This repository contains jobs to convert S1-SLCP products to Coherence Differences (CODs) for creating Damage Proxy Maps v1 (DPMv1).
A COD is created based on the below pairing system, using 3 SLCs --> 2 SLCPs --> 1 COD:
Associated job:
This job does network selection / pairing of pre-event SLCPs with the co-event SLCP in the specified AOI to submit to the S1 COD job (Job 2 below) to create CODs
Type: Individual
Facet: Not needed
User inputs:
Fields | Description | Type | Example |
---|---|---|---|
| Suffix dataset tag to append at the end of the dataset id for differentiating events or settings | string |
|
| Project category for possible queue propagation later | enum |
|
| SLCP version to pair CODs for. Default is | string |
|
| Dataset ID of AOI to find SLCPs for CODs creation based on AOI's coordinates, start, end and event time | string |
|
| Optional) Specific track number (only one allowed) of SLCPs overlapping AOI to process CODs for. If not specified, creates CODs for all tracks overlapping AOI. | int |
|
|
Important outputs:
No products will be directly generated, only S1-COD jobs will be submitted based on pairing results. (Expect 1 job for each subswath)
Notes:
The COD network selector pairs SLCPs to submit to the S1-COD
job based on the following workflow:
Gather pre-event_SLCPs
+ co-event_SLCPs
from GRQ. Search criteria:
SLCPs are pre-event if SLCP_end_time < AOI_event_time
SLCPs are co-event if SLCP_start_time < AOI_event_time < SLCP_end_time
Match pre-event_SLCP and co-event_SLCP pairs if:
pre-event_SLCP[‘master_time’] - co-event_SLCP[‘master_time’] < 1 day
area_intersect(pre-event_SLCP,co-event_SLCP) / area(co-event_SLCP) > min_overlap
Creates and returns minmatch
or valid SLCP pairs for COD based on shortest temporal baseline, where
temporal baseline = co-event_SLC['slave_time'] - pre-event_SLCP['slave_time']
Hence, if there are multiple pre-event SLCPs as per the following:
If minmatch
= 1, only COD 1's job will be submitted
If minmatch
= 2, both COD 2 and COD 1 will be submitted
Type: Individual
Facet: Not needed
User inputs:
Important outputs:
Notes:
The CODs in this PGE are computed as such (from burst_coherence_diff.py
):
where COR = Coherence of given co-registered SLCP
=> Positve values correspond to decreased coherence after event and possible damage due to event.