing water level correctors. Reference water level stations for zoning is provided by 5 permanent water level stations over a 100-mile reach from Harrington Point to Portland with up to 25 miles between stations (Figure 3). Zones may be over 12 miles from a reference station. The Portland District uses tide staffs at an approximate 2-mile interval along the river to control survey and dredging operations which were established by GPS observations or differential leveling. The Columbia River is an ideal candidate for the application of GPS derived water levels for a hydrographic survey since CRD is a chart datum that is defined by orthometric height rather than tidal datums. The only missing link is defining CRD directly to a reference ellipsoid. The process to define CRD relative to ellipsoid heights is simply a matter of combining the relationship between a chart datum relative to orthometric heights with a geoid model which defines orthometric heights to a reference ellipsoid (NAD 83 in this case) into a single model that may be used in existing hydrographic software such as Hypack and Caris HIPS. Use of this technology allows for direct water level measurements at the survey platform (a mobile water level gauge if you will). These previously unmeasured water level changes at the survey platform were conventionally modeled using discrete tidal zoning. Use of GPS water levels with measurements at the survey platform eliminates large vertical errors in the survey resulting from discrete tidal zoning in a river environment subject to significant water level changes resulting from tributary runoff and reservoir control or wind impoundment across open bays. Further reduction of vertical error in hydrographic surveys by application of GPS water levels is the elimination of the need to apply changes in static and dynamic draft. Tables to correct for dynamic draft (settlement and squat) are developed and applied relative to speed over ground. In

6 high current area, such as a river environment, squat tables relative to speed over ground do not accurately reflect actual dynamic draft. Dynamic draft is a function of speed through the water and not speed over ground. While maintaining the same speed through the water, speed over ground increases heading down current and decreases going into the current. The result is errors in the application of dynamic draft. Since the phase center of the GPS antenna and the acoustic center of the multibeam sonar are at a fixed distance and move in unison with vessel loading, settlement and squat, vertical uncertainties in static draft and dynamic draft are eliminated. Fundamentals of GPS Derived Water Levels The basic components that make up the measurement of the vertical in hydrographic surveys relative to the ellipsoid are illustrated in Figure 4. The illustration shows the relation of the GPS antenna to the water line, the separation model between the ellipsoid and chart datum, the chart datum, the reference ellipsoid and the seafloor. Figure 4. Vertical Components of Hydrographic Survey Relative to Ellipsoid Heights. Today s positioning technique for Marine Hydrography involves integrating GPS and inertial navigation systems (INS). The Inertial Measurement Unit (IMU) aids in reducing GPS noise as well as providing high bandwidth, un-interrupted solutions during GPS outages. For high precision surveys (5 cm or better in X, Y and Z), the most common technique has been the post processed Inertial-Aided Kinematic Ambiguity Resolution (IAKAR) mechanization, which requires a reference station in the proximity of the survey area. The reference station helps to mitigate atmospheric and satellite biases and to resolve integer ambiguities. In these cases the GPS rover position is required to be within 20 km of the reference station. Otherwise the atmospheric biases degrade the accuracy of the results, imposing significant limitations and increased survey costs. Since multibeam echosounders first started using Positioning and Orientation Systems (POS), the Heave channel was used as estimate of a vessel s vertical motion. The heave channel is based on

7 high-pass filtered double integrated vertical accelerations. With the recent introduction of the TrueHeave technique, the reliability of the Heave solution from POS has increased dramatically. Traditional Heave measures vessel motion over periods ranging from sub-seconds to 30 seconds. Nearby water level gauges can be used to measure tide and other factors impacting water levels which when combined with Heave can produce an estimate of water depth to the local chart datum. Squat, settlement, dynamic draft and other long period vessel motions must be accounted for independently as they are not included in the heave or water level gauge measurements. With carrier phase based GPS, hydrographers can now fill in the measurement gap between the Heave measurements and the water level gauge measurements by use of ellipsoid heights. The effects of squat, settlement, dynamic draft and changes in static draft can also be mitigated as the GPS is a fixed distance from the sonar acoustic center and vertical effects of these parameters are directly measured with GPS and inertial observations. There are two basic approaches to make use of carrier phase based GPS. These include Real- Time Kinematic (RTK), or Post-Processed Kinematic (PPK). These two methods can be further divided into a single base operation or a virtual base from a network of continuous operating stations. RTK requires the use of a reliable link between the base station and survey vessel. A radio link typically is used for marine applications as it provides a robust link with low latency of correctors. Cellular links have not proven to be as reliable and typically have too much correction latency for a tightly coupled inertial solution. RTK operations have advantages and disadvantages. Advantages Provides real-time water levels that can be used for color swath coverage relative to chart datum Provides real-time kinematic GPS quality control Seamless swath coverage in real-time for improved quality control and quality assurance for bathymetric data Disadvantages Shore Support needed for base stations and radio link Radio link can be problematic in some areas and may result in rejection of the survey if significant loss of corrections is encountered that can not be corrected with the strap down inertial system Range from base station is limited by radio link or 20 km Over large distances or difficult radio environments, post processing or PPK must be employed. Post mission processing removes the effects of gaps in real time correction reception over radio links or data gaps in general. Although such gaps only account for a small percentage of data acquisition they must be removed. Post processing involves the use of base station data recorded on the reference site which when combined with raw data recorded on the hydrographic vessel

8 results in a robust height solution which allows the hydrographer to confidently substitute Heave with ellipsoid height. In addition post processing employs a forward/backward smoother which is applied to the inertial position data. The effect of the smoother will be to significantly reduce error in any periods where GPS data is lost or suboptimum. A smoothed best estimate trajectory (SBET) of position is produced. SBET data provides sub 5 cm ellipsoid heights with a high degree of confidence. PPK operations also have advantages and disadvantages. Advantages Eliminates need for radio link Distance from base can be extended up to 20 km Improved accuracy over GPS outages from forward/backward smoothing Disadvantages Real-time water levels not available for inshore coverage evaluation or quality control of GPS and bathymetric data Range from base station is limited to 20 km The most recent versions of the Applanix POSPac and other post processing tools include the Post Processed Virtual Reference Station (PPVRS) technique, which makes use of GPS network stations to determine atmospheric biases at the rover positions and which tightly integrates GPS with inertial data to provide a continuous, high-precision navigation solution with baselines of up to 100 km. The expected result is a method for hydrographic survey with the necessary precision to eliminate the need for dedicated GPS reference stations for all of the continental U.S. inland waters and near shore areas. PPVRS changes the way hydrographic surveys are acquired by reducing logistics and ensuring sub 5 cm positioning. A comparison of the conventional single station technique with the PPVRS is shown below for a portion of the survey on the Columbia River. The difference in position is well below the error budget involved with tidal zoning and dead band vertical motion associated with the application of Heave.

9 Figure 5. Position Differences between SmartBase PPVRS (WGS 84) and SingleBase (NAD 83) Table 1: Normalized Statistics on Navigation Differences Element RMS SIGMA 68% 95% 100% NPOS cm EPOS cm DPOS cm Note, in Figure 5 there is an offset due to the application of different reference ellipsoids, however, it is the sub 5 cm variation in the vertical which is of interest. The offset is a function of the application of WGS 84 coordinates in the PPVRS and NAD 83 coordinates in the single base. Statistics of the navigation differences are presented in Table 1. It is a NOAA requirement that surveys be acquired on the NAD 83 datum. The optimum method for PPVRS is to compute on the WGS 84 surface using up to date reference site coordinates which compensate for crustal motion using ITRF estimates of reference site coordinates. The remaining challenge then is to convert the final PPVRS coordinates to the NAD 83 datum. Precise datum transformation data which correct for spatial variation in geoidal heights are not always available. For the survey presented in this paper, it was decided to use both RTK and PPK from a single base reference with the base site coordinate in NAD 83, although test areas and statistics are provided using the PPVRS option in conjunction with NAD 83 coordinates. Further, a model was developed to transform NAD 83 ellipsoid heights directly to Columbia River Datum. Testing was done using the SmartBase PPVRS by overriding WGS 84 base coordinates with NAD 83 coordinates and is presented in the results.

10 Columbia River Project Application of GPS Derived Water Levels A critical component for any hydrographic survey using GPS heights is a separation model from the reference ellipsoid to the chart datum. Detailed models of the relationship between chart datum and an orthometric height are required to transform ellipsoid heights to water level data above chart datum. Models such as those found in V-Datum are helpful, but need to incorporate a geoid model (model of ellipsoid heights to orthometric heights) into a single model for use in real-time applications such as Hypack or during post-processing of hydrographic data in Caris. The U. S. Army Corps of Engineers, Portland District provided the profile of Columbia River Datum relative to NAVD 88. The first step in generation of the model was to convert the profile down the river into a Triangular Irregular Network (TIN) spatial model of CRD relative to NAVD 88. The river profile was offset at 2,000-foot intervals perpendicular to the profile and modeled down back channels of islands. Figure 6 illustrates 1 cm contours of CRD relative to NAVD 88 from the resulting model. Note the multiple 1 cm contours that pass through discrete tidal zones. Some zones have more than 7 cm of gradient change across the zone which is not captured in the application of discrete zoning. 2000m Figure 6. Columbia River Datum TIN Model Relative to NAVD 88 (1 cm contours shown in red, 5 cm contours shown in blue and tide zones in black)

11 Data points that define the CRD surface relative to NAVD 88 were converted to NAD 83 ellipsoid heights using Geoid 03. The converted data was inserted back into the TIN model. Figure 7 illustrates 1 cm contours of CRD relative to NAD 83 ellipsoid heights from the resulting model. 2000m Figure 7. Columbia River Datum TIN Model Relative to NAD 83 Ellipsoid (1 cm contours shown in red and 5 cm contours shown in blue)

12 The final step was to convert the TIN model into a high resolution grid model. The grid model used the same format as geoid grid models generated by the National Geodetic Survey which can be used in hydrographic software such as Hypack and Caris HIPS to convert ellipsoid heights directly to a mapping datum. A series of three models were generated at a 3-second arc resolution (Figure 8) in order to capture the high definition of the TIN model in small channels and at the confluence with the Willamette River. 2000m Figure 8. High-Resolution Grid Model of Columbia River Datum GPS base stations were established along the project at intervals not greater than 18 kilometers. This allowed for a maximum range from a base station of not more than one half of the base station spacing or not more than 10 kilometers. Base stations logged 1- second epoch GPS observables and broadcast real-time carrier phase correctors to the two survey vessels every second. A repeater radio was used to relay the signal to reach the 10- kilometer range in some areas. Aboard the survey vessels, the correctors were received and processed by a POS/MV with dual frequency (L1/L2) receivers. As a redundant observation, the correctors were processed with a Trimble MS750 receiver which provided Hypack with real-time ellipsoid heights. The CRD separation model was input in Hypack for realtime water levels (Figure 9). Figure 9. CRD Separation Model in Hypack

13 To verify correctors were providing position data within survey specifications, a Trimble DSM132 receiving differential correctors from the U.S. Coast Guard beacon at Fort Stevens was used for a real-time comparison to RTK position data. As the survey vessels would switch between base stations, the RTK positions and elevations were logged as the vessel was static and a comparison was made from the values obtained from each station. Typical comparison values were sub 2 cm. Figure 10 shows a water level corrected multibeam swath painting in real time with a data display of real time water levels and status of RTK and differential positions. The real-time coverage display was color coded for a distinct color change at the 2-meter and 4-meter depth curves. Survey specifications called for full multibeam coverage to the 4-meter curve and 25-meter skunk striping or single beam coverage inshore of the 2-meter curve. Whereas the swath was corrected to chart datum in real-time, the colored depth curves could be used during the survey to verify coverage requirements were exceeded. Figure 10. Real-Time Hypack Survey Display with GPS Water Levels

14 Data Processing Workflow Data processing followed the typical Caris HIPS CUBE (Combined Uncertainty Bathymetric Estimator) workflow with integration of SBET data through the HIPS load Attitude and Navigation tool. SBET files from single base post processing were created using Applanix POSPac MMS 5.2. The NAD 83 to CRD model in binary format (.bin) was applied during the computation of GPS water levels in HIPS. After GPS water levels were applied, depths were reduced to CRD. Figure 11 shows the basic processing workflow. Figure 11. Workflow for GPS Water Levels Results Static Vessel Observation Adjacent to Gauges In order to evaluate the performance of GPS water level calculations a series of one hour long survey lines were logged while floating in the immediate vicinity of gauges assigned to the project, which included a subordinate gauge set specifically for the project and five CO-OPS water level gauges. GPS water levels were calculated relative to CRD using the standard project workflow for RTK solutions, single base IAKAR and PPVRS IAKAR. GPS water levels for each static observation navigation solution were plotted (Figure 12) with the water level gauge data and compared. The RTK and the single base solutions matched well

15 while there was a notable and unidentified offset in the PPVRS solution of approximately 6 cm. While apparent during this analysis prior comparisons between PPVRS solutions and real-time solutions showed agreement tighter than 6 cm Water Level (m) POS RTK POS RTK (Avg) 0.6 Single Base Single Base (Avg) PPVRS 0.5 PPVRS (Avg) Gauge 0.4 MS750 RTK MS750 (Avg) :12 19:18 19:24 19:30 19:36 19:42 19:48 19:54 20:00 20:06 20:12 20:18 20:24 20:30 20:36 20:42 Figure 12. Comparison of One Hour Series Water Level Observations Direct comparisons between the static vessel observations and water level gauge values were robust enough to enable the authors to detect a possible incorrect offset on one of the survey vessels as well as discover previously unknown datum adjustments at several water level gauges. Table 2 lists the comparisons between the navigation solutions and gauge output for the static vessel observation depicted in the figure above. The statistical analysis shows the GPS standard deviations from gauge values are better than 2 cm for the MS750, POS RTK, POS Single Base, and POS PPVRS although the Mean and Median are just over 5cm with the application of PPVRS. It is unclear why the PPVRS tracks 5 cm different from other solutions and may be related to the application of NAD 83 positions or distance from reference stations. The RTK and PPK Single Base solutions used a nearby base station. However, vertical noise levels for PPVRS and Single Base compare well and are both within specification. The offset is of an unknown nature and may be datum related. It is safe to say that the PPVRS is still within specification and equally mitigates the effect of unknown dynamic draft and long period changes in water level.

16 Table 2. Statistical Analysis of One Hour Series Water Level Observations POS RTK POS Single Base POS PPVRS MS750 Gauge Average Delta Average Delta Average Delta Average Delta 19: : : : : : : : : : : : : Mean Median Std Dev all values in meters GPS Water Levels Compared to Discrete Zoning In order to evaluate the performance of the two methods a sample dataset with depths reduced using GPS water levels was compared to a duplicate dataset using discrete zoning with verified CO-OPS water levels relative to CRD. Comparisons between the two datasets showed differences caused by discrepancies in CRD between the model used to compute GPS water levels and the water level station as well as differences (errors) caused by the inability of zoning from a stationary gauge to capture water level changes at the vessel. As previously mentioned some water level gauges are currently outputting water levels using an incorrect CRD to NAVD88 relationship. For areas where significant discrepancies between published CRD (used by the water level gauges) and interim CRD (used in the model) were present swath to swath agreement between survey lines was evaluated rather than computing a difference surface between the two datasets. Typically, survey line agreement was evaluated between overlap from lines acquired under varying stages of the tide cycle or flow conditions.

17 Comparisons between adjacent swaths showed differences of over 40 cm with zoning applied and virtually no discernable difference with GPS water levels applied (Figure 13). Figure 13. Discontinuities Resulting from Discrete Zoning Figure 14. Grid using zoned tides Figure 15. Grid using GPS water levels Tide zoning artifacts are conspicuous in the final products of a hydrographic survey. Figure 13 illustrates these discontinuities with a 40 cm difference between two lines in which discrete tidal zoning was applied. Figure 14 shows a small portion of a sun illuminated bathymetric grid where a single line of the survey is offset nearly 30 cm from the adjacent survey lines due to the use of discrete tidal zoning. Figure 15 shows the same survey area after application of GPS water levels using the Applanix POSPac Single Base solution.

18 Uncertainty Reduction The use of GPS water levels can drastically reduce the vertical total propagated uncertainty of surveyed depths by removing several vertical uncertainty components from total propagated uncertainty calculations. Under typical scenarios tidal uncertainty is the largest vertical uncertainty component of a hydrographic survey with published recommended values of up to 40 cm used for zoning uncertainty. By removing the need to use discrete zoning and making water level measurements directly at the survey vessel significant reductions in vertical uncertainty can be made. In addition, since GPS height measurements are being made in real-time on the survey platform there is no need to account for changes in vessel loading, draft, and settlement and squat using traditional processing methodology or accounting for these components when computing total propagated uncertainty. Table 3 lists published and project static uncertainty values when using discrete zoning, values that would have been used for the project if discrete zone was used, and the actual static uncertainty values used with GPS water levels. Method Tide Measured Table 3. Water Level Uncertainty Values Traditional m Ranges i 0.05m 0.03m 0.30m 0.20m Project 0.01m 0.30m 0.01m 0.0m 0.01m 0.30m Values Zoned Project 0.03m m Values GPS Water Levels Tide Zoned Discrete m Delta Draft Loading Draft Total Component Values reported at 1 sigma Conclusions GPS derived water levels provide previously unmeasured water levels at the survey platform and vastly improved vertical accuracy to hydrographic surveys over discrete tidal zoning. Since the phase center of the GPS antenna and the acoustic center of the multibeam sonar are at a fixed distance and move in unison with vessel loading, settlement and squat, vertical uncertainties in static draft and dynamic draft are eliminated. However, the prudent hydrographer should maintain a log of static draft observations and squat table such that conventional tides may be applied as a backup method or water level data from the survey can be used to verify or improve modeling of ellipsoid heights relative to chart datum and wind driven water level models used for forecasting water levels. Although some offset was observed in the application of SmartBase PPVRS when compared to RTK or Single Base applications, the offset is of unknown origin and may be datum related in the data set illustrated. There is a strong case for the implementation of a PPVRS solution both in

19 terms of logistics and cost. All of the methods tested provide better vertical accuracies than convention discrete tidal zoning and the application of changes in static and dynamic draft. Application of GPS water levels relative to a separation model to convert ellipsoid heights to chart datum can always be altered as better models become available. However, there are many areas where this relationship is well defined and should be implemented. Lack of a well defined separation model should not preclude logging of GPS carrier phase data. Field operations are the largest expense in updating nautical charts, or any other application of hydrographic data. Resurvey of an area to improve vertical accuracy is cost prohibitive once accurate models become available. At a relatively minor increase in cost, logging of GPS carrier phase data allows hydrographers to collect data today that could be used to improve vertical accuracies on future products derived from their surveys to support integrated mapping efforts (Survey Once, Use Many Times). In the not too distant future, this method would replace discrete tidal zoning and the necessity for installation, maintenance and processing of subordinate water level gauges in support of hydrography and decrease ping-to-chart time. Recommendations/ Future Steps There is a strong need to develop models of chart datum directly related to a reference ellipsoid as was done for the Columbia River project. These models should be compatible with NGS geoid models for integration into current hydrographic software such as Hypack and Caris HIPS. Current chart datum modeling efforts only perform transformations relative to orthometric heights. The application of a second model, currently Geoid 03 in the U.S., is required to transform ellipsoid heights to orthometric heights. An alternative would be hydrographic software providers to enable the input of multiple vertical transformation models. These models need to be of a consistent format (preferably following the NGS geoid model format) for ease of application. Uncertainty can still be inherent in models used to reduce GPS heights relative to the ellipsoid to chart datum. Future work will be required to calculate these model specific uncertainties to ensure that they are properly reported and that processing software allows for application of model uncertainty that may have geographic variability. The benefit of GPS derived water levels is that new models can easily be applied to data after processing has been completed. However, errors from discrete tidal zoning typically exceed ellipsoid modeling errors and can never be removed completely from bathymetric data as measurements were not observed at the survey platform. Full integration of NAD 83 base station coordinates for post processing into the PosPac PPVRS utility is required for effective operations. A less desirable alternative is an SBET transformation tool to allow for WGS 84 post processing and then transformation to NAD 83. i NOAA Field Procedures Manual 2008, Chapter 4 Appendices