Sentinel-1 Extended Timing Annotation Processor (SETAP)
The Extended Timing Annotation Dataset for Copernicus Sentinel-1 (S1-ETAD) is a new auxiliary product developed by ESA (with DLR as contractor), providing users with corrections to improve geometric accuracy of Sentinel-1 (SM and IW) SLC images to centimetric levels.
The product contains analysis-ready layers for removing the atmospheric path delays, the solid Earth tidal deformation, and the Sentinel-1 system specific effects related to the IPF SAR processor (Doppler shift, accurate bi-static correction and height dependent FM-rate mismatch).
To generate such corrections, the SETAP service uses precise information from different ancillary data sources such as IGS and ECMWF.
EO sources supported:
Sentinel-1 Annotations: Sentinel-1 L1A products containing only the XML annotations. The information contained in XML annotation is the only one used by the S1-ETAD processor. The advantage of using S1-Annotation products is that they have a size (in MB) by far smaller than standard ones so they can be downloaded faster.
Please refer to the Sentinel-1 Annotations caching policy section below for more details on characteristics and limitations of this EO source.
Output specifications
This tutorial describes how to submit a processing job for the SETAP service to obtain one or more S1-ETAD products.
The tutorial is addressed to users already familiar with Geohazards thematic Exploitation Platform (GEP), and gives some hints and recommendation for the best service usage experience.
The main user actions are the following:
During the selection of the input data for your processing, take into account that, internally, the processing itself is organized as follows:
Please also consider that, currently, the maximum number of products per job is 35. To process a larger number of input products, it is possible to run multiple jobs but it is strongly recommended to feed all the products belonging to a same data-take to the same job.
Example 1
A job is started with 35 S1-SLC input products belonging to the same data-take (possibly incomplete).
In this case only one S1-ETAD product is generated.
Example 2
The job is started with 35 S1-SLC input products belonging to an interferometric stack (single slice).
In this case each input product belongs to a different data-take and the service generates 35 S1-ETAD products.
Example 3
The job is started with 30 S1-SLC input products belonging to an interferometric stack covering a wide area. 3 SLC slices are necessary to cover the requested area.
In this case the input products are grouped in 10 different groups and the service generates 10 S1-ETAD products.
Once the Product (url) list field is filled with the selected data, it is possible to set the remaining parameters.
Please note that all the remaining parameters have a default value that should be perfectly fine for practically all cases. Nevertheless it is possible to tweak them for very specific purposes.
If set to true this parameter specifies that the global ECMWF dataset have to be used for troposphere computation. By default the value of this parameter is set to false to indicate that only the relevant region of the global ECMWF dataset covering the requested area shall be used for the processing. In this case the data download is typically by far faster. The computation of the data region covering the requested input products is performed automatically by the application.
This parameter can be used to tweak the amount of margin to be applied to the S1 data footprint when the Copernicus DEM is ingested for the processing.
By default the Orbit type parameter is set to AUX_POEORB meaning that orbit with the maximum available accuracy are used for the computation. Precise Orbits are typically available 20 days after the S1 data acquisition.
It is strongly recommended not to change this setting.
Orbit type parameter can also be set to AUX_RESORB that have a lower nominal accuracy w.r.t. to AUX_POEORB, but are typically available 1-2 days after the acquisition.
Important
Support for Restituted Orbits (AUX_RESORB) is considered experimental.
Once the job execution has been completed, it is strongly recommended to share results with other GEP users.
Please refer to the Sharing section for detailed instructions about how to share jobs and the produced ETAD products.
How to use the S1-ETAD products
Basic information on the S1-ETAD product and how to use it are provided in the Product Definition Document. Details on the file format and on the algorithms used to generaate it can be found in the Format specification document and the Algorithm Theoretical Baseline Document respectively.
A Python API that allows to easily access the product and perform basic operations is also available.
In general all main information and news about the S1-ETAD product are collected in the Product Page on the ESA web site.
Sentinel-1 Annotations caching policy
The “Sentinel-1 Annotations” products (cf. “EO Data” menu) benefit from a data caching mechanism on GEP. The cache (Manifest and Annotation files) is managed for a pre-defined set of geographic areas and time span. These areas are specified according to the seismic areas defined by the CEOS Seismic Hazards work group, as follows:
For the launch of the GEP SETAP service (September 2021), the cache is initialised as follows:
Moreover, when a processing Job is started over a spatial and temporal data search outside of the current data caching automation, an opportunistic approach will keep in cache the retrieved input products.
This is resulting in an additional cache capacity, mutualised, that is growing based on user activities (so, without a planification on the GEP side, but with a monitoring of the user interests).
User selected job inputs falling outside current cached products, for spatial and temporal domains with some limitations (see further below the FAQ section, Q1).
For the future of SETAP operations on GEP, a plan with progressive caching based on these CEOS tectonic areas can be achieved according to user needs gathered through user engagement activities. For example:
In terms of performances of a processing job, SETAP will be the fastest when accessing pre-cached input products, a little bit slower if products not yet in cache can be accessed from the Copernicus SciHub, and quite slow for older products only available from other remote data repositories (e.g. the Alaska Satellite Facility) connected to GEP.
Users can provide clear information about their area of interest and time period of interest, sending requests as follows:
Then, send it to s1-etad@esa.int with subject:
Study requirements for optimised data access from the SETAP processing service
FAQ
| Q1: | Are there limitations in terms of geographic areas, product type, acquisition date, maximum number of jobs per user? |
|---|---|
| A1: | Main service limitations are listed below:
|
| Q2: | Do you need KML files for the study regions we are using or is the service global? |
| A2: | The service is global. The area of interest for data selection can be specified by the user by means of the Web UI (see Discover data). If the user already has a KML file for its Area Of Interest (AOI), it can be uploaded in the Web UI to select the AOI (see the “Spatial Filters” section in Geobrowser). |
| Q3: | Which periods are you planning to provide ETAD, or will it be done for new S1 acquisitions in NRT? |
| A3: | NRT product generation is not supported, also, there is no systematic processing on GEP. The product generation have to be triggered by the user for any date compatible with the the limitations described in A1. |
| Q4: | How to apply ETAD corrections to an S1 SLC product? |
| A4: | Basic information about how to apply ETAD corrections are provided in the Product Definition Document (see also the How to use the S1-ETAD products section). |
