NGS releases annual experimental geoid models and gravity interpolation tools

My last column highlighted an ArcGIS web application that incorporates various datasets and data layers to assist surveyors planning vertical control surveys. On Jan, 29, the National Geodetic Survey (NGS) released the latest experimental geoid model, xGeoid20, and a new gravity interpolation tool (see box below, “NGS Releases Annual e& Gravity Interpolation Tools”).

This newsletter will highlight some attributes of these two new products. First, why am I writing about another experimental geoid model. I discussed xGeoid18 in my December 2018 column and xGeoid16 in my June 2017 column. What’s important here is that this will be the last experimental geoid model until 2022, and the dynamic geoid model has also been updated this year in the form of xDGEOID20.

xDGEOID20 is produced by NGS within the Geoid Monitoring Service (GeMS) and is part of the new NAPGD2022. Therefore, users only have a few more years to understand the differences between the hybrid geoid model that is being used today to estimate GNSS-derived orthometric heights and the gravimetric geoid model which will be used to estimate North American-Pacific Geopotential Datum of 2022 (NAPGD2022) GNSS-derived orthometric heights.

NGS also announced a new gravity tool, denoted as “The Experimental Gravity Model 2020 (xGRAV20).” xGRAV20 is designed to provide a full-field gravity value and a digital elevation model height at a-specified location. The xGRAV20 model will be important to users that are computing leveling-derived orthometric heights consistent with NAPGD2022.

It is important to note that the xGEOIDs provide a preliminary but increasingly-accurate view of the changes expected from the upcoming NAPGD2022. Also, the xGEOID20 geoid model is the first combination of the geoid models computed by scientists at NGS and Canadian Geodetic Survey (CGS). One unique element to xGEOID20 is that the differences between the A and the B model are due to the contribution of the GRAV-D airborne gravity and differences in methodology.

The National Geodetic Survey (NGS) has published annual experimental geoid (xGEOID) models since 2014. Each of these experimental geoids demonstrate the improvements provided by the addition of airborne gravity data (GRAV-D data) and by the refinement of geoid computation methods.

First, users can access the xGeoid20 model here. See the box titled Experimental Geoid Models 2020 (xGEOID20).

As the image above indicates, the xGEOID20 is available over a very large area. The box below lists the latitude and longitude boundaries of the areas where xGeoid20 is available.

To use the xGeoid20 Interactive Computation Page, the user can click on the “ACCESS TOOL” button below the map or the Interactive Computation button on the left side of the webpage (see the image above, “Experimental Geoid Models 2020 (xGEOID20)”). I’d like to highlight a statement that NGS added as a mote on the computation page:

1. Coordinates will be processed as IGS14.
2. The epoch should be in decimal year format and reflect the user-specified output epoch. If no epoch is entered, the tool will use a default epoch equal to the epoch of the static geoid model, which is currently 2020.00.

The user needs to know that the epoch is used to compute the xDGEOID20 value. I will demonstrate how this works later in this column.

As in past xGeoid interactive computations web applications, the user can submit data in various formats. The box titled “Input Formats Permitted for xGeoid20 Webtool” provides a list of the permitted formats. It should be noted that inputting an ellipsoidal height, epoch and name are optional. However, the default epoch is 2020.00, so if you want a different epoch, you need to enter the date. Also. the program will only compute an orthometric height if the user provides an ellipsoidal height.

Users have the option of getting the output from the xGeoid20 tool on their computer screen or in the CSV format. The box below is an example of inputting data using the screen option. Once you enter your data, the user clicks on the submit button.

The next image shows an example of the output using the screen option. I have highlighted a few numbers that I’d like to address.

• Your input in NAD83 (2011) epoch 2010.00 (red). I entered my coordinates as NAD 83 (2011), and it assumed that these coordinates are epoch 2010.0.
• Your Result in IGS14 epoch 2020.00 (blue). The routine provides your output coordinates in IGS14, epoch 2020.00. This is the epoch of the static geoid model.
• The geoid height of GEOID18 (with respect to NAD83) and The orthometric height in NAVD88 (based on GEOID18) (green). This NAVD 88 value is for comparison purposes only. It is using GEOID18 and provides an estimate of the differences between the future NAPGD2022 and the current NAVD 88. The orthometric height is computed using the following formula: NAD 83 (2011) ellipsoid height (epoch 2010.0} minus GEOID18.
• Ortho Height (brown). This is the estimation of the orthometric height using the following formula: IGS14 ellipsoid height (epoch 2020.0} minus xGEOID20A (or B).
• Ortho(model)-NAVD88(GEOID18) (purple). These differences are the estimates of the differences between the future NAPGD2022 and the current NAVD 88. It provides the differences for both the xGeoid20A and xGeoid20B model. I look at the B model because it used the GRAV-D data in the development of the model.
• Accuracy (yellow). This is the estimated 95% confidence interval for geoid height.
  Example of Output Format from Screen Option.
  

  ---

xGEOID20 Interactive Computation Output

Note: The GRS80 ellipsoid is used for both NAD83 and IGS14.

N: The geoid height at epoch t0 = 2020.0, which is geocentric and relative to the GRS80 reference ellipsoid.

Accuracy: Estimated 95% confidence interval for geoid height.

DN: The time-dependent geoid change computed between user inputted epoch (t) and t0. To obtain the dynamic geoid height at user inputted epoch (t), add N + DN.
Either Model A or Model B N values may be used for this depending on user preference.

The box below shows an example of inputting data using the CSV option.

---

Example of Output Format from CSV Option

Note: The GRS80 ellipsoid is used for both NAD83 and IGS14.

N: the geoid height at epoch t0 = 2020.0, which is geocentric and relative to the GRS80 reference ellipsoid.

Accuracy: Estimated 95% confidence interval for geoid height.

DN: the time-dependent geoid change computed between user inputted epoch (t) and t0. To obtain the dynamic geoid height at user inputted epoch (t), add N + DN. Either Model A or Model B N values may be used for this depending on user preference.

Cnt,Station,NAD83_Lat,NAD83_Lon,NAD83_Eht,Input_Epoch,IGS14_Lat,IGS14_Lon,IGS14_Eht,Output_Epoch,GEOID18_Ht,Oht_NAVD88,xGEOID20A_Ht,xGEOID20B_Ht,xGEOID20A_Accuracy,Oht_xGEOID20B,Oht_NAVD88,Oht_Diff(xGEOID20A-NAVD88),Oht_Diff(xGEOID20B-NAVD88),DN,Epoch

0,PA,40.616935533762,77.4066810996784,222.425581993569,2010.00,40.6169445389,77.4066880139,221.191,2020.00,-33.685,256.111,-34.475,-34.477,0.039,255.666,255.668,-0.445,-0.443,0.000,2020.000
1,PR,18.2570177272727,66.5508117355371,6.65385123966942,2010.00,18.2570227778,66.5508102806,4.776,2020.00,-39.379,46.033,-41.690,-41.679,0.040,46.466,46.455,0.433,0.422,0.000,2020.000

The printed output from the CSV option looks very confusing, but it can be imported into an excel spreadsheet. The headings and values are all separated by a comma so everything falls into the appropriate columns after importing the data (see image below.)

I stated in the xGeoid20 write up that the dynamic geoid model has also been updated this year in the form of xDGEOID20. This model is produced by NGS within the Geoid Monitoring Service (GeMS) and is part of the new NAPGD2022. For a thorough discussion on GeMS and the time-dependent geoid, view the webinar from NGS’ presentation library. See the box titled “GeMS Webinar by Kevin Ahlgren.”

Also, one of my previous columns described NGS’ GeMS program. The images titled “Examples of the Time-Dependent Geoid Change in Alaska EPOCH 2020.0” and “Examples of the Time-Dependent Geoid Change in Alaska EPOCH 2025.0” show the change in geoid value from Epoch 2020 to Epoch 2025 for two stations in Alaska.

First, looking at the box titled “Examples of the Time-Dependent Geoid Change in Alaska EPOCH 2020.0,” the change between NAPGD2022 and NAVD 88 is approximately 1 meter. Users should note that the GEOID12B is used to establish the NAVD 88 height. Alaska was not included in GEOID18. Comparing the two Alaska labeled boxes, the xDGEOID2022 change between 2020.0 and 2025.0 is –4 mm. I will address this topic in more detail in future newsletters.

As stated by NGS news announcement, “The xGEOID models provide a preliminary but increasingly-accurate view of the changes expected from the upcoming North American-Pacific Geopotential Datum of 2022 (NAPGD2022).” NGS has produced many figures that describe the bias and trend between the future NADGP2022 and NAVD 88. In my June 2017 column I provided a plot that depicted the difference between NAPGD2022 and NAVD 88 based on the GPS on Bench Mark dataset. See the image below.

These figures provide a broad picture of the change but to better understand the changes across the Nation, I used the GPS on Bench Mark dataset, that was involved in the creation of Geoid18 model, to compute an average latitude, longitude, and ellipsoid height for every State. Obviously, this is a fictitious mark but it provides an idea of the average change based on marks that have both a GNSS-derived ellipsoid and a leveling-derived orthometric height. The plot titled “Difference Between the Future NAPGD2022 and NAVD 88” depicts the average difference for each state based on the GPS on Bench Mark data file. These differences were generated using the xGeoid20B values from the output of the xGeoid20 website.

I would encourage everyone to select a couple of marks and compute the differences to understand the change in their particular region. I was the NAVD 88 Project Manager and I informed users of the potential changes between the NGVD 29 and NAVD 88 for about a decade, and I still had surveyors tell me that they didn’t know it was coming. Please take a few minutes to read NGS’ write up on xGEOID20, estimate the differences in your area of interest, and spread the word to your colleagues, friends, and clients.

The last item that I’d like to highlight is that NGS has released a beta version of s surface gravity model consistent with xGEOID20. See the box titled “Experimental Surface Gravity Model 2020 (xGRAV20).” Users can access the beta webtool here.

The access and input to the tool is similar to the xGEOID20 web tool. Saying that, I’d like highlight a few items:

• The input height should be an orthometric type of height not an ellipsoid height.
• If a height is entered, the tool will assume that is correct and use it for the gravity prediction.
• If you do not know the elevation, leave the entry blank. The tool will use the DEM interpolated height if it is blank.

The box below provides the output using the tools sample data.

This gravity tool will be important when users want to incorporate leveling-derived orthometric heights into NAPGD2022. We will address this tool in more detail in future newsletters. I want to emphasis that these two web tools are beta sites. As a beta site, users should verify all information from the site. I encourage everyone to access the tool and check out a few of their favorite marks, and then send an email to NGS informing them of what you like, what you would like to change, and what you would like to see added to the tool.

NGS is releasing this tool as a beta product to get feedback from users. They are interested in your feedback concerning its function and usability as well as how users would like to interact with NGS web tools in the future. Email NGS at ngs.feedback@noaa.gov.

In conclusion, I want to leave you with a thought about change. When I give presentations and seminars, I usually include a slide that probably expresses the thoughts of many individuals.

My brother once told me:

“If you geodesists did it correctly the first time you wouldn’t have to keep performing adjustments and changing the values. Just do it right the first time.”

He’s a doctor and said he must do it right the first time.

My response to my brother and to everyone else is the following:

If you want to improve you have to be willing to change, and if you want to continue to meet future positioning requirements you need to continually change.

Winston Churchill said it better “To improve is to change; to be perfect is to change often.”