Search

JPL's newly developed software for determining terrestrial reference frames, known as SREF (Square-root Reference frame Estimation Filter), has been used to produce JTRF2020, a combined terrestrial reference frame determined from the input SINEX files submitted by the IVS, IGS, ILRS, and IDS for ITRF2020. SREF, being based on a square-root information filter and smoother, determines the reference frame sequentially from the input station position time series. For JTRF2020, 525 GNSS, 194 DORIS, 116 SLR, and 96 VLBI, or a total of 931, station position time series were used. Incorporating process noise in SREF, determined from geophysical fluid loading models, allows the observed station positions to be smoothed between discontinuities caused by earthquakes and equipment changes. Reference frames determined by SREF, like JTRF2020, are represented by this set of smoothed station position time series. SREF also fits a model to the station's observations and uses the model to fill gaps in the station's observing history, to forecast the position of the station after it stopped observing, and to hindcast its position before it started observing. For JTRF2020, the model consists of a piecewise linear trend, an annual periodic term, and a sum of exponential terms to represent postseismic motion for those stations that experience postseismic motion. The result of using SREF to determine JTRF2020 will be presented.

JTRF2020 is the latest TRF solution computed at JPL by assimilating frame input data submitted by the IVS, IGS, ILRS, and IDS for ITRF2020. Determined with SREF (Square-root Reference frame Estimation Filter), a computational code based on a square-root information filter and Dyer-McReynolds smoother algorithm, JTRF2020 adopts a time-series based representation: Its daily time series determined from observations of a multi-technique space-geodetic network of 931 stations describe, through their Cartesian coordinates, the deformation of solid Earth as a function of time. Such coordinates are inherently expressed in Earth’s Center of Mass as sensed by SLR and give access to the quasi-instantaneous scale implied by the underlying frame. JTRF2020, like its predecessor JTRF2014, traditionally adopts a scale relying on VLBI and SLR observations only. In this presentation, we will guide the readers through the process we adopted to define and construct the scale of JTRF2020, from the determination and analysis of the scale differences between VLBI and SLR to the way in which the time-variable scale bias between VLBI and SLR is handled within the J2020 assimilation.

Earth orientation parameters (EOPs) are currently determined from measurements taken by the space-geodetic techniques of SLR, VLBI, GNSS, and DORIS. But each technique has its own unique strengths and weaknesses in this regard. Not only is each technique sensitive to a different subset and/or linear combination of the EOPs, but the averaging time for their determination is different, as is the interval between observations, the precision with which they can be determined, and the duration of the resulting EOP series. By combining the individual series determined by each technique, a series of the Earth's orientation can be obtained that is based upon independent measurements and that spans the greatest possible time interval. Such a combined Earth orientation series is useful for a number of purposes, including a variety of scientific studies and as an a priori series for use in data reduction procedures. However, care must be taken in generating such a combined series in order to account for differences in the underlying reference frames within which each individual series is determined (which can lead to differences in the bias and rate of the Earth orientation series). Traditionally, differences in the underlying reference frames are accounted for by applying a correction to the bias and rate of each individual series being combined with the goal of placing the series within a common reference frame. But there is an uncertainty associated with estimating the bias and rate correction that needs to be applied to each series. In fact, this uncertainty is a major (if not the major) source of error in combined EOP series. However, this source of error can be mitigated by jointly combining the EOP series with the terrestrial reference frame. Recent ITRF and JTRF solutions such as ITRF2020 and JTRF2020 have included EOP series in their determination. In this presentation, the ITRF2020 and JTRF2020 combined polar motion series will be compared to the more traditionally determined IERS Bulletin A and JPL KEOF (Kalman Earth Orientation Filter) combined polar motion series in order to study the improvement attained by jointly determining the combined EOP series with the terrestrial reference frame.

JTRF2020 is a recent contribution by NASA Jet Propulsion Laboratory to IERS as part of their effort to determine ITRF2020. JTRF2020 is estimated daily using a sequential computational method (called "SREF") based on the SRIF and DMCS algorithms which are respective variants of the Kalman filter and smoother. The filtering algorithm allows us to make timely update of the reference frame, as newer geodetic analysis data become available beyond those ingested into JTRF2020. Benefits of such frame updating are examined using hindcasting experiments, which are presented here. Also, some of the standard formulas used in terrestrial reference frame realizations are not immediately suitable for sequential formulation required by the algorithms like the Kalman filter. We thus describe our sequential re-formulation of these formulas, including the intrinsic constraints, week-long EOP arcs, and lack of initial conditions.

JPL's newly developed software for determining terrestrial reference frames, known as SREF (Square-root REference Frame filter), has been used to produce JTRF2020, a combined terrestrial reference frame determined from the input SINEX files submitted by the IVS, IGS, ILRS, and IDS for ITRF2020. SREF, being based upon a square-root information filter and smoother, determines the reference frame sequentially from the input station position time series. Incorporating process noise in SREF, determined from geophysical fluid loading models, allows the observed station positions to be smoothed between discontinuities caused by earthquakes and equipment changes. Reference frames determined by SREF, like JTRF2020, are represented by this set of smoothed station position time series. SREF also fits a model (consisting of a piecewise linear trend, annual and semi-annual periodic terms, and a sum of exponential terms to represent postseismic motion) to the station's observations and uses the model to fill gaps in the station's observing history, to forecast the position of the station after it stopped observing, and to hindcast its position before it started observing. The result of using SREF to determine JTRF2020 will be presented.

The 1754 incident precipitating the French and Indian War took place over a disputed claim between France and England for the Ohio River Valley. A young George Washington commanded a Virginia militia company to the Forks of the Ohio at the site of present-day Pittsburgh. The French duly rousted him and his men. The French had established Fort Duquesne, the latest in a series of forts on the Allegheny River built by them to protect their claim. They based their authority on the 1669 discovery of the Ohio River by an obscure former Jesuit priest, Robert Cavelier, Sieur de La Salle. In the absence of a Procèss Verbal documenting his discovery, the legitimacy of France's claim has long been disputed. New research and insights into somewhat confusing primary sources bear out La Salle's discovery and the primacy of France's claim.