Sentinel 1 interface #5
Open
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
The new observation class allows one to hopefully read in a single "pixel" of BHR data over time to do some quick tests. In essence, it just needs a set of dates, albedos and a gp (this bit is missing) It then needs to pack the data into a BHR structure.
Added SAR forward model, as well as code that crates the linear approximation to the SAR model. The SAR observations have yet to be included in the library. Obviously, untested code.
Currently using NC, but might need changing to GeoTIFF. Might also need reprojecting/resampling to fit state grid etc. Fill value is currently set to 0, which isn't great as values in file span from -ve to +ve. Also data in dB probably needs to be converted to linear scale to work with RT model.
The class seems to work, but untested within KaFKA
Current version has had SAR coupled. One thing that I didn't envisage was that we need the Hessian of the RT model to fully calculate the non-linear terms in the uncertainty, and in this case, we do not have it available, so currently, the system breaks when it tries to calculate the Hessian correction using the GP's Hessian method... Other than that... 1. Observations are read and warped to match the state grid 2. The observation operator results in a linearised problem 3. The linear problem is solved 4. The linear part of the uncertainty is solved 5. It then crashes, but we know why! @McWhity This is where we left it
Fixed issue with gradient calculation. Inversion of SAR data for one single time step has now been tested. The results appear plausible (e.g. the observations are closely fitted and the LAI and SM values are within the expected ranges). I haven't yet tested whether the uncertainty is sensible, but want to merge this back into master.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Merging into master the SAR stuff. So far, what we have here is
For a single image, the code works (in the sense that it is able to closely reproduce the observations by changing the model parameters, and these are within the expected values). I haven't yet checked the uncertainty calculations.