KSA geoid preparation
### LSC computation with data selected in 0.02 d dataa cells.

Readme.

Error estimates.

Scaled error estimates.

Differens wrt FFT..

Diff. ARAMCO.

### Comparisons sea-gravity with gravity model

Differences (mgal).

### Comparison with ARAMCO GPS/levelling.

Id.

Height1

Height1

### Comparison fft - lsc (using 0.01 d data) in 4 areas.

Plots of diferences and gravity stations.

NW area.

NE area.

SW area.

SE area.

Note that the differences are only large where there are no or
little data.

### Error estimation with airborne gravity added.

Using LSC with reduced data close to 0.025 deg. grid points (82569) error estimates
of a (quasi) geoid was computed, see
Figure

The errors are largest where there are no or sparse data or where the gravity variation
is larger than the one used in the covariance modelling.

Grids of Gravity anomalies were generated at 2500 m altitude, covering an area bounded by
25 deg to 28 deg in latitude and 36 deg to 43 in longitude. The spacing in longitude was
in all cases 0.05 deg. In latitude the spacing was 0.2, 0.25, 0.5 and 1.0 deg, respectively.
File names are ksa_airborneX.XX.dat

All error estimates are in the area bounded by 24.0 29.0 35.0 44.0 in a 0.2 deg grid
(at zero altitude) for all 4 examples. The computation is quite easy, since the 82400
first reduced equations are re-used.

The results using a smooth field can be seen from

No airborne .

0.2 deg. spaced .

0.25 deg. spaced .

0.5 deg. spaced .

1.0 deg. spaced .

The calculations were then repeated corresponding to the montaneous NE
area at an altitude of 3.5 km.
A new covariance model was determined.

No airborne .

0.1 deg. spaced .

0.1 deg. spaced
with sea-gravity (5 mgal error assumed) .

0.2 deg. spaced .

0.25 deg. spaced .

0.5 deg. spaced .

1.0 deg. spaced .

### Sea gravity files.

gravsoft format

geodas header

geodas format

Last update 2014-08-28.