This page describes the processes we use to create the data in our GNSS-IR archive.
Obtaining data
GNSS data and metadata are obtained from a variety of sources, particularly Nevada Geodetic Laboratory and SONEL, but also many other local providers. Some sites require registration or permission to use, so we can't redistribute the GNSS data, but we are trying to make it clear where the data has come from.
Processing sites
Sites are analysed and processed using the procedure described in our introduction to GNSS-IR page. We analyse the incoming reflections and all available metadata, including photos and satellite image, and define satellite elevations and azimuths that produce suitable reflections for extracting water height. We use custom Matlab routines for this step, but in principle they work in the same way as Kristine Larson's open source python library.
Creating data files
Results are processed to create the high frequency files. At this stage we also account for the vertical movement of the water as the satellite passes overhead as described in the file format page - this is done at the same stage as a tidal analysis. Data from each satellite frequency is processed separately due to uncertainties in the exact location of the antenna phase centre.
Calculating daily means
Finally we calculate daily means for use in the daily plot on individual site pages (e.g. Bermuda). The procedure used to calculate these means is illustrated in a Python notebook.
