The Sea Level Anomaly (SLA) variable has been present in the DUACS products for many years. It is obtained from the altimeter measurements after applying various geophysical corrections as discussed in the DUACS processing page. This means that different geophysical signals are removed from the measurements. They include, omong other,

• Ocean tide signal: it it corresponds to the barotropic tide frequencies
• Internal tide signal: it corresponds to the coherent part of the baroclinic tide (phase-locked with barotropic tide freq.).
• Dynamic Atmospherical (DAC) signal: its objective is to remove high frequencies signals forced by the atmosphere (pressure and wind) and aliased in altimetric data because of bad temporal sampling of altimeters. Note : At present time S1 (24h signal) and S2 (12h signal) atmospheric signals are removed with a climatology in the DAC processing à DAC contains only the residual stochastic S1S2 signal. The gravitational part, and also radiational part of S1S2 signal is included in ocean tide correction.

Additionally, a residual Long Wavelength Error (LWE) signal is also removed from altimeter measurement. The LWE is an empirical correction that aims at removing residual ocean tide and DAC signal as well as residual orbit error (residual signals induced by the imperfection of the solution used for these corrections)

 

Whereas the physical content of the SLA variable delivered is useful for many applications, it may not fit specific needs. For this reason, different physical variables are available in the DUACS L3 products (see table below),  allowing the users to easily modify the physical content of the SLA. For instance, a SLA field including the ocean tide effects, internal tide and the DAC effect can be obtained that way:

[sla not corrected from ocean tide and dac effect ] = [sla from L3 product]+[dac]+[ocean_tide]+[internal_tide]-[lwe]

The user can also decide to include in the SLA field only the ocean tide effect and/or the DAC effect.

Note that the LWE correction is not a geophysical correction. However, as it aims at reducing the residual signal of dac and ocean tide (see discussion above), we recommend to use it when reintroducing dac and ocean_tide in the SLA field.

The L3 product also includes a “unfiltered” SLA field, i.e. a SLA including  noise measurement that dominates the signal at short wavelenghts . At the opposite, the “filtered” SLA is obtained after low-pass filtering in order to reduce noise measurement effects  (see DUACS processing). As filtering processing can affect the physical signal, we recommend using the “unfiltered” SLA field when combined with other variables to modify the physical content.

A Mean Dynamic Topography (MDT) field is also included. It can be used to compute the Absolute Dynamic Topography (ADT) as discussed in the altimeter reference period page.

 

Table: varius physical variables included in the DUACS L3 products

Physical variable name	description
sla_unfiltered	Raw Sea Level Anomaly (i.e. not filtered). SLA referenced to the [1993, 2012] period
sla_filtered	Sea Level Anomaly, filtered from high frequency signal considered as noises (see DUACS processing). SLA referenced to the [1993, 2012] period
dac	Dynamic Atmospheric correction
ocean_tide	Ocean tide signal
internal_tide	Internal tide correction
lwe	Long wavelength correction
mdt	Mean dynamic topography referenced to the [1993, 2012] period
tpa_correction	TOPEX-A instrumental drift correction derived from altimetry and tide gauges global comparisons (WCRP Sea Level Budget Group, 2018). See dedicated section for more infos.

 

Recommendations for modellers:

Modellers wishing to assimilate DUACS altimeter products and/or use them to compare with modelled sea levels can adjust the physical content of the altimeter measurement to be comparable with the modelled field. The following table gives some recommendations.

Model forced by wind only	Model forced by wind & pressure	Model including tides	Model forced by wind & pressure & includes tides
You should reinject in the SLA from altimetry the high frequency variations induced by wind forcing	You should reinject in the SLA from altimetry the high & low frequency variations induced by wind & pressure forcing	You should reinject in the SLA from altimetry the tidal signal	You should reinject in the SLA from altimetry the high & low frequency variations induced by wind & pressure forcing
SSHmodel  ~  adtalti = sla_unfiltered + (dac – ib_lf) – lwe + mdt Note: here we have an imprecision since we use only the low-frequency part of the IB : we should in fact use the full IB content	SSHmodel  ~  adtalti = sla_unfiltered + dac – lwe + mdt	SSHmodel  ~  adtalti = sla_unfiltered + ocean_tide + iw – lwe + mdt	SSHmodel  ~  adtalti = sla_unfiltered + dac + ocean_tide + iw – lwe + mdt