Gravity and Geoid Modelling for Improved Heights

Texas Height Modernization Workshop Houston, Texas August 28, 2015

Denis Riordan, PSM NOAA, National Geodetic Survey [email protected] Presentation Outline

1. - Introductions. 2. - Mission and Vision of the National Geodetic Survey. 3. - NAVD88 – Still good enough? 4. - GEOIDS – types and definitions. 5. - GRAV-D and what it’s about. 6. - NGS Gravity Survey Plan. 7. - New U.S. Datums for 2022. 8. - Questions. U.S. Department of Commerce National Oceanic & Atmospheric Administration National Geodetic Survey

Mission: To define, maintain & provide access to the National Spatial Reference System (NSRS) to meet our Nation’s economic, social & environmental needs

National Spatial Reference System

* Latitude * Scale * Longitude * Gravity * Height * Orientation

& their variations in time NGS MISSION - The NSRS

Define - National Coordinate Sys. (NSRS)

Maintain - the NSRS

Provide Access The National Geodetic Survey 10 year plan Mission, Vision and Strategy 2013 – 2023 http://www.ngs.noaa.gov/INFO/NGS10yearplan.pdf

• Official NGS policy as of Jan 9, 2008 (updated in 2013) – Modernized agency – Attention to accuracy – Attention to time-changes – Improved products and services – Integration with other fed missions

• 2022 Targets: – Replace NAD 83 and NAVD 88 – Cm-accuracy access to all coordinates – Customer-focused agency – Global scientific leadership GEODETIC DATUMS VERTICAL 1 D (Orthometric Height) (e.g. NGVD 29, NAVD 88, Tidal)

HORIZONTAL 2 D (Latitude and Longitude) (e.g. NAD 27, NAD 83 (1986))

GEOMETRIC 3 D (Latitude, Longitude and Ellipsoid Height) Fixed and Stable - Coordinates seldom change (e.g. NAD 83 (1994), NAD 83 (2007)) also 4 D (Latitude, Longitude, Ellipsoid Height, Velocities) Coordinates change with time (e.g. ITRF00, ITRF05, ITRF08) Datums - Vertical

National Geodetic Vertical Datum of 1929 - Based on heights at 26 tide stations along US & Canadian coast. - Used 75,000 Km (US) & 30,000 Km (Can.) of leveling data. - Adjustment constrained to the 26 tide stations.

North American Vertical Datum of 1988 - Approx. 650,000 Km of new leveling since NGVD 1929. - Based on (constrained to) agreed upon mark in Great Lakes Region (Father’s Point). - Greatly improved vertical accuracy over NGVD 1929 datum. - Defined only on the conterminous North American continent - Does not (and cannot) exist for any place you can’t level to from Father Point, such as: • Guam, American Samoa, Hawaii, Puerto Rico, American Virgin Islands, Commonwealth of the Northern Marianas, Aleutian Islands VERTICAL CONTROL USED FOR NGVD 1929 DATUM North American Vertical Datum of 1988 – June 24, 1993

Federal Register / Vol. 58, No. 120 SUMMARY: This Notice announces a decision by the Federal Geodetic Control Subcommittee (FGCS) to affirm the North American Vertical Datum of 1988 (NAVD 88) as the official civilian vertical datum for surveying and mapping activities in the United States performed or financed by the Federal Government, and to the extent practicable, legally allowable, and feasible, require that all Federal agencies using or producing vertical height information undertake an orderly transition to NAVD 88.

NGVD 29 NAVD 88 • Datum Definition 26 Tide Gauges in Father’s Point the U.S. & Canada Quebec, Canada • Bench Marks 100,000 450,000 • Leveling (Km) 102,724 1,001,500 • Geoid Fitting Distorted to Fit Best Continental Model MSL Gauges Problems with NAVD 88 (Nationally)

• NAVD 88 relies upon bench marks that: – Disappear by the thousands every year – Are not funded for replacement – Are almost never re-checked for movement – Are affected by freeze / thaw, subsidence, uplift, etc. – Are not necessarily in convenient places – Cross-country error build up

NAVD 88 suffers from a zero height surface that: – Has been proven to be ~50 cm biased from the latest, best geoid models (GRACE satellite) – Has been proven to be ~ 1 meter tilted across CONUS (again, based on the independently computed geoid from the GRACE satellite) 10

Can NAVD 88 be fixed?

• Long term fix: Re-level some/all of NAVD 88 –81,500 km of 1st order leveling at least –625,000 km of mixed 1st and 2nd order leveling

• Re-leveling NAVD 88 estimated to cost between $200 Million and $2 Billion

• Time factor in that amount of leveling

• Still would have problems related to passive control

12 Can NAVD 88 be fixed?

Best long term fix: It has been determined that the best long term fix for having a national vertical datum that is accurate and available, is to replace the NAVD 88 datum with a geopotential reference frame (surface). ……..but how is that accomplished?

13 NEW VERTICAL DATUM (Rationale) • A move away from differentially leveled passive control as the defining mechanism of the reference surface • To be consistent with the shift in the geometric reference frame/ellipsoid (2022) • Improvement in our technical abilities in reference surface realization (geopotential gravimetric reference surface - 1cm accuracy of the geoid (GNSS/GRAV-D)) • Goal - ability to establish 2cm orthometric height anywhere in U.S. using a minimum of 15 min. of GNSS data. • The new geopotential reference surface will be aligned with the geometric reference frame/ellipsoid (i.e., no hybrid geoid) Vertical Datum (Status)

• United States Gravimetric Geoid 2012 (USGG12) Made before GRAV-D data • Hybrid Geoid 2012 (Geoid 12B) available

• 2022 Definition of the Vertical Reference Surface • Yearly Experimental Geoids that Include GRAV-D Airborne Data

• The US is not the first adopter of a vertical datum based on a gravimetric geoid! • New Zealand • First adopter in the world (2009) • Finished airborne gravity over country in 2 months of 2014 • New datum available in late 2015 • Canada • Adopted in 2014 Can NAVD 88 be fixed?

Best long term fix: Replace NAVD 88 with a 1cm vertically accurate geopotential reference frame (surface)……..but how?

First, let’s talk about geoids.

16 Definitions: GEOIDS vs GEOID HEIGHTS

• “The equipotential surface of the Earth’s gravity field which best fits, in the least squares sense, (global) mean sea level.”* • Can’t see the surface or measure it directly. • Can be modeled from gravity data as they are mathematically related. • Note that the geoid is a vertical datum surface • A geoid height is the height from an ellipsoidal datum to a geoid. • Hence, geoid height models are directly tied to the geoid and ellipsoid that define them (i.e., geoid height models are not interchangeable).

*Definition from the Geodetic Glossary, September 1986 Why do we need Geoid Models?

• The geoid surface is mathematically GRACE Satellite Gravity related to gravity • Understanding the gravity field is important because it − impacts survey measurements − dictates the direction water flows • Geoid models allow us to relate different kinds of heights Measuring and relating different kinds of heights

• Ellipsoid heights – Inherent to GNSS measurements – Need accurate ellipsoid height control – Better field procedures = better heights

• Orthometric heights – Measure by leveling surveys – Most accurate but most expensive – Difficult to maintain over time, esp. on national scale

• Geoid heights – Derived from model developed from gravity observations – Can provide relationship between e.h. and o.h. The ellipsoid, the geoid, and you Deflection of the vertical

You are here Earth surface

Ellipsoid height, h Orthometric height, H Mean sea level Geoid height, NG

h = H + NG

Note: Geoid height is negative everywhere in the coterminous US Types Uses and History of Geoid Height Models

• Gravimetric (or Gravity) Geoid Height Models – Defined by satellite gravity, EGM08, RTM model – Refined by terrain models (DEM’s) – Purely a scientific model for engineering applications – Does not provide for GPS to NAVD88 relationship

• Composite (or Hybrid) Geoid Height Models – Gravimetric geoid defines most regions – Warped to fit available GPSBM control data – Defined by legislated ellipsoid (NAD 83) and local vertical datum (NAVD 88, PRVD02, etc.) – May be statutory for some surveying & mapping applications USGG2012(Gravimetric Geoid)

• Satellite Gravity Models + EGM08

• 2.6 million terrestrial, ship-borne, altimetric gravity measurements

• 30 arc second Digital Elevation Data

• Computed on 1 x 1 arc minute grid spacing

• GRS-80 ellipsoid Basic Concepts of Hybrid Geoid Modeling (NGS)

• Start with a gravimetric geoid (USGG2012) • Use control data to fit to local datums – Appropriate versions of NAD 83 – Respective local Vertical Datum (if one exists) • Use LSC to determine correlated signal • For complex areas (e.g., CONUS), use MMLSC • Apply grid of correlated signal to USGG2012 • Results in GEOID12 with high frequency nature from USGG2012 but fit to local control GPS Bench Marks used for Geoid 12B (23,961)

GPSBM1999: 6,169 total 0 Canada STDEV 9.2 cm (2σ) GPSBM2003: 14,185 total 579 Canada STDEV 4.8 cm (2σ) GPSBM2009: 18,291 total 576 Canada STDEV 2.8 cm (2σ) TX GPSBM Marks for Geoid 12B (377) Which Geoid for Which NAD 83?

NAD 83(2011) --- Geoid12B

NAD 83(2007) --- Geoid09

NAD 83(19xx) --- Geoid03 --- Geoid99 --- Geoid96 Current and Best NGS Geoid: USGG2012

• GRACE + 2 yrs. of GOCE + other satellite data • + World Gravity Model (EGM2008) • + Estimated Gravity from Topography • + Around Two Million Surface Gravity Points

But even this isn’t good enough for a 1-2 cm accuracy vertical reference surface Problems with Gravity Holdings

20-100 km Terrestrial • Field is not sampled gravity gravity gaps along coast uniformly

• Data range in age and quality, some w/o metadata

• Some surveys have Ship systematic errors New gravity Orleans • Data gaps in littoral areas Can NAVD 88 be fixed?

Best long term fix: Replace NAVD 88 with geopotential reference frame

• GRAV-D: Gravity for the Redefinition of the American Vertical Datum • GRAV-D will: • define the datum as a given geoid model • realize it through GNSS technology • spanning at least the United States and its territories • datum will be tracked (maintained) over time

29 Gravity for the Redefinition of the American Vertical Datum (GRAV-D)

• Replace the Vertical Datum of the USA by 2022 (at today’s funding) • GRAV-D is: – An airborne gravity survey of the entire country and its holdings – A 2022 gravimetric geoid accurate to 1 cm – Long-term monitoring of geoid change over time – Partnership surveys • Working to launch a collaborative effort with the USGS for simultaneous magnetic measurement Gravity and Heights are • Acting Manager: Monica Youngman inseparably connected [email protected] Building a Gravity Field

Long Wavelengths (≥ 250 km)

GRACE/GOCE/Satellite Altimetry +

Intermediate Wavelengths (500 km to 20 km) Airborne Measurement + Short Wavelengths (< 100 km) Surface Measurement and Predicted Gravity from Topography Priority- Greatest Datum Need

• Alaska • Puerto Rico/US Virgin Islands (PRVI) • Coastal US and Great Lakes • Great Lakes • Gulf of Mexico & FL • Eastern Seaboard • Western Seaboard • Hawaii and Pacific Islands • Aleutian Islands • Interior CONUS • Mountainous areas first Data Collection Scope

• Entire U.S. and territories – Total Square Kilometers: 15.6 million – ~200 km buffer around territory or shelf break if possible – Initial target area for 2022 deadline GRAV-D Data Status (May 2015)

CONUS – 44%

AK – 41% PR – 100% Survey and Block Plans

• Data lines spaced 10 km apart • Cross lines spaced 60- 80 km apart • Flight altitude 20,000 ft. • Nominal speed 220-250 knots Requirements

• Geodetic quality results require accurate aircraft positions, velocities, and accelerations

• High-altitude, high-speed, long baseline flights for gravimetry

INS GPS Antenna

Gravimeter Absolute Gravity Tie Methods: Survey Reconnaissance

Parking spot ID

A10 measurement

GPS Base Stations Vertical gravity gradient Data Processing

• Field Quality Control • Final Processing Processing – GPS/IMU combined solution using Precise – GPS/IMU combined solution Point Positioning (PPP) using differential positioning – Gravity processing with Newton (internal – Gravity processing with Issac software) (internal software) – Quality evaluated and data minimally trimmed – Processed data and documentation released Green = EGM08, Blue = Airborne Lake Michigan Online Data Portal http://www.ngs.noaa.gov/GRAV-D/data_products.shtml

Interactive Google Map: – Up-to-date info on all blocks – 27 blocks released (Was 20 in Jan. 2014) – Clicking a block pops up basic info and a link to the block’s page if data is available Validation

• Internal Validation • External Validation – Crossover analysis – Updates to EGM08 compared to GOCE – Line to line comparison – Geoid Slope Validation Surveys – Line reflights (2011, 2014, 2016) Validating Geoid Accuracy

• NGS plans up to 3 surveys to validate the accuracy of the gravimetric geoid model

– GSVS11 • 2011; Low/Flat/Simple: Texas; Done; Success! – GSVS14 • 2014; High/Flat/Complicated: Iowa; Field work Complete – GSVS1x • 2017?; High/Rugged: Colorado? Wyoming? Geoid Slope Validation Survey

Austin • Observe geoid shape (slope) using multiple independent terrestrial survey methods – GPS + Leveling – Deflections of the Vertical • Compare observed slopes (from

325 km terrestrial surveys) to modeled 218 points slopes (from gravimetry or 1.5 km apart satellites) – With / Without new GRAV-D Rockport airborne gravity Surveys Performed

• GPS: 20 identical units, 10/day leapfrog, 40 hrs. ea. • Leveling: 1st order, class II, digital barcode leveling • Gravity: FG-5 and A-10 anchors, 4 L/R in 2 teams • DoV: ETH Zurich DIADEM GPS & camera system • LIDAR: Riegl Q680i-D, 2 pt/m2 spacing, 0.5 km width • Imagery: Applanix 439 RGB DualCam, 5000’ AGL • Other: RTN, short-session GPS, extra gravity marks around Austin, gravity gradients GPS LIDAR/ DoV Imagery Gravity

Leveling Geoid Slope Survey Conclusions

• Including airborne gravity data improves geoid slope accuracy at nearly all distances <325 km

• The NGS geoid in the TX survey meets the 1 cm accuracy objective only if airborne data are included – No other model achieved 1 cm accuracy

• Gravimetric geoid models and GPS are a viable alternative to long-line leveling GSVS11 Results Geoids without GRAV-D

3.5 Predicted Errors of various geoid models over GSVS11 after removal of GPS/Leveling error budget USGG2009 3 EGM2008 xEGM-G 2.5 xEGM-GA xUSGG2011 2 xUSGG-GA-R-K480

Geoids with 1.5 GRAV-D

1 RMS Errors (cm) Errors RMS

0.5 “1 cm geoid”

0

Distances between points (km) 47 GSVS – 2014 (Iowa)

Cedar Rapids, IA

Denison, IA

Central Iowa 330 km . Minimal Trees (GPS) 205 points . No major bridges/water crossings (Leveling) . Along roads (ease of access) . Mostly cloud-free nights . Under GRAV-D coverage (or will be) . With at least some existent bench marks/NAVD 88 level lines . A "significant enough" geoid slope to be interesting . For GSVS11, we wanted "low and flat" topography. For GSVS14 we'll want to step this up to "high and flat" or possibly "medium high, more-or-less flat and gravimetrically complex" GSVS 14 Measurements Performed

• GPS: 20 identical units, 10/day leapfrog, 40 hrs. ea. • Leveling: 1st order, class II, digital barcode leveling • Gravity: FG-5 and A-10 anchors, 4 L/R in 2 teams • DoV: NGS-trained crew with ETH Zurich CODIA GPS & camera (2014) • LIDAR: Riegl Q680i-D, 2 pt/m2 spacing, 0.5 km width • Imagery: Applanix 439 RGB DualCam, 5000’ AGL • Other: Iowa RTN, short-session GPS GSVS14 – GPS Equipment Validation GSVS14

GSVS1X – 3rd Slope Validation Survey

– Currently in planning stages. – Want area of high, rugged terrain, somewhere gravitationally challenging. – Field observations most likely in 2016–17. Moving Toward 2022 - Experimental Geoid Models

NGS will annually produce experimental gravimetric geoids with all available gravity sources (GRAV-D data, etc.)

– Geoid Evolution: • from USGG2012 (last non-GRAV-D gravimetric geoid) • through xGG20_ _ (annual experimental geoid) • to USGG2022 (the final geoid with GRAV-D) – Required annual release, reportable to NGS’ line office (National Ocean Service) – June 2014: First experimental geoid (xGG2014) was released on beta website xGEOID14 – Includes all GRAV-D data released before December 2013. – 6 months to make the model, tools, and website. – Experimental orthometric heights available at all points in NW hemisphere xGEOID14 model. White boxes are areas of GRAV-D airborne data included in the model. Also includes new satellite gravity. Retrieve xGEOID14 Heights

- Points MUST have ellipsoidal heights

- Up to 20 points at once

- Pick a latitude, longitude format - Identify the input horizontal datum - Type in the coordinates, ellipsoidal height, and name of point

- Submit coordinates - Fast computation Notes on xGG2014B

• **NOT FOR ANY OFFICIAL USES** • Only for your information in estimating height changes at points you use. • These heights aren’t exactly the same as will be adopted in 2022, but are similar. Why? – GRAV-D isn’t complete yet and the data gaps affect the geoid model, even in areas with GRAV-D data – Geoid modeling methods continually improving and will be better by 2022, thus changing the heights. – Additional terrestrial and/or satellite data will likely become available by 2022. • First time CONUS + AK + Islands on same geoid Break for Lunch Datums - Horizontal Horizontal North American Datum Of 1927 - Used early triangulation surveys (first called US Std Datum). - Based at Meades Ranch, KS & used Clarke 1866 spheroid.

North American Datum of 1983 (1986) - Used 250,000 points (triangulation, Doppler, a few GPS pts). - Based at geocentric location using Geodetic Ref. Sys. 1980.

Updates* NAD83 (1990’s) – State HARN / FBN surveys.

NAD83 (late 90’s – early 2000’s) – Efforts to establish ellipsoid heights.

NAD83 (2007) – Remove regional distortions, better with align CORS.

NAD83 (2011) – Adjustment to realign CORS / passive control.

* Not a new datums

National Spatial Reference System (NSRS) Improvements over time

NETWORK TIME NETWORK LOCAL SHIFT SPAN ACCURACY ACCURACY

NAD 27 1927-1986 10 meters (1:100,000) 10-200 m

NAD83(86) 1986-1990 1 meter (1:100,000) 0.3-1.0 m

NAD83(199x)* 1990-2007 0.1 meter (1:1 million) 0.05 m “HARN”, “FBN” (1:10 million)

NAD83(NSRS2007) 2007-2011 0.01 meter 0.01 meter 0.03 m

NAD83(2011) 2011 0.01 meter 0.01 meter 0.01 m Why Replace NAD83?

• NAD83 is NON-geocentric & hence inconsistent w/GNSS

• Difficult to maintain consistency between CORS & passive networks

• NAD 83 does not track/report passive mark motion

• Lack of monumentation stability / permanency

• Necessary for agreement with future ubiquitous positioning capability NAD83 Shortcomings

• 2.2 m offset

• CORS & passive network inconsistency

vs. CORS Network

 2000 sites

 225 organizations NAD 83 (2011) epoch 2010.00 76,664 stations “Height Mod” stations

11/19/2013 Height Modernization Surveys 67 OPUS Shared Solutions

11/19/2013 Height Modernization Surveys 68 Future Geometric (3-D) Datum

• Replace NAD83 with new geometric datum – by 2022

• CORS-based, accessed via GNSS observations

• Coordinates & velocities in ITRF & official US datum & relationship

• (NAD83 replacement: plate-fixed or “ITRF-like”)

• Passive control tied to new datum; not a component of new datum

• Address user needs of datum coordinate consistency vs. accuracy New Datum, New Heights

• How much will NSRS ellipsoid height change? – Geometric Datum will be aligned to ITRF – Estimated -1.9 m (Puerto Rico) to +2.0 m (Guam) – Estimated Texas: -1.3 to -1.0 meters (negative)

• How much will NSRS CONUS orthometric height change? – Estimated +0.1 m (Florida) to -1.3 m (Washington) – More than 2 meters of change in Alaska – Estimated Texas: near 0.0 to -0.5 meters (0.0 to – 1.6 feet)

New geometric datum minus NAD 83 (ellipsoid height) New geometric datum minus NAD 83 (horizontal) Estimating 2022 height changes

To estimate the height change of any point:

If you have only a GPS position……. ** Orthometric Heights Above Future Geopotential Datum. - Submit coordinates on xGG2012 website Prototype orthometric heights are now being made available as a precursor to the completion - Theof NAVD88 GRAV-D and the orthometric replacement of NAVD height 88 with a is new similar geopotential to reference the GEOID12 system. The following height reflects the current best estimate of the true orthometric height, based on Hybridthe existing orthometricgravimetric geoid model. height This height is subject to change as data and modeling for the gravimetric geoid change throughout the lifetime of the GRAV-D project, or as new realizations of the ITRF are adopted. However, at the completion of GRAV-D, these heights will supersede the NAVD 88 heights

If you haveAPPROX raw ORTHO GPS HGT: data…….. 23.597 (m) [PROTOTYPE (Computed using USGG2012,GRS80,IGS08)] - Submit data to OPUS and select the extended output. This position and the above vector components were computed without any - USGG12knowledge ortho by the Nationalhgt will Geodetic be Survey similar regarding to the new equipment height or to a few cm. field operating procedures used. How to Plan for 2022 • Move to newest realizations – NAD 83(2011) epoch 2010.00 – GEOID12B (hybrid geoid) • Move to NAVD 88 – understand the accuracy of VERTCON in your area • Utilize passive marks that are up to date – Stay aware of what marks are included in datum updates – Use OPUS & Hgt. Mod procedures to update mark positions • Use OPUS for GPSBMs – Help improve future geoid models & relationship with new datum • NGS Outreach Efforts – Participate in NGS webinars, Geospatial summits, contact NGS Regional Advisor, etc. Metadata is Critical

• Your positional metadata should include: – datum – epoch – source / methods • These will facilitate transforming from current to new datum • Maintaining your original survey data will provide more accurate results

GPS on Bench Marks • http://www.ngs.noaa.gov/heightmod/GPSonBM.shtml • 2014 prioritized bench marks (needing GPS observations) by GEOID12A estimated accuracy, vertical order, stability, and distance from road 2015 - Collaborating with NSPS for another Surveyors Week Campaign

- NGS is making new maps of priority marks

- NGS is making new associated materials to be ready in Feb. 2015

- Becoming year-round campaign Find this article at: http://www.amerisurv.com/content/view/ 13361/153/

Four-part series The first part, covering the NGS Ten Year Plan, OPUS-DB, OPUS-Projects, and Foundation CORS appeared in the January 2015 issue. ?? Questions ??

No gravity, no height. Know gravity, know height.

Denis Riordan, National Geodetic Survey Advisor to Mississippi Regional Advisor to AL, FL, & LA [email protected] (601) 359-5357