[UPDATE] Updated FBC transport with new location and extended timeseries

amocatlas/data_sources/fbc.py

@@ -19,8 +19,6 @@
 from typing import Union
 
 import xarray as xr
-import datetime
-import pandas as pd
 
 # Import the modules used
 from amocatlas import logger, utilities
@@ -34,21 +32,19 @@
 DATASOURCE_ID = "fbc"
 
 # Default list of FBC data files
-FBC_DEFAULT_FILES = [
-    "FBC_overflow_transport.txt",
-]
-FBC_TRANSPORT_FILES = ["FBC_overflow_transport.txt"]
-FBC_DEFAULT_SOURCE = "https://envofar.fo/var/ftp/Timeseries/"
+FBC_DEFAULT_FILES = ["OS_GSR_FBC_D_1995_2024.nc"]
+FBC_TRANSPORT_FILES = ["OS_GSR_FBC_D_1995_2024.nc"]
+FBC_DEFAULT_SOURCE = "https://zenodo.org/records/19554222/files/"
 
 FBC_METADATA = {
-    "project": "Faroe Bank Channel overflow 1995-2015",
-    "weblink": "https://envofar.fo/var/ftp/Timeseries/FBC_overflow_transport.txt",
+    "project": "Faroe Bank Channel overflow 1995-2024",
+    "weblink": "https://zenodo.org/records/19554222",
     "comment": "Dataset accessed and processed via http://github.com/AMOCcommunity/amocatlas",
 }
 
 FBC_FILE_METADATA = {
-    "FBC_overflow_transport.txt": {
-        "data_product": "Daily averaged kinematic FBC-overflow flux (transport) in Sv",
+    "OS_GSR_FBC_D_1995_2024.nc": {
+        "data_product": "Daily average of FBC overflow transport, estimated from an array of moored ADCPs in Sv",
     },
 }
 
@@ -124,7 +120,7 @@ def read_fbc(
 
     added_attrs_per_dataset = [] if track_added_attrs else None
     for file in file_list:
-        if not (file.lower().endswith(".txt")):
+        if not (file.lower().endswith(".nc")):
             log_warning("Skipping unsupported file type : %s", file)
             continue
 
@@ -141,57 +137,11 @@ def read_fbc(
         )
 
         # Open dataset
-        if file.lower().endswith(".txt"):
-            # file.txt
-            try:
-                # column_names, _ = utilities.parse_ascii_header(
-                #     file_path, comment_char="%"
-                # )
-                data_start = utilities.find_data_start(file_path)
-
-                df = pd.read_csv(
-                    file_path,
-                    sep=r"\s+",
-                    encoding="latin-1",
-                    skiprows=data_start,
-                    names=["Decimal year", "Month", "Day", "Flux"],
-                )
-            except Exception as e:
-                log_error("Failed to parse ASCII file: %s: %s", file_path, e)
-                raise FileNotFoundError(
-                    f"Failed to parse ASCII file: {file_path}: {e}"
-                ) from e
-
-            # Time handling
-            try:
-                df = df.apply(
-                    lambda col: col.astype(str)
-                    .str.replace(",", "", regex=False)
-                    .astype(float)
-                )
-                # df['Decimal year'] = df['Decimal year'].astype(str).str.replace(',', '',regex=False).astype(float)
-                df["TIME"] = df["Decimal year"].apply(
-                    lambda x: datetime.datetime(int(x), 1, 1)
-                    + datetime.timedelta(
-                        days=(x - int(x))
-                        * (
-                            datetime.datetime(int(x) + 1, 1, 1)
-                            - datetime.datetime(int(x), 1, 1)
-                        ).days
-                    )
-                )
-                df = df.drop(columns=["Decimal year"])
-                ds = df.set_index("TIME").to_xarray()
-            except Exception as e:
-                log_error(
-                    "Failed to convert DataFrame to xarray Dataset for %s: %s",
-                    file,
-                    e,
-                )
-                raise ValueError(
-                    f"Failed to convert DataFrame to xarray Dataset for {file}: {e}",
-                ) from e
 
+        if file.lower().endswith(".nc"):
+            # Use ReaderUtils for consistent dataset loading
+
+            ds = ReaderUtils.safe_load_dataset(file_path)
             # Attach metadata
             # Attach metadata with optional tracking
 
@@ -222,6 +172,10 @@ def read_fbc(
                     DATASOURCE_ID,
                     track_added_attrs=False,
                 )
+        else:
+            raise ValueError(
+                f"Unsupported file type for {file}. Only .nc files are supported."
+            )
 
         datasets.append(ds)
 

amocatlas/metadata/fbc.yml

@@ -1,8 +1,8 @@
 metadata:
   program: "FBC"
   description: "FBC Overflow transport time series"
-  project: "B. Hansen et al.: A stable Faroe Bank Channel overflow"
-  weblink: https://envofar.fo/data/index.php?dir=Timeseries&sort=N&order=A 
+  project: "B. Hansen, K. M. Larsen: NACLIM - Fluxes: Faroe Bank Channel overflow transport"
+  weblink: https://zenodo.org/records/19554222
   comment: Dataset accessed and processed via http://github.com/AMOCcommunity/amocatlas
   acknowledgment: >
    Funding for the in situ Faroe Bank Channel measurements is from the Environmental Research Programme of the Nordic Council of Ministers (NMR) 1993–1998, from national Nordic research councils, from the Danish DANCEA programme, and from the European Framework Programs, lately under grant agreement no. GA212643 (THOR) and under grant agreement no. 308299 (NACLIM).
@@ -11,29 +11,18 @@ metadata:
   license: 
   featureType: timeSeries
   time_coverage_start: '1995-11-13'
-  time_coverage_end: '2023-05-19'
+  time_coverage_end: '2024-05-18'
   contributing_institutions: "Faroe Marine Research Institute (FAMRI)"
 
 files:
-  FBC_overflow_transport.txt:
-    source_url: https://envofar.fo/var/ftp/Timeseries/FBC_overflow_transport.txt
+  OS_GSR_FBC_D_1995_2024.nc:
+    source_url: https://zenodo.org/records/19554222/files/
     data_product: "Daily averaged kinematic FBC-overflow flux (transport) in Sv"
-    variables_to_remove: [Day, Month]
     variable_mapping:
-      "Flux": TRANS_FBC
+      "FBC_tr": TRANS_FBC
     original_variable_metadata:
-      Flux:
+      FBC_tr:
         long_name: "FBC Overflow"
         description: "FBC Overflow transport time series"
         units: Sverdrup
-        standard_name: ocean_volume_transport_across_line
-      Month:
-        long_name: "Month"
-        description: "Month"
-        units: 
-        standard_name:
-      Day:
-        long_name: "Day"
-        description: "Day"
-        units:
-        standard_name:
+        standard_name: ocean_volume_transport_across_line

tests/test_fbc.py

@@ -29,15 +29,15 @@ def test_default_files_configuration(self):
         """Test default files configuration is reasonable."""
         files = fbc.FBC_DEFAULT_FILES
         assert len(files) > 0
-        assert "FBC_overflow_transport.txt" in files
+        assert "OS_GSR_FBC_D_1995_2024.nc" in files
 
         transport_files = fbc.FBC_TRANSPORT_FILES
         assert len(transport_files) > 0
-        assert "FBC_overflow_transport.txt" in transport_files
+        assert "OS_GSR_FBC_D_1995_2024.nc" in transport_files
 
-        # Files should be text format for FBC data
+        # Files should be NetCDF format for FBC data
         for file in files:
-            assert file.endswith(".txt")
+            assert file.endswith(".nc")
 
     def test_function_exists_and_callable(self):
         """Test that main function exists and is callable."""
@@ -62,4 +62,3 @@ def test_module_imports_successfully(self):
         assert fbc is not None
         # Check required dependencies are accessible
         assert hasattr(fbc, "xr")  # xarray
-        assert hasattr(fbc, "pd")  # pandas