Simulations of direct and reflected wave trajectories for ground-based GNSS-R experiments

Roussel, N.; Frappart, F.; Ramillien, G.; Darrozes, J.; Desjardins, C.; Gegout, P.; Pérosanz, F.; Biancale, R.

doi:https://doi.org/10.5194/gmd-7-2261-2014

Articles | Volume 7, issue 5

https://doi.org/10.5194/gmd-7-2261-2014

© Author(s) 2014. This work is distributed under
the Creative Commons Attribution 3.0 License.

https://doi.org/10.5194/gmd-7-2261-2014

© Author(s) 2014. This work is distributed under
the Creative Commons Attribution 3.0 License.

Articles | Volume 7, issue 5

Model description paper

|

02 Oct 2014

Model description paper |

| 02 Oct 2014

Simulations of direct and reflected wave trajectories for ground-based GNSS-R experiments

N. Roussel, F. Frappart, G. Ramillien, J. Darrozes, C. Desjardins, P. Gegout, F. Pérosanz, and R. Biancale

Download

Final revised paper (published on 02 Oct 2014)
Preprint (discussion started on 24 Jan 2014)

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC C132: 'review', Felipe Nievinski, 17 Mar 2014
- AC C1031: 'Answer first review', Nicolas Roussel, 30 Jun 2014
RC C478: 'Review', Anonymous Referee #2, 02 May 2014
- AC C1032: 'Answer second review', Nicolas Roussel, 30 Jun 2014

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

AR by Nicolas Roussel on behalf of the Authors (30 Jun 2014) Author's response Manuscript

ED: Referee Nomination & Report Request started (03 Jul 2014) by Robert Marsh

RR by Felipe Nievinski (14 Jul 2014)

ED: Reconsider after major revisions (16 Jul 2014) by Robert Marsh

AR by Anna Wenzel on behalf of the Authors (13 Aug 2014) Author's response

ED: Referee Nomination & Report Request started (21 Aug 2014) by Robert Marsh

RR by Felipe Nievinski (31 Aug 2014)

Suggestions for revision or reasons for rejection

Authors are to be commended for their diligent work over the past revision cycles. I find that no major issues remain outstanding in the latest version. I do take the time to indicate a few issues for their consideration. I do not need to see the final version prior to acceptance. Congratulations!

%%%%%%%%%%%%%%%%%
MAJOR
(none)

%%%%%%%%%%%%%%%%%
MODERATE

Authors state that "[the ellipsoid] has been expanded..." The osculating sphere needs to be modified as a consequence of this ellipsoidal expansion. More specifically, the ellipsoid radius of curvature shall be calculated based on the modified ellipsoidal semi-major axes values. This should improve the agreement with the ellipsoidal results slightly.

As a separate issue, and for improved reproducibility, please further clarify how this expansion is achieved. Authors state that "[the ellipsoid] has been expanded until the ellipsoid height of the receiver equals the height of the receiver above the reflecting surface". This is equivalent to expanding the ellipsoid until its surface coincides with the reflecting surface (at nadir of the receiver), which might be called the surface base point. Yet no equations are given. I assume the ellipsoid remains geocentric and its axes are scaled. It'd seem that the modified axes (say, a' and b') cannot be found simply by summation of the original axes (a, b) and the ellipsoidal height (h) of the base point (not of the receiver itself). Rather, there exists a condition to satisfy, that the Cartesian coordinates of the base point (X,Y,Z) remain unchanged, when computed either from the original geodetic coordinates (lamba,phi,h) and ellipsoid constants (a,b) on the one hand, or their modified values (lambda',phi',h';a',b') on the other hand, where lambda'=lambda is the longitude, phi'=phi is the latitude, and h'=0. The modified ellipsoid constants are, then:
a' = (a^2 + h^2 + h*c + a^2*h/c)^0.5 = a*(1 + h^2/a^2 + h*c/a^2 + h/c)^0.5
b' = (b^2 + h^2 + h*c + b^2*h/c)^0.5 = b*(1 + h^2/b^2 + h*c/b^2 + h/c)^0.5
where c = (a^2*cos(phi)^2 + b^2*sin(phi)^2)^0.5
As a check, if h=0, then a'=a and b'=b for any latitude phi.

It wouldn't hurt to stress in the text that this ellipsoidal expansion is only done once as long as the receiver position remains unchanged w.r.t. the reflecting surface; it is redone if the reflecting surface changes (e.g., tidal waters) but it is not redone with changes in the satellite direction.

I find it unnecessary to include ellipsoid vs. plane results in the table and figures, as they are essentially the combination of other already reported results (sphere vs. plane and sphere vs. ellipsoid, dominated by the former).

%%%%%%%%%%%%%%%%%
MINOR

The abstract speaks of "mean deviations" but I suspect the values are maximum absolute deviation, which is what is given in the tables. Please correct. Otherwise, clarify the mean depends on how the satellite elevation angle is sampled (more densely near zenith or near the horizon).

Fig 10: please explain why the mean of differences sometimes is correlated with the tide and other times it is anti-correlated (their arithmetic signs are not always the same).

Fig 11-14: why there is dispersion within results for a fixed elevation angle and fixed pair of algorithms (e.g., multiple green dots intersecting a given vertical line)? Is it a consequence of what source: azimuthal variability in the ellipsoidal radius of curvature (Euler's formula); time-variable ellipsoidal expansion (tidal waters); troposphere correction (I'd recommend disabling it in these plots and keep it enabled only in Fig 16).

Results for the planar Earth model could be mentioned in the abstract.

Hide

ED: Publish subject to technical corrections (31 Aug 2014) by Robert Marsh

AR by Nicolas Roussel on behalf of the Authors (12 Sep 2014) Author's response Manuscript