No. GAPS is a free online tool offered by the University of New Brunswick which is available to the general public free of charge and without any sort of registration.

GAPS accepts submission of files in both RINEX 2.xx or 3.xx formats, Hatanaka compressed files, Receiver RAW files, and several compressed packages. Please note that we do impose a maximum file upload size of 20 Mb. Check our "News" or "Best Practices" pages for more information.

Please do not submit any single-frequency observation files or navigation files for any constellation. GAPS strictly handles dual-frequency GNSS observation files.

There is no limit in terms of total observation time, however, the quality of the solution generally depends on the duration of observations. Check our "Best Practices" page for more information.

This most likely signifies that the orbit/clock product selected was unavailable at the time of the file submission. Please remember that IGS Final orbits and clocks are provided at a latency of 12-18 days, the Rapid products at a latency of 17-41 hours, and the Ultra-Rapid products (that we utilize) at a latency of 6 hours. Typical product latency for IGS MGEX products (which are used for Galileo and/or Beidou processing) is far less predictable, but is usually about 12-18 days.

In this instance, the observable type that was selected is most likely not available within the submitted RINEX observation file. Please make sure that your receiver has been set to track the desired observable type.

No. If a-priori coordinates are not provided, coordinates from the RINEX observation header are used as a priori. If the RINEX file also doesn't provide approximate coordinates, a general point positioning algorithm is utilized within GAPS to provide approximate coordinates. If none of the above are able to provide a reasonable position, an a priori position of (0,0,0) is used.

If a constraint is imposed on the coordinates, then the position will be constrained at the level provided by the user. If no constraints are imposed (i.e., zero was used), the coordinates are used only as approximate coordinates. In the case of unconstrained coordinates, the final solution should always be the same, no matter what a priori values have been used. That said, for ocean tidal loading, the a priori coordinates will be used to determine if there are local tidal coefficients available on our server. If you wish to use our coefficients, a priori coordinates will be required, although they must be within 1000km of a published IGS station's coordinates.

No. If the default option (00:00:00 to 23:59:59) is used, GAPS will process any observation within this time period. The time window option should be used to limit the time to less than what was observed (e.g., to process a specific hour of a 24 hour file).

Partially. Currently, GAPS only supports files with up to 24 hours of observations, realized during the same day (in GPS time), or spanning for a maximum of 2 hours after midnight.

If you didn't receive an e-mail back, the server might be too busy to work on your request immediately. Please try to submit the file again later. If you fail to receive a response upon your second attempt, please contact us at gaps@unb.ca

Within this email response, we have provided you with a unique series of numbers that we use to identify the processing failure. In order to inquire about the failure, please reply to the email remembering to include this series of numbers and we'll look into it for you!

If you open the HTML output file and none of the plots successfully load, this most likely means that a) your browser does not support the file type of the plots or b) the DOP threshold used in processing eliminated a majority of the submitted observations as bad and there was simply not enough data left to provide reasonable output. In the case of the latter, please try to resubmit your observation file using a larger DOP threshold.

In order to properly reference GAPS, please use the most recent citation found on our "PUBLICATIONS" page.

GAPS uses an a-priori value for the neutral atmosphere delay, predicted from ray-tracing through Numerical Weather Prediction models produced by the Canadian Meteorological Centre (CMC), National Centers for Environmental Prediction (NCEP), and European Center fore Midrange Weather Forecasting (ECMWF). These values are provided on a global grid where the input station is interpolated. (see http://mars.hg.tuwien.ac.at/~ecmwf1/ or http://unb-vmf1.gge.unb.ca/). The value of this delay is improved with the estimation of a residual delay, which added to the predicted delay gives the total delay estimated by GAPS. The 'A-priori NAD standard deviation' option allows the user to choose what confidence GAPS should put on the predicted delay. The total delay is computed for every epoch of data processing, as a random walk process. The 'NAD process noise' level allows the user to choose how much GAPS should allow the estimate of the total delay to change over time. The user can set GAPS to not estimate residual delays and therefore rely only on the CMC, NCEP OR ECMWF value, by choosing a very low a-priori standard deviation and setting the process noise to zero. In the case that the NCEP, CMC, OR ECMWF data is not available for a given date, GAPS reverts to using the UNB3m model. (see http://gge.unb.ca/Resources/unb3m/unb3m.html).

In most cases the troposphere is assumed to be symmetric in azimuth. This implies that the delay for an observation taken at an azimuth of 0 degrees and an observation at the same elevation angle but at an azimuth of 180 degrees would experience the same delay due to the troposphere. While this is a good approximation for most of the sky, for observations below ~20 degrees, this is not the case. Tropospheric gradients are a way to model the asymmetry or azimuth-dependence of the tropospheric delay. As is the case for the tropospheric zenith delay, the gradient parameters are most often estimated from the GPS observations themselves. In this case two parameters are estimated, one for the north-south direction (Gn) and one for the east-west direction (Ge). The a-priori value for these parameters are typically left at zero, while the a priori standard deviation is set to 1 cm. Once again, a random walk model is applied which determines how much these parameters can vary over time. A common value for this is 0.3mm/sqrt(hr). Although not normally used in kinematic processing, tropospheric gradients have been shown to systematically improve the accuracy and repeatability of station positions when included in static processing. They can also be a useful meteorological parameter.

We provide this option mainly for those users who are submitting high-altitude observations as we assume that these observations are made near or above the limits of the neutral atmosphere, thus negating the requirement for this correction. Please note that we also impose a cutoff of neutral atmosphere delay corrections once a receiver has achieved an altitude of greater than 15,240 metres (50,000 feet).

In order to obtain integrated water vapor estimates, simply provide us with an appropriate meteorological file relevant to the station for which you are submitting observations. We will use the information in this file along with the estimated neutral atmosphere delay to determine the integrated water vapor.

In order to apply custom antenna calibrations, please follow the example GAPS ANTEX file format found within the link under "User Receiver Antenna Calibration" on our Advanced User submission page. Please note that this example format must be followed exactly for us to apply your custom antenna calibrations.