1、CECW-EP Technical Letter NO. 11 10-1-183 3515789 0814343 539 DEPARTMENT OF THE ARMY U.S. Army Corps of Engineers Washington, DC 203 14-1 O00 Engineering and Design USING DIFFERENTIAL GPS POSITIONING FOR ELEVATION DETERMINATON Distribution Restriction Statement ETL 11 10-1-183 1 April 1998 Approved f
2、or public release; distribution is unlimited. Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,- 3515789 0834344 475 CECW-EP Tech n ka I Letter NO. 1110-1-183 DEPARTMENT OF THE ARMY U. S. Army Corps of Engineers Washington, D.C. 20314-1 O00 1 April 199
3、8 Engineering and Design USING DIFFERENTIAL GPS POSITIONING FOR ELEVATION DETERMINATION 1. Purpose This engineer technical letter provides technical guidance for using Differential Global Positioning System (DGPS) to determine elevations of survey benchmarks for wide- area mapping and GIS database d
4、evelopment applications. Recommended procedural specifications for field DGPS observation sessions are included. 2. Applicability This letter is applicable to major subordinate commands, districts, laboratories, and field operating activities having responsibility for civil works, military construct
5、ion, or environmental restoration projects. These DGPS guidelines and specifications are intended for densifying vertical control over large project areas, such as an entire military installation or watershed basin mapping project. The DGPS methods outlined in this letter are generally not intended,
6、 nor would be cost- effective, for small projects or any type of construction lay out work where vertical grades or benchmarks require an accuracy better than 30 millimeters (mm). In such cases, conventional differential leveling methods should be used. 3. References a. FGCC, (1991), Standards and S
7、pecifications for Geodetic Control Networks, Silver Spring, Maryland. (FGCC is currently known as FGCS) b. Milbert, D.G. and Smith D.A. (1996). Converting GPS Height into NAVD88 Elevation with the Geoid96 Geoid Height Model. National Geodetic Survey, Silver Spring, Maryland. ETL 1110-1-183 e. U.S. A
8、rmy Corps of Engineers (1994), Deformation Monitoring and Control Surveying. Engineer Manual 1110-1-1004, U.S. Army Corps of Engineers, Washington, D.C. d. U.S. Army Corps of Engineers (1996), NAVSTAR Global Positioning System Surveying. Engineer Manual No. 11 10-1-1003, U.S. Army Corps of Engineers
9、, Washington, D.C. e. Zilkoski, D.B., DOnofrio, Joseph D., and Frankes, Stephen J. (1997) Guidelines for Establishing GPS-Derived Ellipsoid Heights (Standards: 2 cm and 5 cm), Version 4.1.1. Silver Spring, Maryland. NGS Unpublished Report. J: Zilkoski, D.B., Richards, J.H., and Young, G.M. (1992). S
10、pecial Report: Results of the General Adjustment of the North American Vertical Datum of 1988, Silver Spring, Maryland. 4. Distribution This information is approved for public release. Distribution is unlimited. 5. Discussion a. Global Positioning System (GPS) surveying produces a set of X-Y-Z coord
11、inates which can be transformed into geodetic latitude, longitude, and ellipsoidal height by using an reference ellipsoid to model the earth. In the US., most GPS ellipsoid heights are measured with respect to North American Datum of 1983 (NAD83) control values, which are based on the Geodetic Refer
12、ence System of 1980 (GRS80) ellipsoid. Published Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-3535789 0834345 30L = ETL 1110-1-183 1 Apr98 orthometric elevations on national vertical control benchmarks in the North American Vertical Datum of 1988
13、(NAVD88) height system are established with respect to the geoid, a model of the earth based on gravity measurements. A determination of a NAVD88 elevation using GPS measurements at a given point requires a transformation between ellipsoid and geoid based height systems. The conversion between the N
14、AD83 GPS ellipsoid and NAVD88 orthometric height is made using the geoidal undulation (also referred to as geoid height) value that represents the geoid-ellipsoid separation distance. b. DGPS may provide an efficient and cost-effective means of densiQing elevation data over large, extended project a
15、reas when compared to conventional differential leveling. Height measurement accuracy that meets most USACE mapping requirements can be successfully achieved from several different GPS surveying techniques. However, DGPS vertical elevation techniques may not be sufficiently accurate for construction
16、 control or may not be cost-effective for small project areas. c. GPS relative vertical positioning and calculated geoid height differences for the determination of NAVD88 orthometric heights may be used when an accuracy no better than 30 mm is required. This GPS height accuracy satisfies feature el
17、evation tolerances specified for most USACE engineering mapping activities. However, it may not be sufficiently accurate for hydraulic engineering studies or construction activities. Guidance for GPS survey accuracies and FOR THE DIRECTOR OF CIVIL WORKS Appendix A Determination of Elevations with GP
18、S Surveying Techniques procedures can be found in EM 1 1 1 O- 1 - 1003. d. Recent advances in geoid modeling have also led to more accurate conversions between NAD 83 GPS and NAVD 88 orthometric height systems. Accuracies of 30 mm or better have been obtained when converting ellipsoid heights from G
19、PS surveys, based on NAD 83 control, to NAVD 88 orthometric heights using the latest geoid model (GEOID96). The initial GPS survey data must be valid for the elevation transfer method to be effective. Guidance for GPS survey accuracies and procedures can be found in EM 1110-1-1003, NAVSTAR Global Po
20、sitioning System Surveying. e. Appendix A presents the basic methodology for using GPS to determine NAVD88 elevations. GPS positioning techniques, coordinate systems, and vertical datum concepts are introduced and discussed along with operational requirements and computational schemes used to obtain
21、 NAVD88 elevations from GPS coordinates. These operationai requirements are based on field test results conducted by U.S. Army Topographic Engineering Center (CETEC) and the National Geodetic Survey (NGS) using several different GPS surveying methods and comparing these results to conventional diffe
22、rential leveling networks. 6. Proponency and Technical Assistance The HQUSACE proponent for this technical letter is CEC W-EP. Technical assistance in performing GPS surveys may be obtained by contacting the U.S. Army Topographic Engineering Center, ATTN: CETEC-TD-G , 770 1 Telegraph Road, Alexandri
23、a, VA 223 15-3864, (703) 428-6767. STEVEN t& L. STOCKTON, P.E. Chief, Engineering Division Directorate of Civil Works 2 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-3515789 081434b 248 ETL 1 1 10-1 -1 83 I Apr98 Appendix A Determination of Elevati
24、ons with GPS Survey ng Techniques A-I. Global Positioning System Ellipsoid Heights u. Recent advances in GPS technology and the current fully operational status for the NAVSTAR GPS have made it possible to accurately measure ellipsoidal height differences from GPS satellites. GPS surveys report vert
25、ical positions in geodetic coordinates defined with respect to the World Geodetic System of 1984 (WGS84) reference ellipsoid, a geocentric, bi-axial, ellipsoid of revolution, that is symmetric about the equatorial axis and whose shape and size are defined by mathematical constants corresponding to t
26、he length of its equatorial and polar axes and selected to best approximate the geoid (currently within 100 meters globally). Although many different geodetic datums exist throughout the world, WGS84 is the reference system most frequently used with off-the-shelf GPS equipment. The ellipsoidal heigh
27、t value at a given point is based on the distance measured along the normal vector from the surface of the reference ellipsoid to the point. The realization and practical accuracy of WGS84 as a vertical reference frame for collecting elevation data depends on the actual ellipsoidal height values ass
28、igned to benchmarks or other physically defined control points. b. In the U.S., final positions from DGPS are established with respect to NAD83. Since NAD83 is based on the GRS80 ellipsoid, ellipsoid heights obtained from GPS surveying using NAD83 control are based on the GRS80 ellipsoid. These heig
29、hts are referred to as NAD83 GPS ellipsoidal heights. Unlike the WGS84 ellipsoid, the GRS8O ellipsoid is not exactly geocentric which can create problems (i.e., large errors) when converting NAD83 GPS ellipsoid heights to orthometric heights using some geoid models, A-2. Orthometric Heights and NAVD
30、88 Elevations u. The orthometric height of a point is the distance from the reference surface to the point, measured along the line perpendicular to every equipotential surface in between. A series of equipotential surfaces can be used to represent the gravity field. One of these surfaces, the geoid
31、, is specified as the referenced system from which orthometric heights are measured. The geoid itself is defined as a potential surface and natural variations in gravity induce a smooth, continuous, curvature to the plumb line and therefore physical equipotential surfaces which are normal to gravity
32、 do not remain geometrically parallel over a given vertical distance (Le. the plumb line is not quite parallel to the ellipsoidal normal). b. The NAVD88 datum is the product of a vertical adjustment of leveled height difference measurements made across North America. NAVD88 was constrained by holdin
33、g fixed the orthometric height of a single primary tidal benchmark at Fathers Point / Rimouski, Quebec, Canada and performing a minimally constrained general adjustment of U.S.-Canadian-Mexican leveling observations. The vertical reference surface is therefore defined by the surface on which the gra
34、vity values are equal to the control point value. NAVD88 elevations are published orthometric heights that represent the geometric distance from the geoid to the terrain measured along the plumb line. Orthometric height corrections were used to enforce consistency between geopotential based vertical
35、 coordinates and measured leveled differences. NAVD88 is the most compatible vertical reference frame available to relate GPS ellipsoidal heights to orthometric heights. Further information on the NAVD88 datum can be found in Zilkoski (1992) and EM 1110-1-1004, Deformation Monitoring and Control Sur
36、veying. A-3. Geoidal Heights a. Geoidal heights or (geoid height values) represent the geoid-ellipsoid separation distance measured along the ellipsoid normal and are obtained by taking the difference between ellipsoidal and orthometric height values (see Figure 1). Knowledge of the geoid height ena
37、bles the evaluation of vertical positions in either the geodetic (ellipsoid based) or the orthometric height A-I Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-3535789 0834347 384 ETL I1 10-1 -1 83 I Apr98 system through the following relation: h=H+
38、N where, h = ellipsoid height H = orthometric height N = geoid height, and by convention N being a positive height above the ellipsoid. at the node points of a regular grid (i.e.? a 2 x 2 grid spacing). Biquadratic interpolation procedures can be used within a grid cell boundary to approximate the g
39、eoid height at a given geodetic latitude and longitude. The NGS GEOID96 model for the United States indicates geoid heights range from a low of -5 1.6 meters in the Atlantic to a high of -7.2 meters in the Rocky Mountains. For more information on geoid modeling, see Milbert (1 996). H = ORTHOMETRIC
40、HEIGHT N = GEOID HEIGHT I h = ELLIPSOIDAL HEIGHT = Ah - AN + HA h=H+N 1 I Figure I. GeoidlEllipsoid Relationship b. Geoid height values at stations where either only h or H is known can be obtained from geoid models which are mathematical surfaces representing the shape of the Earths gravity field.
41、The geoid model is constructed from a truncated functional series approximation using a spherical harmonics expansion and an extensive set of globally available gravity data. The model is determined from the unique coefficients of the finite series representing the geoid surface. Its accuracy depend
42、s on the coverage and accuracy of the gravity measurements used as boundary conditions. Former geoid models produced for general use limit absolute accuracies for geoid heights to no less than 1 meter. More recent geoid models have shown a significant increase in absolute accuracy for geoid heights
43、to a few centimeters. c. In practice the shape of the geoid surface is estimated globally as a function of horizontal coordinates referenced to a common geocentric position. Specific geoid height values are extracted from the model surface A-4. Relative Vertical Positioning with GPS u. DGPS observat
44、ion sessions produce 3-D geodetic coordinate differences that establish the baseline between two given stations. Baseline solutions produce relative positioning results at a greater accuracy than can now be achieved from point positioning. The expected accuracy of such ellipsoidal height difference
45、measurements is based on several factors such as GPS receiver manufacture type, observation session duration, and the measured baseline distance, but it does not depend greatly on prior knowledge of the absolute vertical position of either occupied station. Dual frequency, carrier phase measurement
46、based GPS surveys are usually able to produce 3-D relative positioning accuracies under 30 mm at the 95% confidence level over baseline distances less than 20 km, depending on the type of GPS surveying method used. This situation exists mainly because GPS range biases are physically well correlated
47、over relatively short distances and tend to cancel out as a result of forming double differences for carrier phase data processing. In contrast, GPS absolute code positioning accuracy will contain the full effects of any GPS range measurement errors. The method explained below to obtain NAVD88 eleva
48、tions from satellite surveys is based on the relative vertical positioning capability of GPS. b. Geoidal height differences describe the change in vertical position of the geoid with respect to the ellipsoid between two stations. These relative geoidal heights can be more accurate than the modeled a
49、bsolute separation values within extended areas because the relative geoidal height accuracy is based on the continuous surface characteristics of the geoid model, where only small A-2 Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-W 3515789 0834348 O30 W ETL 11 10-1 -1 83 1 Apr 98 deviations between closely spaced points would be expected. The regional trend or slope of the geoid at a given point will not be highly sensitive to lo
