Fix a bug in projecting water to cryosphere radargrid#253
Open
oberonia78 wants to merge 3 commits intoisce-framework:developfrom
Open
Fix a bug in projecting water to cryosphere radargrid#253oberonia78 wants to merge 3 commits intoisce-framework:developfrom
oberonia78 wants to merge 3 commits intoisce-framework:developfrom
Conversation
oberonia78
commented
Apr 13, 2026
| try: | ||
| dst_srs.SetAxisMappingStrategy(osr.OAMS_TRADITIONAL_GIS_ORDER) | ||
| raster_srs.SetAxisMappingStrategy(osr.OAMS_TRADITIONAL_GIS_ORDER) | ||
| except Exception: |
oberonia78
commented
Apr 13, 2026
Comment on lines
+335
to
+340
| else: | ||
| tx_to_bbox = osr.CoordinateTransformation(raster_srs, dst_srs) | ||
|
|
||
| # transform two neighboring source points to estimate target resolution | ||
| p00 = tx_to_bbox.TransformPoint(x0, y0)[:2] | ||
| p10 = tx_to_bbox.TransformPoint(x0 + dx, y0)[:2] |
Contributor
Author
There was a problem hiding this comment.
Revise from here to line 350
hfattahi
reviewed
Apr 13, 2026
Contributor
hfattahi
left a comment
hfattahi
left a comment
There was a problem hiding this comment.
Thanks @oberonia78 . Please double check how come project_map_to_radar works between 4326 and UTM and not between polar stereo and 4326! Few more comments below as we discussed in PR review.
| raise RuntimeError("Could not determine raster EPSG from projection.") | ||
| raster_epsg = int(raster_epsg) | ||
|
|
||
| xmin, ymin, xmax, ymax = map(float, bbox) |
Contributor
There was a problem hiding this comment.
Is this a type?
Suggested change
| xmin, ymin, xmax, ymax = map(float, bbox) | |
| xmin, ymin, xmax, ymax = map(bbox) |
Comment on lines
+265
to
+268
| try: | ||
| raster_srs.AutoIdentifyEPSG() | ||
| except Exception: | ||
| pass |
Contributor
There was a problem hiding this comment.
what is this helping with?
Comment on lines
+280
to
+284
| try: | ||
| dst_srs.SetAxisMappingStrategy(osr.OAMS_TRADITIONAL_GIS_ORDER) | ||
| raster_srs.SetAxisMappingStrategy(osr.OAMS_TRADITIONAL_GIS_ORDER) | ||
| except Exception: | ||
| pass |
Contributor
There was a problem hiding this comment.
Looks like the order of axis is important for us. If that is the case then you should not pass but raise an exception.
Comment on lines
+298
to
+299
| xmin_src, xmax_src = min(xs), max(xs) | ||
| ymin_src, ymax_src = min(ys), max(ys) |
Contributor
There was a problem hiding this comment.
Does this work for crossing dateline? Please see the stage_dem.py where we do similar transformations and I think we were more careful crossing dateline.
Comment on lines
+346
to
+350
| # fallback in case estimate becomes zero or numerically unstable | ||
| if not np.isfinite(est_dx) or est_dx <= 0: | ||
| est_dx = max((xmax - xmin) / 1000.0, 1e-6) | ||
| if not np.isfinite(est_dy) or est_dy <= 0: | ||
| est_dy = max((ymax - ymin) / 1000.0, 1e-6) |
Contributor
There was a problem hiding this comment.
This does not make sense. What is going here?
Comment on lines
+329
to
+344
| # Use source pixel size magnitude to define output resolution in target CRS. | ||
| # This keeps behavior simple and avoids very strange defaults from Warp. | ||
| # For your case (bbox_epsg=4326), output grid becomes lon/lat. | ||
| if bbox_epsg == raster_epsg: | ||
| out_xres = abs(dx) | ||
| out_yres = abs(dy) | ||
| else: | ||
| tx_to_bbox = osr.CoordinateTransformation(raster_srs, dst_srs) | ||
|
|
||
| # transform two neighboring source points to estimate target resolution | ||
| p00 = tx_to_bbox.TransformPoint(x0, y0)[:2] | ||
| p10 = tx_to_bbox.TransformPoint(x0 + dx, y0)[:2] | ||
| p01 = tx_to_bbox.TransformPoint(x0, y0 + dy)[:2] | ||
|
|
||
| est_dx = abs(p10[0] - p00[0]) | ||
| est_dy = abs(p01[1] - p00[1]) |
Contributor
There was a problem hiding this comment.
Please double check and clean up.
xhuang-jpl
reviewed
Apr 14, 2026
Comment on lines
+1226
to
+1228
| if "water" in mask_type and filter_bool: | ||
| water_mask_path = cfg[ | ||
| "dynamic_ancillary_file_group"]["water_mask_file"] |
Contributor
There was a problem hiding this comment.
what if the water_mask_path is None? please do the check
Comment on lines
+375
to
+385
| warped_band = warped_ds.GetRasterBand(1) | ||
| if warped_band is None: | ||
| raise RuntimeError("Failed to access warped raster band.") | ||
|
|
||
| arr = warped_band.ReadAsArray() | ||
| if arr is None: | ||
| raise RuntimeError("Failed to read warped raster window.") | ||
|
|
||
| warped_gt = warped_ds.GetGeoTransform() | ||
| if warped_gt is None: | ||
| raise RuntimeError("Warped raster geotransform is missing.") |
Contributor
There was a problem hiding this comment.
In addition to raising the RuntimeError, please use journal to log those error.
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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.
This PR fixes three issues related to the use of water masks in ionospheric processing:
When applying the water mask for ionosphere estimation on the main and side bands, the process failed because Frequency A was hard-coded. This PR updates the logic to dynamically select the appropriate projection based on whether the Frequency B radar grid is used.
The processing also failed when using water masks in EPSG:3413 or EPSG:3031, because the code incorrectly assumed an EPSG:4326 projection. This PR fixes the issue by properly handling the input projection and performing the necessary coordinate transformation.
insar.pydeletes the intermediate files but some of them is required for water projection. This PR moves the steps deleting the intermediate files.