The focus of my GPS research relates to understanding unmodelled systematic errors in GPS, including those at sub-daily periods, longer-period errors on short baselines, and ionospheric and tropospheric signals/corrections.
Active Projects
Sub-daily systematic errors: it is now well known that sub-daily systematic errors propagate to longer periods, such as semi-annual, annual and fortnightly, in conventional 24 h batch solutions. Recent work has investigated the level of this propagation by comparing kinematic PPP solutions with conventional 24 h PPP solutions. Time-variable sub-daily signals were observed, and these propagated, in part, into annual and semi-annual signals with amplitudes of ~0.5-1.0mm. Depending on the network geometry, they may bias estimates of low-degree spherical harmonics at the 5-10% level.
Long-period systematic errors: aside from propagated signals mentioned above, long-period errors also exist as is seen in long-running short baselines. Ten such baselines have been studied and each show a surprising level of one or more of annual, quasi-secular and undocumented jumps
This work is being led by my current PhD student, Liz Petrie. Liz has coded higher-order ionospheric corrections into GAMIT and is in the process of performing a global reanalysis (1996-2007) to examine its effects on coordinate time series and GPS-defined reference frame parameters
Precipitable Water Vapour (PWV) from GPS
This work is being led by my recent PhD student (and now postdoc), Ian Thomas and focuses on Antarctica, analysing long time series from our own global reprocessing in GAMIT (1995-2006) and comparing to radiosonde and recent satellite-derived (AIRS, MODIS, AMSR-B) products
Work on airborne positioning has focused on application to long-baseline positioning of lidar instruments in Svalbard. Long-baseline accuracy tests have recently been conducted and show that, with parameterised tropospheric zenith delay (TZD), kinematic GPS positioning precision is near-constant (~0.15m) for baselines from ~20 km to several hundred km. Biases remain near-zero for all baseline lengths. Solutions using commercial software where TZD is not parameterised degrade to a precision of ~0.5m for baselines longer than 50 km and exhibit elevation-dependent bias. Further work is required to rigorously identify the various error components of airborne lidar and GPS-related errors, notably tropospheric and multipath errors, remain dominant.
Recent Paper Highlights
King, M.A., C.S. Watson, N.T. Penna, and P.J. Clarke. 2008. Subdaily signals in GPS observations and their effect at semiannual and annual periods, Geophys. Res. Lett., 35, L03302, doi:10.1029/2007GL032252. Limited electronic copies are available at http://www.agu.org/journals/gl/gl0803/2007GL032252/ using login&password: 61193892
Penna, N., M. King, and M. Stewart. 2007. GPS height time series: short period origins of spurious long period signals, Journal of Geophysical Research, 112, B02402, doi:10.1029/2005JB004047.
This site was last updated
27-Jul-2010