SWOT Gravimetry data

[Esteban82]

I'd like to propose replacing the current @earth_faa dataset served with the new SWOT-based gravity grid released by David Sandwell.
The new grid (grav_SWOT_03.nc) is now available from the TOPEX FTP, and David confirmed via email that we should indeed replace the old dataset with this one.

I can do it. The only issue is that there is no source grid to update the FAA Uncertainty. So I should ask David for that grid or remove that dataset from GMT Remote Data sets. I would choose to remove it.

[joa-quim]

Yes, this looks a good update.

[anbj]

Yes, great!

[Esteban82]

Ok, I will do it. I asked David for the uncertainty grid but there is no one yet.

[anbj]

Let me know if you need any help, I'll be glad to.

[Esteban82]

I have some doubts how to set these two values (for the recipes). Specially the SCALE. (I have to change them because the range is around twice bigger).

# DST_SCALE=0.025
# DST_OFFSET=300

Is there a place where it is explained in detail?

[seisman]

What's the z-range of this dataset?

[Esteban82]

The range of -712.481506348 to +942.2193604 mGal

[Esteban82]

In the recipt it says # Scale is also chosen so that 1/scale is a finite-decimal number (here 0.05 = 1/20)

So, I know I can use 0.0625 (1/16). My main doubt is if I use a scales like 1/17, 1/18 or 1/19. The three has infinite decimal numbers.

1/16=0.0625
1/17=0.05882352941176470
1/18= 0.05555555555555560
1/19=0.05263157894736840
1/20=0.05

[joa-quim]

From your comments alone (did not check the recipe), I think the scale should be 1/40

(942.2193604 + 712.481506348) / 2^16 = 0.02524873148724365

and

1/0.02524873148724365 = 39.60595012486973

[Esteban82]

Ok, I will try.

Another question. Do you agree that this is the proper title of the dataset? I just add SWOT.
# SRC_TITLE=IGPP_Earth_SWOT_Free_Air_Gravity_Anomalies_v3

[joa-quim]

It seems a bit long but don't know were that is used.

Note that the DST_OFFSET may need to be changed too.

[Esteban82]

From your comments alone (did not check the recipe), I think the scale should be 1/40

(942.2193604 + 712.481506348) / 2^16 = 0.02524873148724365

and

1/0.02524873148724365 = 39.60595012486973

It didn't work. I think I have to

Convert /tmp/32029/grav_SWOT_03.nc to earth/earth_faa/earth_faa_01m_p.grd=ns+s0.025+o300
grdconvert [ERROR]: Cannot write format ns = GMT netCDF format (16-bit integer), CF-1.7.
grdconvert [ERROR]: The packed z-range, [-40499,25689], exceeds the maximum representable size. Adjust scale and offset parameters or remove out-of-range values.
grdconvert (gmtapi_export_grid): NetCDF: Numeric conversion not representable [earth/earth_faa/earth_faa_01m_p.grd=ns+s0.025+o300]

What is wrong with packed z-range? What is the maximum representable? Should I modify the offset, the scale or both?

[joa-quim]

The maximum representable size in 16bits is (if I'm not mistaken with the signed part) [-32768 32767]. That is why the scale must be 1/40 and the offset must be adjusted too. Not sure what is the value that should be used. Something that when we revert the scalling operation gets us the minimum grid value.