Determine path forward for Sentinel-1 processing

[jhkennedy]

~Right now, processing Sentinel-1 ITS_LIVE velocity granules costs ~$0.20 / pair whereas Optical products are generally $0.01 or less. Accordingly, e have only processed 6 and 12 day pairs through Dec. 31, 2022, and thus have an almost 2 year gap in the data record right now, and no ability to do low-latency production.

Ideally, we could get S1 processing down to ~$0.05/pair for V2, fill the product record, and turn on low latency production.

To do that, we need to determine which processing option is the best path forward. There are a couple options:

1. Adapt the S1 correction workflow to process SLC pairs
2. Adapt the S1 correction workflow to process SLC BURST pairs
3. Adapt the S1 correction workflow to process GRD pairs
4. Migrate the ISCE2 workflow to ISCE3 to process SLC pairs
5. Migrate the ISCE2 workflow to process SLC BURST pairs
6. Migrate the ISCE2 workflow to ISCE3 to process SLC BURST pairs
7. Stick with what we've got and deal with it.

Can start with this pair:

S1A_IW_SLC__1SSH_20170221T204710_20170221T204737_015387_0193F6_AB07
S1B_IW_SLC__1SSH_20170227T204628_20170227T204655_004491_007D11_6654

References:

1. https://www.mdpi.com/2072-4292/13/4/749: Yang's original autoRIFT paper, which contains the scenes he used for validation
2. https://essd.copernicus.org/articles/14/5111/2022/essd-14-5111-2022.html: Sentinel-1 processing methodology

[jhkennedy]

For the SLC pair above, here are the associated burst pairs:

S1_191586_IW2_20170221T204734_HH_AB07-BURST S1_191586_IW2_20170227T204652_HH_6654-BURST 
S1_191586_IW1_20170221T204733_HH_AB07-BURST S1_191586_IW1_20170227T204651_HH_6654-BURST 
S1_191585_IW3_20170221T204732_HH_AB07-BURST S1_191585_IW3_20170227T204650_HH_6654-BURST 
S1_191585_IW2_20170221T204731_HH_AB07-BURST S1_191585_IW2_20170227T204649_HH_6654-BURST
S1_191585_IW1_20170221T204730_HH_AB07-BURST S1_191585_IW1_20170227T204648_HH_6654-BURST 
S1_191584_IW3_20170221T204729_HH_AB07-BURST S1_191584_IW3_20170227T204647_HH_6654-BURST 
S1_191584_IW2_20170221T204728_HH_AB07-BURST S1_191584_IW2_20170227T204646_HH_6654-BURST 
S1_191584_IW1_20170221T204728_HH_AB07-BURST S1_191584_IW1_20170227T204645_HH_6654-BURST 
S1_191583_IW3_20170221T204727_HH_AB07-BURST S1_191583_IW3_20170227T204645_HH_6654-BURST 
S1_191583_IW2_20170221T204726_HH_AB07-BURST S1_191583_IW2_20170227T204644_HH_6654-BURST 
S1_191583_IW1_20170221T204725_HH_AB07-BURST S1_191583_IW1_20170227T204643_HH_6654-BURST 
S1_191582_IW3_20170221T204724_HH_AB07-BURST S1_191582_IW3_20170227T204642_HH_6654-BURST 
S1_191582_IW2_20170221T204723_HH_AB07-BURST S1_191582_IW2_20170227T204641_HH_6654-BURST 
S1_191582_IW1_20170221T204722_HH_AB07-BURST S1_191582_IW1_20170227T204640_HH_6654-BURST 
S1_191581_IW3_20170221T204721_HH_AB07-BURST S1_191581_IW3_20170227T204639_HH_6654-BURST 
S1_191581_IW2_20170221T204720_HH_AB07-BURST S1_191581_IW2_20170227T204638_HH_6654-BURST 
S1_191581_IW1_20170221T204719_HH_AB07-BURST S1_191581_IW1_20170227T204637_HH_6654-BURST 
S1_191580_IW3_20170221T204718_HH_AB07-BURST S1_191580_IW3_20170227T204636_HH_6654-BURST 
S1_191580_IW2_20170221T204717_HH_AB07-BURST S1_191580_IW2_20170227T204635_HH_6654-BURST 
S1_191580_IW1_20170221T204716_HH_AB07-BURST S1_191580_IW1_20170227T204634_HH_6654-BURST 
S1_191579_IW3_20170221T204716_HH_AB07-BURST S1_191579_IW3_20170227T204633_HH_6654-BURST 
S1_191579_IW2_20170221T204715_HH_AB07-BURST S1_191579_IW2_20170227T204633_HH_6654-BURST 
S1_191579_IW1_20170221T204714_HH_AB07-BURST S1_191579_IW1_20170227T204632_HH_6654-BURST 
S1_191578_IW3_20170221T204713_HH_AB07-BURST S1_191578_IW3_20170227T204631_HH_6654-BURST 
S1_191578_IW2_20170221T204712_HH_AB07-BURST S1_191578_IW2_20170227T204630_HH_6654-BURST 
S1_191578_IW1_20170221T204711_HH_AB07-BURST S1_191578_IW1_20170227T204629_HH_6654-BURST 
S1_191577_IW3_20170221T204710_HH_AB07-BURST S1_191577_IW3_20170227T204628_HH_6654-BURST 

[jhkennedy]

The easiest way to get all the bursts for a particular SLC will be to use asf_search to find the SLC's groupID:

>>> import asf_search as asf

>>> results = asf.product_search('S1A_IW_SLC__1SSH_20170221T204710_20170221T204737_015387_0193F6_AB07-SLC')
>>> results[0].properties['groupID']
'S1A_IWSH_0227_0232_015387_090'

which can then be applied to a burst search in vertex using the groupID filter, or, using doing a burst search in asf_search:

results = asf.search(
    dataset='SLC-BURST',
    groupID='S1A_IWSH_0227_0232_015387_090'
)

[jhkennedy]

Note, cost target is to be able to process all 6 and 12 day pairs for every land ice intersecting burst from 2023-01-01 to "today" (whatever day we submit the pairs for processing) with $30K.