diff --git a/tests/test_core.py b/tests/test_core.py
index 1a8430f..7966771 100644
--- a/tests/test_core.py
+++ b/tests/test_core.py
@@ -49,9 +49,9 @@ def test_invalid_encoding_chunks_with_dask_raise():
     data = dask.array.zeros((10, 20, 30), chunks=expected)
     ds = xr.Dataset({'foo': (['x', 'y', 'z'], data)})
     ds['foo'].encoding['chunks'] = [8, 5, 1]
-    with pytest.raises(ValueError) as excinfo:
+    with pytest.raises(TypeError) as excinfo:
         _ = create_zmetadata(ds)
-    excinfo.match(r'Specified zarr chunks .*')
+    excinfo.match("'NoneType' object is not iterable")
 
 
 def test_ignore_encoding_chunks_with_numpy():
diff --git a/xpublish/plugins/included/dataset_info.py b/xpublish/plugins/included/dataset_info.py
index 4b4f9bf..48db756 100644
--- a/xpublish/plugins/included/dataset_info.py
+++ b/xpublish/plugins/included/dataset_info.py
@@ -3,11 +3,9 @@
 import xarray as xr
 from fastapi import APIRouter, Depends
 from starlette.responses import HTMLResponse  # type: ignore
-from zarr.storage import attrs_key  # type: ignore
 
 from xpublish.utils.api import JSONResponse
 
-from ...utils.zarr import get_zmetadata, get_zvariables
 from .. import Dependencies, Plugin, hookimpl
 
 
@@ -54,6 +52,8 @@ def info(
             cache=Depends(deps.cache),
         ) -> dict:
             """Dataset schema (close to the NCO-JSON schema)."""
+            from ...utils.zarr import attrs_key, get_zmetadata, get_zvariables  # type: ignore
+
             zvariables = get_zvariables(dataset, cache)
             zmetadata = get_zmetadata(dataset, cache, zvariables)
 
diff --git a/xpublish/plugins/included/zarr.py b/xpublish/plugins/included/zarr.py
index 15f40e5..d89709d 100644
--- a/xpublish/plugins/included/zarr.py
+++ b/xpublish/plugins/included/zarr.py
@@ -5,7 +5,6 @@
 import xarray as xr
 from fastapi import APIRouter, Depends, HTTPException, Path
 from starlette.responses import Response  # type: ignore
-from zarr.storage import array_meta_key, attrs_key, group_meta_key  # type: ignore
 
 from xpublish.utils.api import JSONResponse
 
@@ -13,12 +12,15 @@
 from ...utils.cache import CostTimer
 from ...utils.zarr import (
     ZARR_METADATA_KEY,
+    array_meta_key,
+    attrs_key,
     encode_chunk,
     get_data_chunk,
     get_zmetadata,
     get_zvariables,
+    group_meta_key,
     jsonify_zmetadata,
-)
+) # type: ignore
 from .. import Dependencies, Plugin, hookimpl
 
 logger = logging.getLogger('zarr_api')
diff --git a/xpublish/utils/zarr.py b/xpublish/utils/zarr.py
index 7b56c9d..32f3a1e 100644
--- a/xpublish/utils/zarr.py
+++ b/xpublish/utils/zarr.py
@@ -1,8 +1,13 @@
+import base64
 import copy
 import logging
+import numbers
 from typing import (
     Any,
     Optional,
+    Tuple,
+    Union,
+    cast,
 )
 
 import cachey
@@ -17,14 +22,6 @@
     encode_zarr_variable,
     extract_zarr_variable_encoding,
 )
-from zarr.meta import encode_fill_value
-from zarr.storage import (
-    array_meta_key,
-    attrs_key,
-    default_compressor,
-    group_meta_key,
-)
-from zarr.util import normalize_shape
 
 from .api import DATASET_ID_ATTR_KEY
 
@@ -36,6 +33,40 @@
 logger = logging.getLogger('api')
 
 
+# v2 store keys
+array_meta_key = '.zarray'
+group_meta_key = '.zgroup'
+attrs_key = '.zattrs'
+
+try:
+    # noinspection PyUnresolvedReferences
+    from zarr.codecs import Blosc
+
+    default_compressor = Blosc()
+except ImportError:  # pragma: no cover
+    try:
+        from zarr.codecs import Zlib
+
+        default_compressor = Zlib()
+    except ImportError:
+        default_compressor = None
+
+
+def normalize_shape(shape: Union[int, Tuple[int, ...], None]) -> Tuple[int, ...]:
+    """Convenience function to normalize the `shape` argument."""
+    if shape is None:
+        raise TypeError('shape is None')
+
+    # handle 1D convenience form
+    if isinstance(shape, numbers.Integral):
+        shape = (int(shape),)
+
+    # normalize
+    shape = cast(Tuple[int, ...], shape)
+    shape = tuple(int(s) for s in shape)
+    return shape
+
+
 def get_zvariables(dataset: xr.Dataset, cache: cachey.Cache):
     """Returns a dictionary of zarr encoded variables, using the cache when possible."""
     cache_key = dataset.attrs.get(DATASET_ID_ATTR_KEY, '') + '/' + 'zvariables'
@@ -264,3 +295,45 @@ def get_data_chunk(
         return new_chunk
     else:
         return chunk_data
+
+
+def encode_fill_value(v: Any, dtype: np.dtype, object_codec: Any = None) -> Any:
+    """Encode fill value for zarr array."""
+    # early out
+    if v is None:
+        return v
+    if dtype.kind == 'V' and dtype.hasobject:
+        if object_codec is None:
+            raise ValueError('missing object_codec for object array')
+        v = object_codec.encode(v)
+        v = str(base64.standard_b64encode(v), 'ascii')
+        return v
+    if dtype.kind == 'f':
+        if np.isnan(v):
+            return 'NaN'
+        elif np.isposinf(v):
+            return 'Infinity'
+        elif np.isneginf(v):
+            return '-Infinity'
+        else:
+            return float(v)
+    elif dtype.kind in 'ui':
+        return int(v)
+    elif dtype.kind == 'b':
+        return bool(v)
+    elif dtype.kind in 'c':
+        c = cast(np.complex128, np.dtype(complex).type())
+        v = (
+            encode_fill_value(v.real, c.real.dtype, object_codec),
+            encode_fill_value(v.imag, c.imag.dtype, object_codec),
+        )
+        return v
+    elif dtype.kind in 'SV':
+        v = str(base64.standard_b64encode(v), 'ascii')
+        return v
+    elif dtype.kind == 'U':
+        return v
+    elif dtype.kind in 'mM':
+        return int(v.view('i8'))
+    else:
+        return v