implement Eriks review

Author: reint-fischer

docs/examples/tutorial_Argofloats.ipynb

@@ -256,13 +256,6 @@
     "fig.colorbar(cb, label=\"Temperature (°C)\")\n",
     "plt.show()"
    ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": []
   }
  ],
  "metadata": {

docs/examples/tutorial_delaystart.ipynb

@@ -136,7 +136,7 @@
    "source": [
     "%%capture\n",
     "\n",
-    "ds_out_out = xr.open_zarr(\"DelayParticle_time.zarr\", decode_times=False)\n",
+    "ds_out = xr.open_zarr(\"DelayParticle_time.zarr\")\n",
     "\n",
     "fig = plt.figure(figsize=(7, 5), constrained_layout=True)\n",
     "ax = fig.add_subplot()\n",
@@ -146,25 +146,23 @@
     "ax.set_xlim(31, 33)\n",
     "ax.set_ylim(-32, -30)\n",
     "\n",
-    "timerange = np.unique(ds_out_out[\"time\"].values[np.isfinite(ds_out_out[\"time\"])])\n",
+    "timerange = np.unique(ds_out[\"time\"].values[np.isfinite(ds_out[\"time\"])])\n",
     "\n",
     "# Indices of the data where time = 0\n",
-    "time_id = np.where(ds_out_out[\"time\"] == timerange[0])\n",
+    "time_id = np.where(ds_out[\"time\"] == timerange[0])\n",
     "\n",
-    "sc = ax.scatter(ds_out_out[\"lon\"].values[time_id], ds_out_out[\"lat\"].values[time_id])\n",
+    "sc = ax.scatter(ds_out[\"lon\"].values[time_id], ds_out[\"lat\"].values[time_id])\n",
     "\n",
-    "t = str(timerange[0].astype(\"timedelta64[h]\"))\n",
+    "t = timerange[0].astype(\"datetime64[h]\")\n",
     "title = ax.set_title(f\"Particles at t = {t}\")\n",
     "\n",
     "\n",
     "def animate(i):\n",
-    "    t = str(timerange[i].astype(\"timedelta64[h]\"))\n",
+    "    t = timerange[i].astype(\"datetime64[h]\")\n",
     "    title.set_text(f\"Particles at t = {t}\")\n",
     "\n",
-    "    time_id = np.where(ds_out_out[\"time\"] == timerange[i])\n",
-    "    sc.set_offsets(\n",
-    "        np.c_[ds_out_out[\"lon\"].values[time_id], ds_out_out[\"lat\"].values[time_id]]\n",
-    "    )\n",
+    "    time_id = np.where(ds_out[\"time\"] == timerange[i])\n",
+    "    sc.set_offsets(np.c_[ds_out[\"lon\"].values[time_id], ds_out[\"lat\"].values[time_id]])\n",
     "\n",
     "\n",
     "anim = FuncAnimation(fig, animate, frames=len(timerange), interval=100)"
@@ -278,7 +276,7 @@
    "source": [
     "%%capture\n",
     "\n",
-    "ds_out = xr.open_zarr(\"DelayParticle_releasedt.zarr\", decode_times=False)\n",
+    "ds_out = xr.open_zarr(\"DelayParticle_releasedt.zarr\")\n",
     "\n",
     "fig = plt.figure(figsize=(7, 5), constrained_layout=True)\n",
     "ax = fig.add_subplot()\n",
@@ -295,12 +293,12 @@
     "\n",
     "sc = ax.scatter(ds_out[\"lon\"].values[time_id], ds_out[\"lat\"].values[time_id])\n",
     "\n",
-    "t = str(timerange[0].astype(\"timedelta64[h]\"))\n",
+    "t = timerange[0].astype(\"datetime64[h]\")\n",
     "title = ax.set_title(f\"Particles at t = {t}\")\n",
     "\n",
     "\n",
     "def animate(i):\n",
-    "    t = str(timerange[i].astype(\"timedelta64[h]\"))\n",
+    "    t = timerange[i].astype(\"datetime64[h]\")\n",
     "    title.set_text(f\"Particles at t = {t}\")\n",
     "\n",
     "    time_id = np.where(ds_out[\"time\"] == timerange[i])\n",
@@ -384,14 +382,15 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "outtime_infile = xr.open_zarr(outfilepath, decode_times=False).time.values[:]\n",
-    "print(outtime_infile.astype(\"timedelta64[s]\").astype(\"timedelta64[h]\"))\n",
-    "\n",
-    "# TODO: fix decode_times issue\n",
-    "# assert (\n",
-    "#     outtime_expected[np.isfinite(outtime_expected)]\n",
-    "#     == outtime_infile[np.isfinite(outtime_infile)]\n",
-    "# ).all()"
+    "outtime_infile = (\n",
+    "    xr.open_zarr(outfilepath).time.values[:] - ds.time.values[0]\n",
+    ")  # subtract initial time to convert from datetime64 to timedelta64\n",
+    "print(outtime_infile.astype(\"timedelta64[h]\"))\n",
+    "\n",
+    "assert (\n",
+    "    outtime_expected[np.isfinite(outtime_expected)]\n",
+    "    == outtime_infile[np.isfinite(outtime_infile)]\n",
+    ").all()"
    ]
   },
   {
@@ -424,16 +423,9 @@
     "        dt=np.timedelta64(1, \"h\"),\n",
     "        output_file=output_file,\n",
     "    )\n",
-    "    outtime_infile = xr.open_zarr(outfilepath, decode_times=False).time.values[:]\n",
-    "    print(outtime_infile.astype(\"timedelta64[s]\").astype(\"timedelta64[h]\"))"
+    "    outtime_infile = xr.open_zarr(outfilepath).time.values[:] - ds.time.values[0]\n",
+    "    print(outtime_infile.astype(\"timedelta64[h]\"))"
    ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": []
   }
  ],
  "metadata": {

docs/examples/tutorial_output.ipynb

@@ -86,7 +86,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Parcels saves some metadata in the output file with every simulation (Parcels version, CF convention information, etc.). This metadata is just a dictionary which is propogated to `xr.Dataset(attrs=...)` and is stored in the `.metadata` attribute. The user is free to manipulate this dictionary to add any custom, xarray-compatible metadata relevant to their simulation. Here we add a custom metadata field `date_created` to the output file."
+    "Parcels saves some metadata in the output file with every simulation (Parcels version, CF convention information, etc.). This metadata is just a dictionary which is propogated to `xr.Dataset(attrs=...)` and is stored in the `.metadata` attribute. We are free to manipulate this dictionary to add any custom, xarray-compatible metadata relevant to their simulation. Here we add a custom metadata field `date_created` to the output file."
    ]
   },
   {
@@ -95,7 +95,15 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "output_file.metadata[\"date_created\"] = datetime.now().isoformat()"
+    "output_file.metadata[\"date_created\"] = datetime.now().isoformat()\n",
+    "output_file.metadata"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "To write the metadata to the output_file, we need to add it before running `pset.execute()` which writes the particleset including the metadata to the output_file."
    ]
   },
   {
@@ -119,7 +127,7 @@
    "source": [
     "## Reading the output file\n",
     "\n",
-    "Parcels exports output trajectories in `zarr` [format](https://zarr.readthedocs.io/en/stable/). Files in `zarr` are typically _much_ smaller in size than netcdf, although may be slightly more challenging to handle (but `xarray` has a fairly seamless `open_zarr()` method). Note when when we display the dataset we can see our custom metadata field `date_created`.\n"
+    "Parcels exports output trajectories in `zarr` [format](https://zarr.readthedocs.io/en/stable/). Files in `zarr` are typically _much_ smaller in size than netcdf, although may be slightly more challenging to handle (but `xarray` has a fairly seamless `open_zarr()` method). Note when we display the dataset we can see our custom metadata field `date_created`.\n"
    ]
   },
   {
@@ -376,7 +384,8 @@
    "source": [
     "import cartopy.crs as ccrs\n",
     "import cartopy.feature as cfeature\n",
-    "import matplotlib"
+    "import matplotlib\n",
+    "from matplotlib.animation import FuncAnimation"
    ]
   },
   {
@@ -526,7 +535,7 @@
     "\n",
     "\n",
     "# Create animation\n",
-    "anim = matplotlib.animation.FuncAnimation(fig, animate, frames=nframes, interval=100)\n",
+    "anim = FuncAnimation(fig, animate, frames=nframes, interval=100)\n",
     "anim"
    ]
   },