1	National Spatial Reference System NORTH AMERICAN VERTICAL DATUM OF 1988 (NAVD 88) SEMINAR January 15, 2003 Catskill, New York Edward J. McKay
2	OUTLINE Vertical Datums Height Systems NAVD 88 Project NAVD 88 Implementation FEMA & NAVD 88 NAVD 88 Conversion Techniques
3	NATIONAL SPATIAL REFERENCE SYSTEM The National Spatial Reference System (NSRS) is the name given to all geodetic control contained in the National Geodetic Survey (NGS) Data Base. This includes: A, B, First, Second and Third-Order horizontal and vertical control, Geoid models such as GEOID 99, precise GPS orbits and Continuously Operating Reference Stations (CORS), observed by NGS as well as data submitted by other Federal, State, and local agencies, academic institutions and the private sector
4	VERTICAL DATUMS SEA LEVEL DATUM OF 1929 NATIONAL GEODETIC VERTICAL DATUM OF 1929 (As of July 2, 1973) NORTH AMERICAN VERTICAL DATUM OF 1988 (As of June 24, 1993)
5	COMPARISON OF VERTICAL DATUM ELEMENTS NGVD 29 NAVD 88 DATUM DEFINITION 26 TIDE GAUGES FATHER’S POINT/RIMOUSKI IN THE U.S. & CANADA QUEBEC, CANADA BENCH MARKS 100,000 450,000 LEVELING (Km) 102,724 1,001,500 GEOID FITTING Distorted to Fit MSL Gauges Best Continental Model
6	NORTH AMERICAN VERTICAL DATUM 88 WHAT IS A VERTICAL CONTROL NETWORK? An Interconnected System of Bench Marks Each Bench Mark Is Assigned A height Referenced To A Common Surface
7	NORTH AMERICA VERTICAL WHY DO WE NEED A VERTICAL CONTROL NETWORK? Reduces The Amount Of Future Leveling Required Enables Surveyors To Check Their New Leveling Provides Backups For Destroyed Or Disturbed Bench Marks Assists In Monitoring Changes In Local Areas Provides A Common Framework
8	HEIGHT SYSTEMS FIVE STEPS TO CREATING A VERTICAL CONTROL NETWORK Recon level line and set new bench marks Observe height differences between bench marks Correct observations for known systematic effects Minimize discrepancies in the results obtained by leveling along different routes between the same two points Define the surface datum to which heights may be referred
9
10
11
12	Heights Based on Geopotential Number Normal Height (NGVD29) H* = C / g g = Average normal gravity along plumb line Dynamic Height (IGLD55,85) H^dyn= C / g₄₅ g₄₅= Normal gravity at 45° latitude Orthometric Height H = C / g g = Average gravity along the plumb line Helmert Height (NAVD 88) H = C / (g + 0.0424 H) g = Surface gravity measurement (mgals)
13	The Geoid The geoid is the equipotential surface of the earth’s attraction and rotation which, on the average, coincides with mean sea level in the open ocean.
14	Execution of Surveys; Sources of Error Errors may be characterized as random, systematic, or blunders Random error represents the effect of unpredictable variations in the instruments, the environment, and the observing procedures employed Systematic error represents the effect of consistent inaccuracies in the instruments or in the observing procedures Blunders or mistakes are typically caused by carelessness and are detected by systematic checking of all work through observational procedures and methodology designed to allow their detection and elimination
15	Geodetic Control Network of Monumented Points Precisely Measured in Accordance with Standard Procedures Meet Accuracy Specifications Adjusted to Tie Together Documented for Multiple Use
16	HEIGHT SYSTEMS GEOPOTENTIAL NUMBERS Although geopotential numbers are useful for the adjustment of vertical networks, for many purposes “true” orthometric heights above a physically defined reference surface are still necessary A geopotential number can be converted to a “true” orthometric height by dividing the geopotential number by the mean value of gravity along the plumb line between the point and the reference surface H = C/gm Since the mean value of gravity cannot be directly measured (because the reference surface lies within the Earth beneath the point), a model must be used to derive the value as a point, and other variables
17	HEIGHT SYSTEMS GEOPOTENTIAL NUMBERS The geopotential number of a point is a measure of the difference in potential from the reference surface to the equipotential surface passing through the point The geopotential number is numerically equivalent to the work required to raise a mass of 1 Kg against gravity (g) through the orthometric height (H) to the point: H Geopotential number (C) = g dH 0 The difference in height (dh) measured during each setup of leveling can be converted to a difference in potential by multiplying dh by the mean value of gravity (gm) for the setup Geopotential difference = gm*dh
18	HEIGHT SYSTEMS GEOPOTENTIAL NUMBERS The geopotential number C is measured in geopotential units (gpu) 1 gpu = 1 Kgal meter = 1000 gal meter g = 0.98 Kgal \ c @ 0.98 H (Reference: “Physical Geodesy” by Heiskanen and Moritz)
19	HEIGHT SYSTEMS SEA LEVEL HEIGHTS Heights measured above local mean sea level The National Tidal Datum epoch is a particular 19 - year series over which the phases (such as mean lower low water) are determined. Encompasses all significant tidal periods Including the 18.6 - year period for the regression of the Moon’s nodes Averages out practically all of the meteorological, hydrological, and oceanographic variability Leveling is used to determine the relationship between bench marks and tidal gages
20	HEIGHT SYSTEMS DATUMS Any surface defined as the reference surface from which heights are measured, can be called a datum International Great Lakes Datum (IGLD)1955 Defined by one height (Father Point) Water - level transfers used to connect leveling across the Great Lakes Dynamic heights H - C/GÎ; Ge - 980.6294 gals (Normal gravity at 45 degrees latitude as defined in 1955)
21	HEIGHT SYSTEMS DATUMS National Geodetic Vertical Datum of 1929 (NGVD 29) Defined by heights of 26 tidal stations in the U.S. and Canada Tide gages were connected to the vertical network by leveling from tide gage staffs to bench marks Water - Level transfers used to connect leveling across the Great Lakes Normal orthometric heights H - C/Ga; Ga - Normal gravity based on formula
22	HEIGHT SYSTEMS DATUMS North American Vertical Datum of 1988 (NAVD 88) Defined by one height (Father Point/Rimouski) Water-level transfers used to connect leveling across the Great Lakes Geopotential Numbers Helmert orthometric heights Hhel - C/Ga; Ga = Mean value of gravity along the plumb line between the geoid and surface, estimated using Helmert’s reduction, I.e., g + 0.0424xHo. g = gravity at the surface in gals Ho = approximate height in kilometers
23	HEIGHT SYSTEMS DATUMS INTERNATIONAL Great Lakes Datum (IGLD) 1985 Same as NAVD 88, except published in Dynamic Heights Dynamic Heights Hdym = C/Go; Go = 980.6199 gals (Normal gravity at 45 degrees latitude as defined in 1985)
24
25	EXECUTION OF SURVEYS SOURCES OF ERROR Errors may be characterized as random, systematic, or blunders Random error in leveling results represent the effect of unpredictable variations in the instruments, the environment, and the procedure of leveling Random error cannot be completely eliminated, although it can be kept small Therefore, it represents the “noise level,” a limit on the accuracy with which leveling may measure elevation differences
26	EXECUTION OF SURVEYS SOURCES OF ERROR Errors may be characterized as random, systematic, or blunders Systematic error represents the effect of consistent inaccuracies in the instruments or in the leveling procedures Systematic error may be small in a single measurement; it accumulates when measurements made under similar circumstances are totaled Therefore, it can result in a significant discrepancy in the height differences measured between two control points by different leveling systems and/or routes For leveling to provide accurate height differences, systematic error must be minimized, either by procedure or by applying corrections to the data
27	VERTICAL DATA REDUCTION COMPUTATIONS Systematic errors which cannot be sufficiently controlled by instrumentation or observational techniques are minimized by applying appropriate corrections to the observed data. (See Balazs and Young, 1982). NGS applies seven corrections Level Collimation Scale Imperfections Refraction Curvature Tidal Accelerations Gravity Field Magnetic Fields
28	EXECUTION OF SURVEYS SOURCES OF ERROR Blunders The sources of error in leveling can be classified into three groups: Those affecting the line of sight Those affecting the heights computed Blunders
29	Error Sources Associated With Differential Leveling Error Source Typical Size of Error in mm Per 1 km Section Blunders: Forward pin or plate movement between setups……… 10.0 One rod unit or larger error in reading the rod……….. 5.0 Systematic Errors: Rod verticality error …………………………………... 1.0 Rod scale error…………………………………………. 2.0 Thermal expansion of Invar rod ………………………. 0.2 Rod index error ……………………………………….. 1.0 Movement of tripod during setup (if set up correctly)… 0.2 Gradual movement of turning points: during setups ……………………………………………… 0.6 between setups …………………………………………… 0.6
30	Error Sources Associated With Differential Leveling Error Source Typical Size of Error in mm Per 1 km Section Systematic Errors Continued: Collimation ……….. ………………………………………. 2.4 Under and over compensation ………………………….. 0.4 Refraction ………………………………………………….. 2.0 Refraction change during setup …………………………. 0.6 Diurnal Earth tides ………………………………………… 0.1 Earth’s magnetic field …………………………………….. 1.0 NI 002 parallax ……………………………………………. 0.6
31	Error Sources Associated With Differential Leveling Error Source Typical Size of Error in mm Per 1 km Section “Quasi” Random Errors: Scintillation, short-period …………………………………. 1.0 Scintillation, long-period ……...………………………….. 5.0 Pointing error (experienced observer)..…...……………….. 0.4 Rod error in individual graduations ..……………………. 0.1 *** NOTE: Assumes 50 meter sight lengths and 10 setups per 1 kilometer section.
32	NAVD 88 DATUM DEFINITION AND RESULTS
33
34	NAVD 88 PROGRAM DEFINITION NAVD 88 is a program which combined 1,300,00 kilometers of leveling surveys held in the NGS National Spatial Reference System (NSRS) data base, into a single least squares adjustment to provide users with improved heights for over 500,000 vertical control points distributed throughout the United States, on a common datum.
35
36
37	PRESENT NETWORK FOR NAVD 88 ORIGINAL LEVELING 700,000 KM REPEAT LEVELING 200,000 KM NEW BNA LEVELING 81,500 KM NEW OUTSIDE LEVELING 20,000 KM TOTAL FOR NAVD 88 1,001,500 KM (620,000 MILES)
38
39
40	NEW YORK VERTICAL NETWORK NGVD 29 bench marks . . . . . . . . . 12,927 NAVD 88 bench marks . . . . . . . . . 14,529 (INCLUDES “POSTED” DATA) POSTED bench marks . . . . . . . . . . 609 Bench marks without NAVD 88 heights . . . . . . . . . . 599* *Includes TBM’s, some RESETS, and new marks on lines not included in NAVD 88 general adjustment
41
42
43
44	NORTH AMERICALN VERTICAL DATUM OF 1988 (NAVD 88) THE U.S. PORTION OF THE PROJECT INCLUDED THE REMONUMENTATION AND REOBSERVATION OF AN 80,000 KILOMETER SUBSET OF THE VERTICAL CONTROL PORTION OF THE NATIONAL SPATIAL REFERENCE SYSTEM. A MINIMUM-CONSTRAINT LEAST SQUARES ADJUSTMENT OF LEVELING DATA INVOLVING 709,000 MARKS WAS PERFORMED.
45	NORTH AMERICAN VERTICAL DATUM OF 1988 (NAVD 88) IN ORDER TO MINIMIZE THE EFFECTS ON USGS NATIONAL MAPPING PRODUCTS (NMPs), AS REQUESTED BY USERS, NGS SELECTED THE NEW INTERNATIONAL GREAT LAKES DATUM OF 1985 (IGLD 85) LOCAL MEAN SEA LEVEL HEIGHT VALUE AT MINIMUM-CONSTRAINT DATUM POINT FOR NAVD 88. THE DATUM POINT IS LOCATED AT THE MOUTH OF THE ST. LAWRENCE RIVER IN QUEBEC, CANADA. USING FATHER POINT/RIMOUSKI AS THE DATUM POINT FOR BOTH IGLD 85 AND NAVD 88 MINIMIZES THE IMPACT ON NMPs, AND ALLOWS NAVD 88 TO REPLACE BOTH NGVD 29 AND IGLD 55.
46	NORTH AMERICAN VERTICAL DATUM OF 1988 (NAVD 88) FISCAL THE GENERAL ADJUSTMENT DID NOT INCLUDE APPROXIMATELY 25 PERCENT OF THE VERTICAL CONTROL NETWORK. BENCH MARKS IN “STABLE” AREAS WHICH WERE REMOVED FROM THE ADJUSTMENT (DENOTED AS “POSTED”) BECAUSE OLDER DATA DID NOT FIT WITH THE LATEST DATA. THIS DATA WAS INCORPORATED INTO THE NAVD 88 DURING YEARS 1992-1993.
47	NORTH AMERICAN VERTICAL DATUM OF 1988 (NAVD 88) NAVD 88 DOES NOT CONTAIN USGS, COE, OR STATE DOT THIRD-ORDER LEVELING DATA. USGS PERSONNEL HAVE PERFORMED PILOT STUDIES TO DETERMINE HOW TO BEST INCORPORATE THEIR THIRD-ORDER DATA INTO NAVD 88 (ABOUT A 5-10 YEAR PROGRAM)
48	NAVD 88 DATUM DEFINITION Vertical datum based upon an equipotential surface Minimally constrained adjustment held fixed at one point, Father Point/Riouski (Point-au-Pere) 1.3 million kms of leveling data used Heights of 585,000 permanent bench marks estimated. Both orthometric heights and geopotential numbers have been published
49	NAVD 88 GENERAL ADJUSTMENT COMPLETION DATE OF JUNE 1991: WHAT DOES THIS REALLY MEAN? The general adjustment of NAVD 88 was completed in June 1991. This means that bench marks included in the NAVD 88 Helmert blocking phase (approximately 80 percent of the total) have final adjusted heights available. Bench marks in “stable” areas which were removed from the adjustment (denoted as “POSTed”) because older data did not fit with the latest data was incorporated into NAVD 88 during fiscal years 1992-1993.
50	NAVD 88 GENERAL ADJUSTMENT COMPLETION DATE OF JUNE 1991: WHAT DOES THIS REALLY MEAN? Bench marks “POSTed” in large crustal movement areas, e.g., southern California, Phoenix, Arizona, Houston, Texas, and southern Louisiana was published as special reports after the final adjustment was completed. This is an on-going, long-term task which was started in January 1992. It is important to note that some bench marks in crustal movement areas, i.e., bench marks which were included in the NAVD 88 Helmert blocking phase, is available. The heights of these bench marks will be based on the latest available data, but still may be influenced by crustal movement effects.
51	Most surveying applications should not be significantly affected because the changes in relative height between adjacent bench marks should be less than 1 cm. As stated above, the absolute height values will change much more, but this should not be a major concern to the surveyor. The greatest problem the surveyor will have is ensuring that all height values of bench marks in the project area are referenced to the same vertical datum, preferably NAVD 88 and labeled correctly (metadata). Other agencies’ bench marks, e.g., COE, FL Department of Transportation, FL Department of Environmental Protection, and USGS, were incorporated into NAVD 88 by NGS as these agencies provided, and still do, their data in computer-readable form. However, the leveling data associated with over 500,000 third-order bench marks established by USGS have not been placed in computer-readable form and do not have NAVD 88 heights. In addition, COE has established hundreds of thousands of bench marks across the nation which do not have NAVD 88 heights.
52	IMPACT OF NAVD 88 Data Bases containing heights referenced to NGVD 29 will have to be updated to NAVD 88 Depending upon the accuracy required, in many areas a Bias Factor could be used for bench marks not included in the readjustment In “Moving” areas a Bias Factor probably will not be sufficient for most applications
53	IMPACT OF NAVD 88 Published Heights of Bench Marks Have Changed Published height values has shifted as much as 5 decimeters In “Stable” areas, Relative height changes between adjacent bench marks should only be millimeters In “Moving” areas, Relative height changes have been dependent upon the reasons for the movements.
54	IMPACT OF NAVD 88 Maps depicting NGVD 29 Heights will have to be modified for NAVD 88 Heights In many areas a Single Bias Factor, Describing the Difference between NGVD 29 and NAVD 88, could be used for most Mapping Applications In “Moving” areas, maps depicting the rates of movements will have to be compiled
55	REASONS TO CONVERT PRODUCTS TO NAVD 88 Surveys between bench marks will often close better NAVD 88 has provided a better reference to compute GPS-Derived Orthometric Heights 40,000 Additional bench marks of First-Order accuracy is available on NAVD 88 Data and NAVD 88 adjusted height values is readily available and accessible in a convenient format from NGS’s web site: http://www.ngs.noaa.gov Federal Surveying and Mapping agencies will stop publishing on NGVD 29 and will publish only on NAVD 88 Surveys performed for the Federal Government requires the use of NAVD 88
56	REASONS TO CONVERT PRODUCTS TO NAVD 88 THE AMERICAN CONGRESS ON SURVEYING AND MAPPING (ACSM) AND THE FEDERAL GEODETIC CONTROL SUBCOMMITTEE (FGCS) RECOMMEND NAVD 88. National Geodetic Survey no longer adjust to NGVD 29
57	BENEFITS OF NAVD 88 Improved set of heights on a single vertical datum for North America Improved FGCS Leveling procedures with higher production and lower error rates All NGS National Spatial Reference System data is validated in a single data base, with easy access by users for crustal motion studies, adjustments, latest official heights, and descriptions Removal of height discrepancies caused by inconsistent adjustment constraints
58	BENEFITS OF NAVD 88 (CONTINUED) Detection and Removal of height errors due to blunders Minimization of effects of systematic errors in leveling data Replacement of both NGVD 29 and IGLD 55 with a single datum Remonumentation and incorporation of 80,000 km of new leveling data not previously adjusted to NGVD 29 Orthometric Heights compatible with GPS-Derived Orthometric Heights computed using the High-Resolution Geoid Model called Geoid99
59
60	NAVD 88 IMPLEMENTATION
61	NAVD 88 IMPLEMENTATION Published and distributed NAVD 88 height values Processed and distributed height values for “POSTed” data FGCS Vertical Workgroup input from ACSM Ad Hoc Committee USGS third-order vertical data FEMA/National Flood Insurance program
62	FGCS VERTICAL WORK GROUP MEMBERS: National Geodetic Survey (Chair) U.S. Geological Survey Federal Highway Administration International Boundary Commission Bureau of Land Management U.S. Army Corps of Engineers U.S. Forest Service Federal Emergency Management Agency
63	ACSM AD HOC COMMITTEE GEOGRAPHIC MAKEUP East Coast (Florida to Massachusetts) Gulf Coast Interior Southern States Great Lakes area Plains and Mountain States Pacific Coast (California to Washington)
64	ACSM AD HOC COMMITTEE DISCIPLINE MAKEUP Land Surveyors Geodetic Surveyors Mappers ACSM Private Members ACSM Government Members
65	FEMA’S RESPONSE TO NAVD 88
66	FEMA’S Response to NAVD 88 Local Mean Sea Level (LMSL) determined at individual tide gages Sea Level Datum (SLD) of 1929 constrained at 26 tide gages in the U.S. and Canada National Geodetic Vertical Datum of 1929 (NGVD 29) renamed from SLD of 1929 to avoid confusion with LMSL North American Vertical Datum of 1988 (NAVD 88) constrained only at Pointe au Pere gage on St. Lawrence River
67	FEMA’s Response to NAVD 88 FEMA mapped and prepared Flood Insurance Studies (FISs) for thousands of communities with flood elevations vertical reference is the datum as defined by NGS FISs contain flood profiles Flood Insurance Rate Maps (FIRMs) contain flood elevations and Elevation Reference Marks (ERMs) Letters of Map Amendment and Revision (LOMAs and LOMRs) are issued based on elevation comparisons
68	FEMA’s Response to NAVD 88 FEMA Users Include: Banks and mortgage institutions (lenders) Flood insurance agents Surveyors, engineers, architects, and planners Community floodplain, planning, and zoning officials FEMA Contractors Include: Federal and State water resources agencies Regional water resources commissions Private architectural and engineering firms
69	FEMA’s Response to NAVD 88 Lenders initiate flood insurance purchase requirement based on FIRMs Surveyors provide Elevation Certificates for flood insurance agents and lenders Community officials enforce floodplain management regulations, which are based on FIS and FIRM Federal contractors must know how and when to implement conversion
70	FEMA’s Response to NAVD 88 Responsibility of Map Users ensure use of datum consistent with FIS and FIRM Responsibility of FEMA Contractors adherence to FEMA guidelines for conversion documentation of datum used in FIS and FIRM ensure datum consistency throughout FIS and FIRM
71	FEMA’s Response to NAVD 88 How Will FEMA Accomplish Conversion? Educate staff* Educate contractors* Educate users* Close coordination with NGS *FEMA has published two documents: Appendix 6, Conversion to the North American Vertical Datum 1988 Converting the NFIP to the NAVD 88
72	FEMA’s Response to NAVD 88 FEMA’s Original Plan New Studies - FY ‘93 FISs (scope of work April ‘92) Map actions FY’93 as practicable FEMA’s Current Proposal Update Appendix 6, Conversion to the NAVD 88 Refine strategy for an orderly transition of FISs and FIRMs to NAVD 88 Gradually convert based on opportunities to republish FISs and FIRMs for other reasons. Ultimate goal is to convert all FISs to NAVD 88
73
74
75
76
77
78
79
80
81
82
83
84
85
86	NGS’ RESPONSIBILITIES Performed procedures to officially replace NGVD 29 with NAVD 88 Compiled documentation to brief Congress and State officials on NAVD 88 impacts and benefits to minimize problems with uniformed users Provided documentation and publication of NAVD 88 final results
87	NGS’ RESPONSIBILITIES Estimated conversion (bias) shifts between NGVD 29 and NAVD 88 Analyzed bias shift computations to determine where other data, e.g., COE and/or USGS data, may be required (in computer-readable form) to improve the estimate of the bias factor Analyzed the vertical control network to determine “local” areas where height changes are due to crustal movement
88	NGS’ RESPONSIBILITIES Analyzed the vertical control network to separate bias shifts into components: changes due to datum definition, crustal movement, improved corrections applied to leveling data to account for systematic errors, and removal of adjustment distortions in NGVD 29 Incorporate other data, e.g., COE and/or USGS data, into NAVD 88 (data must be in computer-readable form) Educate NAVD 88 users
89	NAVD 88 USER’S RESPONSIBILITIES Provide Kinds Of Data, Reports, Routines, and Training Required To Implement NAVD 88 Relay (In A Timely Manner) To NGS Problems with Implementation Of NAVD 88
90	NAVD 88 CONVERSION TECHNIQUES
91	BASIC CONVERSION TECHNIQUES Estimation of bench mark heights by incorporating the original leveling data into NAVD 88 using least squares adjustment techniques A rigorous transformation of bench mark heights for a particular project using datum conversion correctors estimated from the project’s original adjustment constraints and their differences between NAVD 88 and NGVD 29 A simplified transformation of bench mark heights using an average bias shift for the area (VERTCON)
92	CONVERSION TECHNIQUES (CONTINUED) Technique number 1 is the most rigorous technique because the bench mark heights will retain their original relative accuracy. These heights will be useful to all users. In addition, NGS will adjust and publish the results if the data are submitted to NGS in computer-readable form. Technique number 2 may meet many users’ requirements, but depending upon the accuracy requirements and the complexity of the users’s leveling network, may prove to NGS to process. Technique number 3 should be the easiest method to implement, but in general is only sufficiently accurate enough to meet mapping requirements.
93	CONVERSION TECHNIQUES The use of GPS data and a high-resolution geoid model (Geoid99) to estimate accurate GPS-derived orthometric heights will be directly associated with the implementation of NAVD 88. It is important that users initiate a program to convert their products to NAVD 88. The conversion process is not a difficult one, but will require time and resources. There will be several different conversion techniques available. The technique used will depend on the accuracy requirement of the user, I.e., procedures developed for conversion of less accurate GIS/LIS products will be different than procedures developed for conversion of USGS NGVD 29 published height values.
94
95	Review Vertical Datums Height Systems NAVD 88 Project NAVD 88 Implementation FEMA & NAVD 88 NAVD 88 Conversion Techniques
96