update zarr section to match

Author: TomNicholas

intermediate/storage_formats.ipynb

@@ -55,12 +55,13 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "- Tree of arbitrary groups\n",
-    "- Each holds arbitrary data in the form of arrays + metadata\n",
-    "- No relationship enforced between groups\n",
-    "- No relationship enforced between arrays within a group\n",
-    "- No concept of \"coordinates\" vs \"data\"\n",
-    "- No references from one group to another"
+    "* **Tree of groups** – Tree of arbitrary groups.\n",
+    "\n",
+    "* **Separate groups** – No relationship enforced between groups, and no references from one group to another.\n",
+    "\n",
+    "* **Separate arrays** – No relationship enforced between arrays within a group.\n",
+    "\n",
+    "* **Arbitrary JSON metadata** – Each holds arbitrary data in the form of arrays + metadata."
    ]
   },
   {
@@ -79,17 +80,21 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "How does zarr relate to `xarray`?\n",
+    "### How does zarr relate to `xarray`?\n",
+    "\n",
+    "* **Arrays <-> `Variables`** - zarr arrays map well to `xarray.Variables`\n",
+    "  - Especially as zarr v3 includes (optional) `dimension_names`\n",
+    "\n",
+    "* **Groups <-> `Datasets`** - zarr groups map reasonably well to `xarray.Dataset` objects\n",
+    "  - Open a single zarr group in xarray via `xr.open_dataset(store, group='/path', engine='zarr')`\n",
     "\n",
-    "- zarr arrays map well to `xarray.Variables`\n",
-    "  - especially because zarr v3 includes (optional) `dimension_names`\n",
-    "- zarr groups map reasonably well to `xarray.Dataset` objects\n",
-    "  - `xr.open_dataset(store, group='/path', engine='zarr')`\n",
-    "  - but `xarray.Dataset`s require that all arrays in the Dataset have aligned dimensions\n",
-    "    - so it is possible to create a zarr group that is not a valid `xarray.Dataset`, if the group contains arrays with non-aligning dimensions\n",
-    "  - Also zarr has no concept of \"coordinate\" vs \"data\" variables\n",
-    "    - so xarray has to save this piece of information as an additional piece of metadata \n",
-    "- zarr store has a tree of groups\n",
+    "* **Groups must be alignable** - But `xarray.Dataset`s require that all arrays in the Dataset have aligned dimensions\n",
+    "  - so it is possible to create a zarr group that is not a valid `xarray.Dataset`, if the group contains arrays with non-aligning dimensions\n",
+    "\n",
+    "* **No \"coordinates\"** – No arrays are special, so Zarr has no intrinsic concept of \"coordinate\" vs \"data\" variables.\n",
+    "  - So xarray has to save this piece of information as an additional piece of zarr metadata.\n",
+    "\n",
+    "* **Tree of groups <-> `DataTree`** - zarr store has a tree of groups\n",
     "  - maps to either a set of independent `xarray.Datasets`\n",
     "    - `xr.open_groups(store)`\n",
     "  - or to a single `xarray.DataTree`\n",
@@ -192,7 +197,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "TIFF (Tag Image File Format) is a *flexible* raster container widely used in biosciences, remote sensing and GIS. \n",
+    "TIFF (Tag Image File Format) is a raster container widely used in biosciences, remote sensing and GIS. \n",
     "\n",
     "A **GeoTIFF** is simply a TIFF that stores additional additional georeferencing information tags (CRS, affine transform, etc.) so geospatial software knows where each pixel sits on Earth. \n",
     "\n",
@@ -206,25 +211,26 @@
     "\n",
     "* **Compression / tiling** – DEFLATE, LZW, etc. Tiling lets software fetch small windows efficiently.\n",
     "\n",
-    "### Practical notes for xarray users\n",
-    "\n",
-    "* **Read** – use `rioxarray.open_rasterio()` (wraps rasterio) to get an immediate, Dask-chunked DataArray.\n",
-    "\n",
-    "* **Write** – `DataArray.rio.to_raster(\"out.tif\")`; choose compression + tiling via driver_kwargs.\n",
-    "\n",
-    "* **Dimensionality** – TIFF is inherently 2-D per band; no native time or vertical axis. If you need 4-D data, NetCDF or Zarr is usually a better fit.\n",
-    "\n",
-    "* **Metadata depth** – single-level tags only (no nested groups). For rich hierarchies, stick to HDF5 / NetCDF-4.\n",
-    "\n",
-    "* **Cloud-optimized GeoTIFF (COG)** – same format, arranged so HTTP range requests can stream windows efficiently; xarray handles it transparently when rasterio is compiled with libcurl.\n"
+    "* **Cloud-optimized GeoTIFF (COG)** – same format, arranged so HTTP range requests can stream windows efficiently; xarray handles it transparently when rasterio is compiled with libcurl."
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
     "### How does TIFF relate to xarray?\n",
-    "\n"
+    "\n",
+    "* **Dimensionality** – Each raster image maps well to a single `xarray.Variable`, but TIFF is inherently 2-D per band; no native time or vertical axis. If you need 4-D data, NetCDF or Zarr is usually a better fit.\n",
+    "\n",
+    "* **No named dimensions** - TIFFs don't have named dimensions for the two axes of the raster.\n",
+    "\n",
+    "* **IFDs as groups** -  IFDs can be mapped to groups, which may be useful for multi-resolution TIFFs (also known as \"overviews\") and multi-page TIFFs.\n",
+    "\n",
+    "* **Metadata depth** – single-level tags only (no nested groups). For rich hierarchies, stick to HDF5 / NetCDF-4.\n",
+    "\n",
+    "* **Read** – use `rioxarray.open_rasterio()` (wraps rasterio) to get an immediate, Dask-chunked DataArray. However `rioxarray` is for interacting with GeoTIFFs, not general TIFFs.\n",
+    "\n",
+    "* **Write** – `DataArray.rio.to_raster(\"out.tif\")`; choose compression + tiling via driver_kwargs."
    ]
   },
   {