A 1.974Mb PDF of this article as it appeared in the magazine—complete with images—is available by clicking HERE
Updating NGS 58 and NGS 59
The advent and evolution of new technology such as Global Navigation Satellite Systems (GNSS) over the last thirty years has allowed NGS, other Federal, State and Local Agencies, and the Private Sector to determine geodetic positions with greater speed, better accuracy, and less personnel than ever before. Many of these coordinates and their accompanying observations have been used to extend and densify the NSRS. Standards and specifications are required to ensure that the methods for acquiring and processing the data provide results of sufficient quality for reliably and accurately augmenting the NSRS.
Determining accurate orthometric heights (elevations) is an important and challenging aspect of updating the NSRS. The vertical reference frame of the NSRS in North America—the North American Vertical Datum of 1988 (NAVD 88)—is currently realized (or “accessed”) through a nationwide network of passive marks with official (“published”) heights determined primarily, but not exclusively, through leveling observations. Expansion of this realization through traditional optical and digital leveling survey techniques, while very precise, is costly and time consuming. This is particularly a problem in dynamic regions where vertical changes (such as subsidence) require re-observation on a frequent basis.
In order to take advantage of GPS surveying to economically densify vertical control, NGS published guidelines for using static, post-processing GPS technology to establish accurate ellipsoid heights (NGS 58 in 1997) and accurate orthometric heights (NGS 59 in 2008) on passive control. Since those guidelines were published, GNSS equipment and data processing techniques have improved, more precise GNSS orbits are available, additional frequencies and codes have been added to GPS, real-time GNSS reference networks cover much of the U.S., additional navigation satellite constellations have been launched, and significant improvements in geoid computation have been realized. These advances prompted NGS to re-examine how GNSS may best be used to optimally determine ellipsoid heights and orthometric heights, whether or not it is applied to extending passive vertical control. Working with GCT Engineering and The Ohio State University, data are being collected and processed as part of an overall re-examination of NGS 58 and NGS 59.
This re-examination will include a re-evaluation of the effects of factors contributing to the accuracy and precision of GPS; in particular on the determination of ellipsoid heights. These factors include for example: the distance to existing passive and active control, the occupation time, and the data processing strategies. Determining how these factors impact ellipsoid height accuracy will allow NGS to publish new standards and specifications to achieve various desired ellipsoid height accuracies and precisions.
Additionally, this effort will examine how these same factors, plus others, may be used to determine orthometric heights. Current practice as specified in NGS 59 requires significant field time and (usually) the construction of additional passive control, as well as the inclusion of existing NAVD 88 benchmarks as constraints in a least squares adjustment of a local network, while using the most recently available hybrid geoid model. This analysis will aid in determining if the NGS 59 standards likewise need to be updated.
GPS On Bench Marks Campaign
One key aspect of existing hybrid geoid models is their reliance upon the quality and distribution of the control data used to define them. These control data take the form of GPS-derived NAD 83 ellipsoid heights located at the same place as leveling-based NAVD 88 orthometric heights. At each such point, the separation between the datums is known and can be used to control the surface of the hybrid geoid models, such as GEOID12A. Unfortunately equitable spatial distribution of such points is difficult to manage. The hybrid geoid modeling techniques will interpolate over any gaps in coverage, but gaps greater than 30 km will produce interpolation errors on the order of many centimeters. Gaps as large as 100 km have been seen to cause errors between 5 and 10 cm. Hence, locating and filling such gaps in coverage and remain the one of the best means of improving future hybrid geoid models.
A GPS occupation of a leveled benchmark in such a gap can be easily made available to NGS by sharing the solution in OPUS-DB. This will provide a value that can be tested and potentially used in the next hybrid geoid model. Optimally though, the data should be Bluebooked and loaded into the NGS IDB. However, even data of potentially lower quality is much better than no data at all.
As part of NGS’ focus on filling such gaps, last spring, volunteers across the United States celebrated National Surveyors Week by participating in a national "GPS on Bench Marks Campaign." Surveyors collected GPS field data and shared it via OPUS-DB. These submissions helped set a new monthly record for March 2014 when 282 GPS data files and solutions were shared, along with photos and descriptions of the survey markers. The solutions are now publicly available at http://www.noaa.gov/OPUS/view.jsp.
Future GPS on Bench Mark Campaign efforts will help ensure a transition tool is available to transform heights between the North American Vertical Datum of 1988 (NAVD 88) and the new vertical datum when it is released in 2022. While the primary method of defining and accessing the new vertical datum will be through GNSS technology, the current NAVD 88 is founded on historic geodetic leveling surveys on thousands of bench marks that are often difficult and expensive to access. Establishing GNSS coordinates on NAVD 88 bench marks will create a direct tie between the official leveling-based height system and the new vertical datum.
By adding GNSS coordinates to existing NAVD 88 bench marks, NGS can also improve the hybrid geoid models, until 2022 when the use of a hybrid geoid model will no longer be necessary. The current hybrid geoid, GEOID12A, includes many "GPS on Bench Mark" observations, but future models would benefit from additional data collection in many areas. NGS expects to coordinate with the National Society of Professional Surveyors and others around the country to complete future GPS on Bench Mark Campaigns. To learn more, visit the Campaign’s web-page at http://www.ngs.noaa.gov/heightmod/GPSonBM.shtml.
Datum Transformation Tools
In 2012, NGS made the GEOCON and GEOCON11 (versions 1.0) tools available on a test basis (on the NGS "Beta" web site). These tools were developed in response to user requests that a transformation tool (akin to NADCON) be made available which connects pre-2007 realizations of NAD 83 to NAD 83(NSRS2007) and to NAD 83(2011). While both tools transform latitude and longitude (like NADCON) they also can transform ellipsoid heights. Both tools worked in three areas: CONUS, Alaska and Puerto Rico/Virgin Islands. When initially released, the description for these tools stated that GEOCON11 would transform between NAD 83(NSRS2007) and NAD 83(2011), while GEOCON would transform between NAD 83(HARN) and NAD 83(NSRS2007). To be more specific, that use of "NAD 83(HARN)" was a bit of a misnomer. The connection in GEOCON (version 1.0) was between NAD 83(NSRS2007) and "the most recent post-1986, pre-2007 published NAD 83 coordinate on a point". Thus a mix of coordinates from a given state’s HARN, FBN or both or even neither (with no distinction between them) was put into GEOCON version 1.0. In August of 2014, GEOCON and GEOCON11 (version 1.0) exited beta testing and both became official products of NGS.
During the two years of beta testing, the two primary pieces of feedback that NGS received on GEOCON and GEOCON11 were that (a) users wanted the software to work with more formats than FGCS’s Bluebook format and (b) users in many states that distinguish between their "HARN" and "FBN" realizations of NAD 83 wanted the software to recognize that distinction. The first request was fulfilled by the creation of GEOCON version 1.1 and GEOCON11 version 1.1, wherein "free format" was added as an option to both input and output of both tools. This version was released on the NGS Beta website on September 30, 2014.
The second request represented a significant philosophical change to the GEOCON software, and took longer to develop. It was given the name GEOCON v2.0. Because of the significant changes necessary to change GEOCON, a number of other improvements were also made at the same time. Among the changes coming are the following:
1. For every state and territory in the nation, a limited, rigidly defined set of "supported realizations" was created. For most states, there was just one "supported realization" between 1986 and 2007, which was given the moniker "NAD 83(FBN)", for purposes that will be clear later. However, for 19 other states, two realizations were supported, and given the monikers "NAD 83(HARN)" and "NAD 83(FBN)" in chronological order. In one territory (Puerto Rico) it was necessary to support 3 realizations between 1986 and 2007, as the initial HARN was found to be in systematic error and re-released years after it was already in use.
2. All data in a state that was not on a "supported realization" (due to "feathering" one state to another state’s HARN while the HARNs and FBNs were being expanded) were not used. For example, in Ohio, the only "supported realization" between 1986 and 2007 is NAD 83(1995). However, data does exist in Ohio with the datum tags of 1992, 1993, 1994, 1996 and 1997. These data represent points in Ohio used in other (surrounding states) in the creation of their HARNs and/or FBNs. Such data points (17% of all Ohio data with a published realization between 1986 and 2007) were discarded in the creation of GEOCON v2.0.
3. The tool was expanded to cover three new regions: Hawaii, American Samoa and Guam/CNMI, including support for NAD 83(PA11) and NAD 83(MA11).
4. In those states where 2 "supported realizations" existed (3 for PR), the tool allows for transformation from the HARN to the FBN as a separate transformation, valid only in that state. However, once users transform into the FBN, then the "regional approach" that began with version 1.0 remains valid–that is, transforming from FBN to NSRS2007 will be allowed on a regional multi-state basis. The reason for this, as mentioned in the version 1.0 documentation, is that there is significant continuity in the FBNs from one state to the next. This is why, in states with just one supported realization, it was decided to refer to that single realization as "NAD 83(FBN)".
5. GEOCON and GEOCON11 were combined into one tool, now called GEOCON v2.0.
GEOCON v2.0 is currently scheduled for release to public for beta testing in the spring of 2015.
NGS Data Explorer
The National Geodetic Survey Data Explorer is a web mapping application which NGS introduced in 2012. It allows users to view NGS geodetic control data across the United States and its territories using Google Maps. It replaced the previous tool "The NGS Map", and is significantly easier to use.
As an interactive tool, this application enables users to explore the extensive geodetic control network of the NSRS as well as the NGS datasheet documentation for the specified control marks. Users can find control in their area of interest searching by geographic location, coordinates or the PID of individual marks. NGS Data Explorer also provides access to control mark information including the latitude, longitude, elevation, position source, and other available data.
Dr. Dru Smith has been the Chief Geodesist at NGS since 2005, and most recently led the development of the NGS Ten Year Strategic Plan. During his years at NGS, he has been involved in geoid modeling, ionosphere research and most recently in updating the datum transformation software GEOCON.
Dr. Gerald Mader is Chief of the Geosciences Research Division at NGS and has worked extensively developing static and kinematic GPS data processing techniques and software. He has been the driving force behind the development of the entire OPUS suite of tools as well as the NGS antenna calibration program.
Brian Shaw is a Geodesist at the National Geodetic Survey in the Geodetic Services Division. He holds a BS in Computer Science and an MS in Geographic Information Systems.
Christine Gallagher recently became the Constituent Resource Manager at NGS after helping coordinate the National Height Modernization Program for the past five years. She has worked with programs across federal agencies that aim to improve decision-making by using accurate geospatial information, and she previously earned a Masters in Engineering and Public Policy at the University of Maryland.
A 1.974Mb PDF of this article as it appeared in the magazine—complete with images—is available by clicking HERE