Depth average velocity

.gitignore

@@ -1,4 +1,5 @@
 .DS_Store
+gliderdac*
 
 # Python-generated files
 __pycache__/

pyproject.toml

@@ -31,6 +31,7 @@ build-backend = "hatchling.build"
 
 [dependency-groups]
 dev = [
+    "compliance-checker>=5.4.2",
     "ipykernel>=6.29.5",
     "matplotlib>=3.10.0",
     "mypy>=1.15.0",

scripts/netcdf2csv.example.py

@@ -6,9 +6,28 @@
 
 out = Path("../tests/data").resolve()
 
-xr.open_dataset(Path(out, "osu684.sbd.nc")).drop_dims("j").to_pandas().to_csv(
-    Path(out, "osu684.sbd.csv")
-)
-xr.open_dataset(Path(out, "osu684.tbd.nc")).drop_dims("j").to_pandas().to_csv(
-    Path(out, "osu684.tbd.csv")
-)
+# Commented out because I didn't keep the input files...
+# xr.open_dataset(Path(out, "osu684.sbd.nc")).drop_dims("j").to_pandas().to_csv(
+#     Path(out, "osu684.sbd.csv")
+# )
+# xr.open_dataset(Path(out, "osu684.tbd.nc")).drop_dims("j").to_pandas().to_csv(
+#     Path(out, "osu684.tbd.csv")
+# )
+
+input_sbd = [
+    "osu685-2025-056-0-27.sbd.nc",
+    "osu685-2025-056-0-28.sbd.nc",
+    "osu685-2025-056-0-29.sbd.nc",
+    "osu685-2025-056-0-30.sbd.nc",
+]
+input_tbd = [
+    "osu685-2025-056-0-27.tbd.nc",
+    "osu685-2025-056-0-28.tbd.nc",
+    "osu685-2025-056-0-29.tbd.nc",
+    "osu685-2025-056-0-30.tbd.nc",
+]
+
+for f in input_sbd + input_tbd:
+    xr.open_dataset(Path("../", f)).drop_dims("j").to_pandas().to_csv(
+        Path(out, f.replace(".nc", ".csv"))
+    )

src/glide/assets/config.yml

@@ -9,11 +9,18 @@
 
 --- ### PART 1 GLOBAL CONFIGURATION
 
-# This part contains global attributes. 
-
+# Edit the name for your deployment
+trajectory: 
+  name: YYYYMMDD_gliderSN
+  attributes: # Don't change the attributes, they are required by IOOS
+    cf_role: trajectory_id
+    comment: "A trajectory is a single deployment of a glider and may span multiple data files."
+    long_name: "Trajectory/Deployment Name"
+
+# Add CF standard global attributes here
 netcdf_attributes:
   title: Example Slocum glider dataset generated by the python package glide.
-  comment: glide is developed at https://github.com/OSUGliders/glide
+  comment: glide is developed by the OSU glider group at https://github.com/OSUGliders/glide
   conventions: "CF-1.6, Unidata Dataset Discovery v1.0"
 
 --- ### PART 2 VARIABLE DEFINITIONS, QC CONTROLS, AND METADATA
@@ -128,6 +135,68 @@ m_gps_lon:
     valid_min: -180.0
     coordinate_reference_frame: urn:ogc:crs:EPSG::4326
 
+u:
+  dtype: f4
+  CF:
+    comment: "The depth-averaged current is an estimate of the net current measured while the glider is underwater. The value is calculated over the entire underwater segment, which may consist of 1 or more dives."
+    long_name: "Depth-Averaged Eastward Sea Water Velocity"
+    observation_type: "calculated"
+    platform: "platform"
+    source_sensor: "m_water_vx"
+    standard_name: "eastward_sea_water_velocity"
+    units: "m s-1"
+    valid_max: 10.
+    valid_min: -10.
+
+v:
+  dtype: f4
+  CF:
+    comment: "The depth-averaged current is an estimate of the net current measured while the glider is underwater. The value is calculated over the entire underwater segment, which may consist of 1 or more dives."
+    long_name: "Depth-Averaged Northward Sea Water Velocity"
+    observation_type: "calculated"
+    platform: "platform"
+    source_sensor: "m_water_vy"
+    standard_name: "northward_sea_water_velocity"
+    units: "m s-1"
+    valid_max: 10.
+    valid_min: -10.
+
+time_uv:
+  CF:
+    calendar: "gregorian"
+    comment: "The depth-averaged current is an estimate of the net current measured while the glider is underwater. The value is calculated over the entire underwater segment, which may consist of 1 or more dives."
+    long_name: "Depth-Averaged Time"
+    observation_type: "calculated"
+    source_sensor: "m_present_time"
+    standard_name: "time"
+    units: "seconds since 1970-01-01T00:00:00Z"
+
+lat_uv:
+  dtype: f4
+  CF:
+    comment: "The depth-averaged current is an estimate of the net current measured while the glider is underwater. The value is calculated over the entire underwater segment, which may consist of 1 or more dives."
+    long_name: "Depth-Averaged Latitude"
+    observation_type: "calculated"
+    platform: "platform"
+    source_sensor: "m_gps_lat"
+    standard_name: "latitude"
+    units: "degrees_north"
+    valid_max: 90.
+    valid_min: -90.
+
+lon_uv:
+  dtype: f4
+  CF:
+    comment: "The depth-averaged current is an estimate of the net current measured while the glider is underwater. The value is calculated over the entire underwater segment, which may consist of 1 or more dives."
+    long_name: "Depth-Averaged Longitude"
+    observation_type: "calculated"
+    platform: "platform"
+    source_sensor: "m_gps_lon"
+    standard_name: "longitude"
+    units: "degrees_east"
+    valid_max: 180.
+    valid_min: -180.
+
 heading:
   source: m_heading
   conversion: rad_to_deg

src/glide/cli.py

@@ -118,10 +118,9 @@ def l2(
     """
     conf = config.load_config(config_file)
 
-    flt = process_l1.parse_l1(flt_file)
+    flt_raw = process_l1.parse_l1(flt_file)  # Keep raw for velocity extraction
+    flt = process_l1.format_l1(flt_raw.copy(), conf)
     sci = process_l1.parse_l1(sci_file)
-
-    flt = process_l1.format_l1(flt, conf)
     sci = process_l1.format_l1(sci, conf)
 
     flt = process_l1.apply_qc(flt, conf)
@@ -133,6 +132,10 @@ def l2(
 
     out = process_l1.get_profiles(merged, shallowest_profile, profile_distance)
 
+    out = process_l1.assign_surface_state(out, flt=flt_raw)
+
+    out = process_l1.add_velocity(out, conf, flt=flt_raw)
+
     out = process_l1.enforce_types(out, conf)
 
     out.attrs = conf["globals"]["netcdf_attributes"]
@@ -295,6 +298,130 @@ def concat(
     ds.to_netcdf(out_file)
 
 
+@app.command()
+@log_args
+def backfill(
+    l2_files: Annotated[
+        list[str], typer.Argument(help="L2 files to check for missing velocity.")
+    ],
+    raw_dir: Annotated[
+        str,
+        typer.Option("-r", "--raw-dir", help="Directory containing raw sbd/dbd files."),
+    ],
+    extra_files: Annotated[
+        int,
+        typer.Option(
+            "-n",
+            "--extra",
+            help="Number of extra raw files to load after last L2 file.",
+        ),
+    ] = 3,
+) -> None:
+    """
+    Backfill depth-averaged velocity to L2 files.
+
+    Uses the glider state variable in L2 files to identify dive cycles, then
+    looks up velocity from the corresponding raw flight files. Updates velocity
+    if missing or if the new estimate differs significantly from the existing one.
+
+    File naming convention: L2 files should be named like 'basename.l2.nc'
+    where 'basename.sbd.nc' or 'basename.dbd.nc' is the corresponding raw file.
+    """
+    raw_path = Path(raw_dir)
+
+    # Sort L2 files by name to ensure chronological order
+    l2_files_sorted = sorted([Path(f) for f in l2_files])
+
+    # Filter to files that have velocity variables
+    files_to_update = []
+    for l2_file in l2_files_sorted:
+        try:
+            with nc.Dataset(str(l2_file), "r") as ds:
+                if "time_uv" in ds.variables:
+                    files_to_update.append(l2_file)
+        except Exception as e:
+            _log.warning("Could not read %s: %s", l2_file, e)
+            continue
+
+    if not files_to_update:
+        typer.echo("No L2 files with time_uv variable found.")
+        return
+
+    typer.echo(f"Processing {len(files_to_update)} L2 files.")
+
+    # Extract base names from L2 files (remove .l2.nc suffix)
+    def get_base_name(l2_file: Path) -> str:
+        name = l2_file.name
+        for suffix in [".l2.nc", ".L2.nc", ".nc"]:
+            if name.endswith(suffix):
+                return name[: -len(suffix)]
+        return name.rsplit(".", 1)[0]
+
+    first_base = get_base_name(files_to_update[0])
+    last_base = get_base_name(files_to_update[-1])
+
+    # Get all sbd/dbd files in raw directory, sorted
+    # Look for .sbd.nc, .sbd.csv, .dbd.nc, .dbd.csv patterns
+    raw_files = sorted(
+        list(raw_path.glob("*.sbd.nc"))
+        + list(raw_path.glob("*.sbd.csv"))
+        + list(raw_path.glob("*.dbd.nc"))
+        + list(raw_path.glob("*.dbd.csv"))
+    )
+
+    if not raw_files:
+        typer.echo(f"No raw flight files found in {raw_dir}")
+        return
+
+    # Extract base names from raw files (remove .sbd.nc, .sbd.csv, etc.)
+    def get_raw_base_name(raw_file: Path) -> str:
+        name = raw_file.name
+        for suffix in [
+            ".sbd.nc",
+            ".sbd.csv",
+            ".dbd.nc",
+            ".dbd.csv",
+            ".tbd.nc",
+            ".tbd.csv",
+        ]:
+            if name.endswith(suffix):
+                return name[: -len(suffix)]
+        return name.rsplit(".", 1)[0]
+
+    raw_names = [get_raw_base_name(f) for f in raw_files]
+
+    # Find indices of first and last files to load using exact base name match
+    try:
+        first_idx = raw_names.index(first_base)
+    except ValueError:
+        typer.echo(f"Could not find raw file matching {first_base}")
+        return
+
+    try:
+        last_idx = raw_names.index(last_base)
+    except ValueError:
+        last_idx = first_idx
+
+    # Load from first to last + extra_files
+    end_idx = min(last_idx + extra_files + 1, len(raw_files))
+    raw_files_to_load = [str(f) for f in raw_files[first_idx:end_idx]]
+
+    typer.echo(f"Loading {len(raw_files_to_load)} raw flight files...")
+
+    # Update each L2 file
+    updated = []
+    for l2_file in files_to_update:
+        if process_l1.backfill_velocity(str(l2_file), raw_files_to_load):
+            updated.append(str(l2_file))
+
+    if updated:
+        typer.echo(f"\nUpdated {len(updated)} files:")
+        for f in updated:
+            typer.echo(f"  {f}")
+    else:
+        typer.echo("No files were updated.")
+
+
 @app.command()
 @log_args
 def cfg(