Online Positioning User Service Review and new developments OPUS Online Positioning User Service Review and new developments Scott Lokken NGS State Advisor (NC) Scott.lokken@noaa.gov 919-733-3836 work 240-678-2167 cell

Positions and Elevations from GPS Static Occupations Different Flavors of OPUS OPUS Rapid Static (15min to 2 hrs) OPUS Static (2 hours +) Sharing (Optional) OPUS Projects (Network Least Squares solution) OPUS Net (Under development) GNSS (Under development)

OPUS Static limitations Dual-frequency data (L1/L2) [recommended] Minimum 2 hrs of data (maximum 48—only cross midnight once) No kinematic data No Glonass. Galileo will be discussed as the constellation becomes available Some of the OPUS Guidelines are the data must be dual frequency (for now) it should be at least 2 hours in duration No kinematic or rapid static No glonass You must supply the correct antenna type and height, or I wouldn’t bet on the results of the vertical component And the file name should be the standard 8.3 CORS naming convention. Accurate height requires: correct antenna type correct antenna height

How Does OPUS Work? Data submitted through NGS web page Processed automatically with NGS computers & software Positioned (Tied) to suitable CORS Solution via email (usually in minutes) Don’t forget—it is FREE OPUS works by Submitting data through the NGS Web page It is Processed automatically with respect to 3 CORS 3 independent single base line solutions are performed in the ITRF (International Terrestrial Reference Frame) using the PAGES software and then averaged together. The Same thing but different is done for the Nad83 computations. You can read all about it under the “discussion” button on the OPUS home page If everything is working as it should you can usually expect results in just a few minutes.

How Does OPUS Static Compute Position? OPUS Chooses Three best CORS Then 3 single baselines computed 3 positions averaged — simple mean (equal weights) Differences between positions include any errors in CORS coordinates

OPUS-RS search algorithm 250 km limit <50 km Sort stations in CORS network by distance from rover. Select up to nine CORS that are less than 250 km from rover and that have suitable data. No solution is attempted if fewer than three CORS selected. No solution attempted if distance from rover to polygon enclosing selected CORS is greater than 50 km.

Recent Changes -New Web page and interface -Absolute Antenna Calibrations -MYCS coordinates available

OPUS Static Metadata statistics metadata NAD83 ITRF SPC and UTM

OPUS Rapid Static Obs used >60 quality ind. < 1 might have problem Normalize residual<1 is desirable

What to look for? your antenna type and antenna height are correct orbit used = precise or rapid OPUS-static 90% data used 50% Ambiguities solved RMS< 3cm Peak to Peak < 5cm 3cm horz 5cm vert OPUS-rs Obs used >60 quality ind. < 1 might have problem Normalize RMS <1 is desirable Rapid Static: -No warning messages. -Quality indicators that are suspiciously low Normalized RMS that is suspiciously high. -Coordinate standard deviations that are suspiciously high.

Trouble fixing ambiguities Short session Trouble fixing ambiguities Result: High peak to peak errors (poor solution)

Extended Output Note: Peak to peaks are fairly large; hoz @ 15 cm Vrt @ 23 cm Example of extended output. Solution was from applanix base in middle of median going to Wilmington. Poor visability and examining the PDOP after the fact, a poor window for most of the session.

Running the poor APPLANIX base session again ommitting Castle Hayne

How Can I Improve My Results? -The best way to get more accurate results is to observe longer sessions. -Data sets of at least four hours have been shown to result in more reliable results. -Pre-plan your survey. Make sure your not observing in a poor GPS window. -Set your GPS in an unobstructed area including away from and higher than your vehicle. (preferably use a fixed height pole) -Use an antenna with a ground plane.

Calculating horz/vert accuracies OPUS Static solution report Horizontal positional accuracy calculation at 95% confidence Vertical positional accuracy calculation at 95% confidence VertAccuracy = height peak to peak value

OPUS Rapid Static OPUSrs reports Root Mean Square Error (RMSE) results at 1 sigma (68% confidence interval). Horizontal positional accuracy at 95% confidence: *2 methods--dependent on the RMSE values* #1 if: RMSElat = RMSElong Horizontal accuracy = 1.7308 x RMSE #2. if: 0.2 < RMSEsmaller/RMSElarger <1.0 Horizontal accuracy = 2.4477 x 0.5(RMSElat + RMSElong)

OPUS Rapid Static (cont) OPUS reports Root Mean Square Error (RMSE) results at 1 sigma (68%). Vertical accuracy at 95% confidence: Vertical accuracy = 1.96 x (RMSEorthometric height)

Slide showing database and search

OPUS Projects

OPUS Projects Submit Data to OPUS static Include recovery info, description and photographs Combine multiple observations into a session Perform a least squares adjustment of all sessions

Future developments OPUS NET Will Replace OPUS static Use new algorithm for processing Least Squares instead of mean Different style output Includes usFt in SPC! In testing for over a year, and rumored to go live soon. GNSS Working on adding to both NET and RS

Questions? Questions??? Breath!!!