Paper for a communication at the U.S. Hydro 2003

Transcription

1 Paper for a communication at the U.S. Hydro 2003 A Student Initiated Hydrographic Survey in a Riverine Environment, The Pearl River 2002 Field Project Experience David Dodd, Stephan Howden, David Wells, Denis Wiesenburg and the Hydrographic Science Masters Degree Graduating Class of 2002 Hydrographic Science Program Department of Marine Science The University of Southern Mississippi Abstract The University of Southern Mississippi s (USM) Hydrographic Science Program (HSP) has recently graduated its third class. The final task for students graduating from the program is the completion of a six-week field project, which entails a complete hydrographic survey for charting purposes. Students take full control of this project where they are responsible for all aspects of this survey from site selection, initial data mining, and specification development, through field data collection, processing and analysis, to a final report and field chart. The final products are sent to the National Ocean Service (NOS) for critical analysis and assessment. If deemed worthy, the information is used to update existing charts. Presently, data from the 2001 field program in St. Louis Bay have been accepted by NOS and data from the 2002 field program on the Pearl River are under review. The Hydrographic Science program s 2002 field project was performed on the Pearl River near the Stennis Space Center in Mississippi. Existing spatial information was used for project planning. GPS and traditional survey techniques were used to establish horizontal and vertical control. Water levels were monitored above and below the survey area. Depth observations and side-scan data were collected using a small skiff as well as a multibeam data collection platform. Analysis of the data indicated that IHO S-44, 4 th edition, first order standards were met. Introduction A common complaint regarding many academic programs is that there is a disconnection between theory and real-world application. The Hydrographic Science Degree program at the University of Southern Mississippi goes to great lengths to bridge that disconnection. At the completion of all classroom course work, students must conduct a complete hydrographic survey suitable for charting purposes. In all regular courses, theory is backed up by practical exercises. Many of these practical field exercises focus on the hydrographic survey field project. The University of Southern Mississippi s (USM) hydrographic science graduate degree program started as a joint venture between USM s Department of Marine Science (DMS) and the Naval Oceanographic Office (NAVO). USM contributes the academic environment including professors and instructors experienced in the field of hydrography, as well as research and educational facilities and equipment. NAVO s contributions include: fieldwork support in the form of equipment loans and a multibeam data collection platform; and experienced personnel as

2 instructors, tutors, guest lecturers and students. The program received Category A certification from the FIG/IHO International Advisory Board on Standards of Competence for Hydrographic Surveyors in April of 2000 and graduated its first class in August To date, the program has had three graduating classes, awarding a total of 34 master s degrees in hydrographic science. During the course of the four years that the program has been operational, it has progressed from 100% of enrollment through NAVO (both uniform and civilian), to a very diverse mixture of students from NAVO, NIMA, USACE, NRL and foreign navies and hydrographic offices. A new addition to the Stennis Space Center facilities of USM is the Hydrographic Science Research Center (HSRC). The primary goal of HSRC is to address NAVO operational hydrographic research needs. The HSRC will compliment the HSP and will lead to excellent opportunities for graduate students who wish to pursue research and acquire a thesis based M.S. in Hydrographic Science. Most of the students in the program follow the accelerated one-year track. This includes two semesters of theory and application followed by one semester of application, combining for about 50 continuous weeks of study. Some students have opted to complete the program in two years and conduct research with the HSRC. The degree awarded to graduates is a courseintensive non-thesis Masters of Hydrographic Science. The final semester of this program is essential for rounding out the educational experience. The Hydrographic Science Field Project consumes the last six weeks of the program. Although the actual project is conducted at the end of the program, many of the practical exercises completed during the year for other courses are directed towards the area where this project will be executed. This includes preliminary reconnaissance surveys. The project is comprised of a complete nautical charting hydrographic survey including: Data mining for all existing spatially-related information for the project area Planning, preparation and the development of Hydrographic Survey Specifications for the project area Data collection (Multibeam, singlebeam, side-scan sonar, tides, sound velocity, shoreline, navigation aids and all ancillary data) Data processing and analysis Quality assurance checks to determine if IHO S-44 specifications have been met Final field sheet and field chart production Completion of a comprehensive report of Survey (ROS) Presentation of the project and results to the examination committee The final charts and report of survey are submitted to the National Ocean Service (NOS) for evaluation and assessment. If deemed acceptable, the information is used to update exiting nautical charts. This NOS submission process has been implemented for the 2001 and 2002 projects. Some of the data from the 2001 survey will be used in the next update of the affected chart and the 2002 field project data is currently under NOS review [Martin (2002)]. Students take complete control of the field project. They are responsible for all aspects of the survey from initial planning to final product, with very little input from faculty, although faculty guidance is always available. Once the project is completed they present the results of their 2

3 experience to the Hydrographic Program faculty. This paper is a summary of the field project conducted by the graduating class of Most of the material for this paper was taken from the report of survey and final presentations. The 2002 graduating class includes: Name Affiliation Name Affiliation Charles Baptiste NAVO Wesley Lewis NAVO Michael Bendzlowicz NAVO James McGaughran NIMA Glen Boyer NAVO Susan Moffett NAVO David Brazier NAVO Sheldon Powe NAVO Glen Carson NAVO Frank Schenk Navy Rich Delgado Navy Karim Taga Tunisian Navy Kim Jones NAVO Eddie Wiggins USACE The 2002 Field Project The 2002 field project was conducted on the East Pearl River in Mississippi, from the Highway 90 Bridge north to the I-10 Bridge in June and July of 2002 (see Figure 1). This region was selected mainly because of its proximity to the Stennis Space Center, where the DMS is located. It also presented some unique challenges in the areas of tide (river), wildlife (snakes, alligators, mosquitoes) and navigation hazards (sunken logs, barges and debris). The project area covers about 10 km of the river. There is a channel that ranges in depth from about 6 m to about 24 m. The width of the river varies between 100 and 200 meters. I-10 Stennis Space Center Mississippi Pearl River HWY 90 New Orleans Gulf of Mexico Figure 1: Pearl River survey site 3

4 The resources available for the field project were extensive. In addition to USM's resources a very strong working relationship with the NAVO provided USM with access to more and varied hydrographic surveying tools. The resources available to the HSP through USM and NAVO include; vessels, multibeam and singlebeam echosounders, side scan-sonars (SSS), sound velocity profilers, GPS receivers (DGPS and RTKGPS), tide gauges, computers, plotters and the latest high performance hydrographic software (for data collection, processing, analysis and visualization). Two vessels were used in the field project, a USM skiff and a NAVO hydrographic survey launch (HSL). Through a working agreement with NAVO the HSP has access to the HSL with a fully integrated multibeam system. In addition to the multibeam, the HSL also has a singlebeam sounder system, attitude sensor, wide area differential GPS, CTD probe and a side-scan sonar (SSS). It is 46 long and has a draft of about 5 (see Figure 2). Figure 2: NAVO Bertram, Training HSL The USM skiff is a shallow draft craft that is used to simulate a vessel of opportunity. Students build their own singlebeam and side-scan sonar data collection platform, without permanently modify the vessel, or drilling holes. It is a 21 flat bottom skiff with a draft of about 1. The following USM resources are used on this vessel (see Figure 3): Knudsen 320MP dual frequency echosounder, Trimble DSM212 DGPS receiver (with a Coast Guard Beacon receiver). Marine Sonic dual frequency SSS. Hypack MAX integrated navigation system Honda 3Kw generator. In addition to the equipment used directly on the survey vessels, other USM equipment used includes: Two Sutron tide gauges (both with bubbler and pressure sensors). OTT float and encoder tide gauge. Static and RTK GPS receivers (Trimble 750, Ashtech Z-Extremes, Novatel OEM 4/DL) Lieca Total Station. 4

5 Lieca Digital Levels. Among the USM software resources used are: CARIS GIS, HOM, DOM, HIPS and SIPS. HYPACK MAX. Area Based Editor Fledermaus. Figure 3: Skiff Equipment Project Preparation As previously mentioned, project preparation began during the fall and spring terms, with exercises geared towards the final project. These exercises included installing tide gauges, geodetic control, and data mining. Two tide gauges were established for this project. One at the Baxter Boat Camp, near the southern end of the project and the other at the I-10 Bridge, near the northern extent. Both of these locations were sites for temporary tide gauges installed in the late 1970s, from which tidal datums (Mean Lower Low Water - MLLW) were determined. Tidal benchmarks established to reference these datums were recovered and used to reference the new gauge installations. As a result, existing tidal datums were recovered for use in the project. 5

6 The tidal benchmarks for the Baxter site had been tied into the local vertical control (NAVD 88), however, the I-10 Bridge benchmarks had not. As part of this project both sites were tied together, and to the primary horizontal control (USM 1) using static GPS. The NAVD 88 datum from the Baxter site was transferred to the I-10 site using GPS observations and the GEOID 99 geoid/ellipsoid undulation model. The difference in orthometric height (NAVD 88) between MLLW of the two sites was found to be 25 cm [HYD ROS (2002)]. The Baxter tide gauge was established during the HSP Tides and Water Levels course in March of It consisted of a digital recorder with a bubbler gauge and a pressure transducer. The I- 10 tide gauge was established in June of It consisted of a float and shaft encoder. One month of simultaneous observations were used to determine the tidal regime. A tidal range of about 0.5 m was observed at both stations. Tide exhibited a distinctly diurnal trend. There was a phase difference of about 48 minutes between stations and an amplitude difference of about 10 cm. To accommodate for these differences, four tide zones were established [HYD ROS (2002)]. During the HSP Kinematic Positioning course the students started the process of establishing an NGS order 1 GPS control point (USM 1) near the USM HSP facilities. This point was accepted by the NGS and is in the Blue Book of geodetic control points (PID DE6235). It was used as the primary horizontal GPS control for the survey. The NAVO HSL was positioned with DGPS using the Fugro SeaStar Wide Area Network for differential corrections. The USM skiff was also positioned with DGPS using the Coast Guard Beacon Service for differential corrections. As a result, local horizontal control was not needed for vessel positioning. The USM 1 control point was used to observe the positioning accuracy of the Trimble DSM (using the Coast Guard Beacon Service). As a confidence check data were collected with the DSM mounted over the USM 1 reference geodetic point on June 5 and July 11, The GGA string data were collected and the standard deviation computed using SA Watch, version The actual accuracy (at 95%) was determined to be 0.7 m. Most of the data mining for this project area was conducted during the HSP Nautical Cartography and GIS course. High-resolution (1 meter) aerial photographic imagery in the form of Digital Ortho Quarter Quads (DOQQ) for the area were obtained from the Mississippi Automated Resource Information System (MARIS) website. During the HSP Practical Hydrography course, which directly precedes the field project, the students wrote a set of Hydrographic Survey Specifications (HSS) for the planned survey. This document described the survey and survey area and specified how it should be conducted in order to meet desired standards (IHO Order 1). The HSS was written in accordance with NAVO standards [HYD HSS (2002)]. Calibration surveys were conducted for the USM Skiff singlebeam platform to measure dynamic draft and navigation latency. Dynamic draft was observed using the rod and level technique. 6

7 Navigation latency was observed by running a series of lines at different speeds in the same direction. A bar check was conducted at the beginning and end of every survey day. Calibration surveys for the NAVO HSL multibeam platform were conducted prior to the project and then again at the completion of data collection. The calibration survey was designed to measure navigation latency and transducer roll, pitch and azimuth orientation offsets. Execution The aerial photography imagery was verified by a shoreline survey with the USM skiff and by static GPS observations of prominent objects. The verification procedure showed agreement to within 10 meters and usually much less. The shoreline for the project was digitized directly from the imagery. Sound velocity observations were performed daily for both the singlebeam and Multibeam systems. Two systems were used to collect this data: a conductivity, temperature and pressure probe (CTD) and a velocimeter. Both systems showed similar form, but were offset by 2 m/s. This offset translated to an error of 1.5 cm for the outer beams of the multibeam sonar, and was not considered to be significant. The sound velocity during the project was around 1510 m/s. The NAVO Bertram Training HSL was used for surveying the main channel. It was used to collect both Multibeam and side-scan sonar data. It was in operation for 6 days, including transit and calibration exercises. The center of the channel was surveyed first and the remaining lines were run parallel to it. In many cases the Multibeam swath reached the shore. At least two hundred percent coverage was achieved. Due to restricted maneuverability, check-lines were run in a zigzag pattern, at 45 to the main lines. All of the Multibeam data was archived to 4 mm data tapes in the Generic Sensor Format (GSF). The USM skiff was used for collecting near-shore singlebeam data as well as side-scan data. It was also used for collecting bottom samples, stream gauging, shorelining, monitoring tide gauges and conducting shoal examinations. Near-shore lines were run perpendicular to the shore, with 20 meter spacing and overlapping the Multibeam data. The singlebeam data was logged and archived in the HYPACK raw file format (in ASCII text). The side-scan data was stored and archived in the Marine Sonic Tiff format. Two-hundred percent coverage was achieved with the SSS. Processing and Analysis A total of about 2 Gb of singlebeam and Multibeam data were collected along with about 300 Mb of side-scan sonar data from the Marine Sonics SSS. Eighteen hours of Multibeam data collection translated into 56 hours of processing. Twenty-five hours of single beam data collection translated into five hours of data processing. The Area Base Editor (ABE) was used for the initial Multibeam data cleaning. This editor was used to remove depth spikes. This data was then ingested into CARIS HIPS version 5.2 for attitude and position sensor editing and the application of tides, using the zones created as a result of the tidal analysis. The cleaned soundings were then exported to a CARIS Map, using 7

8 one-meter shoal bias binning, for quality control and field chart production. The singlebeam data was processed entirely in HIPS and exported to the same CARIS map. The multibeam data was also exported to Fledermaus for 3-D visualization and flythrough production. The CARIS Quality Control tool was used to compare the Multibeam check-lines with the main production lines. This tool was used to compare individual soundings from each beam of the check-lines with a Digital Terrain Model (DTM) created from the production lines. This tool computed percentage values for the number of soundings that fell within the desired IHO S44 criteria (Order 1). The results of this analysis indicated that 95% of the soundings passed IHO Order 1 criteria for beams 25 through 100. The outer beams pass rate was just slightly better than 90%. SSS data were collected with the NAVO HSL as well as with the USM Skiff. Due to data format issues is was not possible to process the HSL SSS data except on board the vessel. The Marine Sonic data collected with the skiff were processed with CARIS SIPS. The main purpose for collecting the SSS data was to identify and mark possible hazards for further investigation (see Figure 4). The Pearl River was once used for the transportation of logs, and the bottom is littered with debris (see Figure 5). Over two hundred targets were identified as possible hazards. These were recorded as targets in SIPS. Many of these targets were logs and vertical piles that did not show up in the Multibeam data as well (if at all) as in the side-scan data. Although the intension was to investigate these targets, time ran out and the students were unable to perform the necessary shoal examinations. Figure 4: Side-scan and Multibeam View of Sunken Barge. 8

9 5 m Figure 5: Side-scan Sonar Image of Logs Final Products The primary final product was a field chart (see Figure 6 and Figure 7). The NAVO field chart production specifications were used as the template for the field charts. Two charts were produced, each at a scale of 1:4000. The field charts were created in CARIS GIS 4.4 with the intention of building both paper and digital (S-57) chart products. The CARIS map layering and feature code settings were organized so that conversion to IHO S-57 (ENC) could be facilitated. Due to time constraints only the field charts were produced. However, the experts at CARIS managed to convert both field charts into preliminary S-57 products within a matter of hours, which was a testament to initial CARIS map organization and the abilities of CARIS personnel. The final step for any major project is the final report. Producing a pretty picture without the necessary information to back it up is just that a pretty picture. The final report of survey (ROS) format followed the NAVO guidelines. This report included all information necessary for the validation of the data collected, and was one of the primary sources of information for this paper. 9

10 Figure 6: North Section Field Chart 10

11 Figure 7: Field Chart Blowup Comments The primary purpose of the field project is to give students the opportunity to combine and apply hydrographic concepts, and it is an essential learning tool. However, it is more than just an academic exercise. The students approached the entire process with the intention of developing a quality product that could be of use to the NOS for updating existing nautical charts. Acrobat PDF versions of the paper charts and the ROS were delivered to the NOS for review. Acceptance of this hydrographic information by the NOS is the ultimate validation of the processes and techniques used to create the final product and ROS. One of the constraints to producing acceptable data is the time component. This course is only six weeks long, and the class graduates within a week of presenting the final results of their project. One of the most consistent comments presented in the final briefs is if we had more time. The following is a list of suggestions and lessons learned gleaned from the final briefs: Plan to use known target to calibrate the systems on the boat each day of the survey. Stage equipment and test before using. Bench test gages. All system functions should be learned before deployment. Research into existing control needs to start earlier. Significant time was invested in recovering NOS benchmarks. Tide gages need to be installed earlier. 11

12 Needed more research into existing tidal datum. Keep plenty of notes and make sure notes are properly kept each day. Obtain GeoDAS software to process side-scan data from the Bertram. Negotiate an agreement with NAVO to obtain software to process Bertram side-scan data in Bldg New stronger transducer mount as spare. Implement the breakaway transducer mount coupling for river use. RTK principals are understood. The challenge is getting the equipment to understand RTK principals. Bottom line more emphasis on equipment configuration. Process data at the end of every survey day! Prosecute hazardous targets promptly after identification. Outfit Carolina Skiff with gyro. Bertram/HSL survey system is not Black Box. Direct measurement of transducer offsets not practical, however, could do horizontal uncertainty testing with submerged object of known position. Independent verification of positional accuracy. More monitoring analysis of GPS solution quality daily analysis of files. More near real-time quality assurance of MBES data. Difficult in this survey without predicted tides. Shoreline from Imagery was invaluable. Accuracy assessment of rectification and heads-up digitization process should be done early in product development. Start writing ROS from the beginning, not at the end. References Martin, E. (2002). Personal communication on September 05, Head of Gulf Coast Nautical Charting Branch, National Ocean Service, Silver Springs, MD. HYD ROS (2002). ROS 2002 Pearl River. Unpublished report of survey for the Department of Marine Science, University of Southern Mississippi, Stennis Space Center, Mississippi. HYD HSS (2002). HSS 2002 Pearl River. Unpublished hydrographic Survey Specifications for the Department of Marine Science, University of Southern Mississippi, Stennis Space Center, Mississippi. 12