Add planktivore variables from the Backseat (_) Group

.vscode/launch.json

@@ -458,7 +458,9 @@
             // Processing truncated at 2025-01-12T15:15:43 instead of 2025-01-13T17:43:58 - seems to happen with several other log_files too
             //"args": ["-v", "1", "--log_file", "pontus/missionlogs/2025/20250107_20250123/20250111T233141/202501112331_202501131744.nc4", "--no_cleanup"]
             // Test sipper_odv() with a log_file that has an ESP Sample
-            "args": ["-v", "1", "--log_file", "daphne/missionlogs/2026/20260316_20260318/20260317T191958/202603171919_202603181628.nc4", "--no_cleanup"]
+            //"args": ["-v", "1", "--log_file", "daphne/missionlogs/2026/20260316_20260318/20260317T191958/202603171919_202603181628.nc4", "--no_cleanup"]
+            // Test sipper_odv() with a log_file that has an ESP Sample
+            "args": ["-v", "1", "--log_file", "ahi/missionlogs/2025/20250414_20250418/20250415T040019/202504150400_202504152346.nc4", "--no_cleanup"]
         },
 
     ]

src/data/create_products.py

@@ -190,6 +190,10 @@ def __init__(  # noqa: PLR0913
         "proxy_adinos": "algae",
         "proxy_diatoms": "Purples",
         "proxy_hdinos": "Blues",
+        "avgrois": "YlOrRd",
+        "casetemp": "thermal",
+        "casehumidity": "PuBu",
+        "casepress": "deep",
     }
 
     def _open_ds(self):
@@ -407,6 +411,20 @@ def _get_biolume_plot_variables(self) -> list:
             return self._get_lrauv_biolume_variables()
         return self._get_dorado_biolume_variables()
 
+    def _get_planktivore_plot_variables(self) -> list:
+        """Get planktivore + context variables for plot_planktivore_2column()."""
+        return [
+            ("backseat_planktivore_hm_avgrois", "linear"),
+            ("backseat_planktivore_lm_avgrois", "linear"),
+            ("backseat_planktivore_casetemp", "linear"),
+            ("backseat_planktivore_casehumidity", "linear"),
+            ("density", "linear"),
+            ("wetlabsbb2fl_particulatebackscatteringcoeff470nm", "linear"),
+            ("wetlabsbb2fl_particulatebackscatteringcoeff650nm", "linear"),
+            ("wetlabsbb2fl_mass_concentration_of_chlorophyll_in_sea_water", "linear"),
+            ("backseat_planktivore_casepress", "linear"),
+        ]
+
     def _plot_nighttime_indicator(
         self,
         fig: matplotlib.figure.Figure,
@@ -1750,12 +1768,11 @@ def plot_biolume_2column(self) -> str:  # noqa: C901, PLR0912, PLR0915
 
         self._open_ds()
 
-        # Early return if no biolume variables present
-        biolume_prefix = "wetlabsubat_" if self._is_lrauv() else "biolume_"
-        if not any(v.startswith(biolume_prefix) for v in self.ds.variables):
+        # Early return if no biolume plot variables present in dataset
+        plot_variables = self._get_biolume_plot_variables()
+        if not any(var in self.ds for var, _ in plot_variables):
             self.logger.warning(
-                "No %s* variables found in dataset, skipping plot_biolume_2column",
-                biolume_prefix,
+                "No biolume plot variables found in dataset, skipping plot_biolume_2column",
             )
             return None
 
@@ -1868,6 +1885,143 @@ def plot_biolume_2column(self) -> str:  # noqa: C901, PLR0912, PLR0915
         self.logger.info("Saved biolume 2column plot to %s", output_file)
         return str(output_file)
 
+    def plot_planktivore_2column(self) -> str:  # noqa: C901, PLR0912, PLR0915
+        """Create 2-column planktivore plot with map, ROIs, engineering, and context.
+
+        Layout (5 rows x 2 columns, column-major order):
+          (0,0) track map       (0,1) density
+          (1,0) hm_avgrois      (1,1) backscatter470
+          (2,0) lm_avgrois      (2,1) backscatter650
+          (3,0) casetemp        (3,1) chlorophyll
+          (4,0) casehumidity    (4,1) casepress
+        """
+        # Skip plotting in pytest environment - too many prerequisites for CI
+        if "pytest" in sys.modules:
+            self.logger.info("Skipping plot_planktivore_2column in pytest environment")
+            return None
+
+        self._open_ds()
+
+        # Early return if no planktivore plot variables present in dataset
+        planktivore_vars = [
+            v
+            for v, _ in self._get_planktivore_plot_variables()
+            if v.startswith("backseat_planktivore_")
+        ]
+        if not any(var in self.ds for var in planktivore_vars):
+            self.logger.warning(
+                "No backseat_planktivore plot variables found in dataset, "
+                "skipping plot_planktivore_2column",
+            )
+            return None
+
+        idist, iz, distnav = self._grid_dims()
+        if idist.size == 0 or iz.size == 0 or distnav.size == 0:
+            self.logger.warning(
+                "Skipping plot_planktivore_2column due to missing gridding dimensions"
+            )
+            return None
+
+        fig, ax = plt.subplots(nrows=5, ncols=2, figsize=(18, 10))
+        plt.subplots_adjust(hspace=0.15, wspace=0.04, left=0.05, right=0.97, top=0.96, bottom=0.06)
+
+        best_ctd = None
+        if self._is_lrauv():
+            self.logger.info("LRAUV mission detected for planktivore 2column plot")
+            self._compute_density_lrauv()
+        else:
+            self.logger.info("Dorado mission detected for planktivore 2column plot")
+            best_ctd = self._get_best_ctd()
+            self._compute_density(best_ctd)
+
+        self._plot_track_map(ax[0, 0], ax[1, 0])
+
+        if self.auv_name and self.mission:
+            try:
+                gulper_locations = self._get_gulper_locations(distnav)
+            except FileNotFoundError as e:
+                self.logger.warning("Error retrieving gulper locations: %s", e)  # noqa: TRY400
+                gulper_locations = {}
+        else:
+            try:
+                gulper_locations = self._get_sipper_locations(distnav)
+            except FileNotFoundError as e:
+                self.logger.warning("Error retrieving sipper locations: %s", e)  # noqa: TRY400
+                gulper_locations = {}
+
+        try:
+            profile_bottoms = self._profile_bottoms(distnav)
+        except (TypeError, ValueError) as e:
+            self.logger.warning("Error computing profile bottoms: %s", e)  # noqa: TRY400
+            profile_bottoms = None
+
+        try:
+            bottom_depths = self._get_bathymetry(
+                self.ds.cf["longitude"].to_numpy(),
+                self.ds.cf["latitude"].to_numpy(),
+            )
+        except ValueError as e:  # noqa: BLE001
+            self.logger.warning("Error retrieving bathymetry: %s", e)  # noqa: TRY400
+            bottom_depths = None
+
+        row = 1
+        col = 0
+
+        plot_variables = self._get_planktivore_plot_variables()
+
+        for var, scale in plot_variables:
+            self.logger.info("Plotting %s...", var)
+            if var not in self.ds:
+                self.logger.warning("%s not in dataset, plotting with no data", var)
+
+            self._plot_var(
+                var,
+                idist,
+                iz,
+                distnav,
+                fig,
+                ax,
+                row,
+                col,
+                profile_bottoms,
+                scale=scale,
+                gulper_locations=gulper_locations,
+                bottom_depths=bottom_depths,
+                best_ctd=best_ctd,
+            )
+            if row != 4:  # noqa: PLR2004
+                ax[row, col].get_xaxis().set_visible(False)
+            else:
+                ax[row, col].set_xlabel("Distance along track (km)")
+
+            if row == 4 and col == 0:  # noqa: PLR2004
+                row = 0
+                col = 1
+            else:
+                row += 1
+
+        self._plot_nighttime_indicator(fig, ax[0, 1], distnav)
+
+        if self._is_lrauv():
+            netcdfs_dir = Path(BASE_LRAUV_PATH, f"{Path(self.log_file).parent}")
+            output_file = Path(
+                netcdfs_dir,
+                f"{Path(self.log_file).stem}_{self.freq}_2column_planktivore.png",
+            )
+        else:
+            images_dir = Path(BASE_PATH, self.auv_name, MISSIONIMAGES, self.mission)
+            Path(images_dir).mkdir(parents=True, exist_ok=True)
+            output_file = Path(
+                images_dir,
+                f"{self.auv_name}_{self.mission}_{self.freq}_2column_planktivore.png",
+            )
+        plt.savefig(output_file, dpi=100, bbox_inches="tight")
+        plt.show()
+        plt.close(fig)
+
+        self.logger.info("Saved planktivore 2column plot to %s", output_file)
+        return str(output_file)
+
     def _get_best_ctd(self) -> str:
         """Determine best CTD to use for ODV lookup table based on metadata"""
         # LRAUV doesn't use multiple CTDs, return None
@@ -2333,6 +2487,7 @@ def process_command_line(self):
     p_start = time.time()
     cp.plot_2column()
     cp.plot_biolume_2column()
+    cp.plot_planktivore_2column()
     if cp.mission and cp.auv_name:
         cp.gulper_odv()
     if cp.log_file:

src/data/nc42netcdfs.py

@@ -77,6 +77,15 @@
         {"name": "VolumeScatCoeff117deg700nm"},
         {"name": "mass_concentration_of_petroleum_hydrocarbons_in_sea_water"},
     ],
+    "_": [
+        {"name": "planktivore_diatoms"},
+        {"name": "planktivore_dinoflagellates"},
+        {"name": "planktivore_caseTemp"},
+        {"name": "planktivore_caseHumidity"},
+        {"name": "planktivore_casePress"},
+        {"name": "planktivore_HM_AvgRois"},
+        {"name": "planktivore_LM_AvgRois"},
+    ],
     "WetLabsUBAT": [
         {"name": "average_bioluminescence"},
         {"name": "flow_rate"},
@@ -205,9 +214,12 @@ def extract_groups_to_files_netcdf4(self, log_file: str) -> Path:
 
             # Extract all other groups
             all_groups = list(src_dataset.groups.keys())
+            if "_" in all_groups:
+                self._extract_backseat_group(log_file, src_dataset, netcdfs_dir)
+
             for group_name in sorted(SCIENG_PARMS):
-                if group_name == "/" or group_name not in all_groups:
-                    if group_name != "/" and group_name not in all_groups:
+                if group_name in {"/", "_"} or group_name not in all_groups:
+                    if group_name not in {"/", "_"} and group_name not in all_groups:
                         self.logger.warning("Group %s not found in %s", group_name, input_file)
                     continue
                 self._extract_single_group(log_file, group_name, src_dataset, netcdfs_dir)
@@ -243,6 +255,32 @@ def _extract_root_group(
         else:
             self.logger.warning("No requested variables found in root group '/'")
 
+    def _extract_backseat_group(
+        self, log_file: str, src_dataset: netCDF4.Dataset, output_dir: Path
+    ):
+        """Extract variables from the '_' group to <stem>_{GROUP}_Backseat.nc."""
+        backseat_parms = SCIENG_PARMS.get("_", [])
+        if not backseat_parms:
+            return
+
+        try:
+            src_group = src_dataset.groups["_"]
+        except KeyError:
+            self.logger.warning("Backseat group '_' not found in %s", Path(log_file).name)
+            return
+
+        self.logger.info("Extracting backseat group '_' as Backseat")
+        vars_to_extract, requested_vars = self._get_available_variables(src_group, backseat_parms)
+
+        if vars_to_extract:
+            output_file = output_dir / f"{Path(log_file).stem}_{GROUP}_Backseat.nc"
+            self._create_netcdf_file(log_file, "Backseat", src_group, vars_to_extract, output_file)
+            self.logger.info("Extracted backseat group '_' to %s", output_file)
+        else:
+            self.logger.warning(
+                "No requested variables (%s) found in backseat group '_'", requested_vars
+            )
+
     def _extract_single_group(
         self,
         log_file: str,

src/data/process.py

@@ -700,6 +700,7 @@ def create_products(self, mission: str = None, log_file: str = None) -> None:
         cp.logger.addHandler(self.log_handler)
 
         cp.plot_biolume_2column()
+        cp.plot_planktivore_2column()
         cp.plot_2column()
         if mission and "dorado" in cp.auv_name.lower():
             cp.gulper_odv()

src/data/resample.py

@@ -1987,10 +1987,14 @@ def resample_variable(  # noqa: PLR0913
                 surface_exclusion_depth_m,
             )
         else:
-            self.df_o[variable] = self.ds[variable].to_pandas()
-            self.df_o[f"{variable}_mf"] = (
+            orig_series = self.ds[variable].to_pandas()
+            mf_series = (
                 self.ds[variable].rolling(**{timevar: mf_width}, center=True).median().to_pandas()
             )
+            # Reset df_o per variable to avoid index alignment issues when
+            # variables within the same instrument have different time axes
+            # (e.g., Backseat group has diatoms at ~10s and case* at ~1s)
+            self.df_o = pd.DataFrame({variable: orig_series, f"{variable}_mf": mf_series})
             # Resample to center of freq https://stackoverflow.com/a/69945592/1281657
             self.logger.info(
                 "Resampling %s with frequency %s following %d point median filter",
@@ -2006,19 +2010,11 @@ def resample_variable(  # noqa: PLR0913
                 )
                 dt_index = pd.date_range(mission_start, mission_end, freq=freq.lower())
                 self.df_r[variable] = pd.Series(np.nan, index=dt_index)
-                instr_data = (
-                    self.df_o[f"{variable}_mf"]
-                    .shift(0.5, freq=freq.lower())
-                    .resample(freq.lower())
-                    .mean()
-                )
+                instr_data = mf_series.shift(0.5, freq=freq.lower()).resample(freq.lower()).mean()
                 self.df_r.loc[instr_data.index, variable] = instr_data
             else:
                 self.df_r[variable] = (
-                    self.df_o[f"{variable}_mf"]
-                    .shift(0.5, freq=freq.lower())
-                    .resample(freq.lower())
-                    .mean()
+                    mf_series.shift(0.5, freq=freq.lower()).resample(freq.lower()).mean()
                 )
         return ".mean() aggregator"
 

src/data/test_process_i2map.py

@@ -32,7 +32,7 @@ def test_process_i2map(complete_i2map_processing):
     # update the expected size here.
     EXPECTED_SIZE_GITHUB = 63137
     EXPECTED_SIZE_ACT = 63106
-    EXPECTED_SIZE_LOCAL = 64642
+    EXPECTED_SIZE_LOCAL = 64650
     if str(proc.args.base_path).startswith("/home/runner"):
         # The size is different in GitHub Actions, maybe due to different metadata
         assert nc_file.stat().st_size == EXPECTED_SIZE_GITHUB  # noqa: S101