Frequently Asked Questions#

Your documentation keeps mentioning pandas. What is pandas?#

pandas is a very popular data analysis package in Python with wide usage in many fields. Our API is heavily inspired by pandas — this is why there are so many references to pandas.

Do I need to know pandas to use xarray?#

No! Our API is heavily inspired by pandas so while knowing pandas will let you become productive more quickly, knowledge of pandas is not necessary to use xarray.

Should I use xarray instead of pandas?#

It’s not an either/or choice! xarray provides robust support for converting back and forth between the tabular data-structures of pandas and its own multi-dimensional data-structures.

That said, you should only bother with xarray if some aspect of data is fundamentally multi-dimensional. If your data is unstructured or one-dimensional, pandas is usually the right choice: it has better performance for common operations such as groupby and you’ll find far more usage examples online.

Why is pandas not enough?#

pandas is a fantastic library for analysis of low-dimensional labelled data - if it can be sensibly described as “rows and columns”, pandas is probably the right choice. However, sometimes we want to use higher dimensional arrays (ndim > 2), or arrays for which the order of dimensions (e.g., columns vs rows) shouldn’t really matter. For example, the images of a movie can be natively represented as an array with four dimensions: time, row, column and color.

pandas has historically supported N-dimensional panels, but deprecated them in version 0.20 in favor of xarray data structures. There are now built-in methods on both sides to convert between pandas and xarray, allowing for more focused development effort. Xarray objects have a much richer model of dimensionality - if you were using Panels:

  • You need to create a new factory type for each dimensionality.

  • You can’t do math between NDPanels with different dimensionality.

  • Each dimension in a NDPanel has a name (e.g., ‘labels’, ‘items’, ‘major_axis’, etc.) but the dimension names refer to order, not their meaning. You can’t specify an operation as to be applied along the “time” axis.

  • You often have to manually convert collections of pandas arrays (Series, DataFrames, etc) to have the same number of dimensions. In contrast, this sort of data structure fits very naturally in an xarray Dataset.

You can read about switching from Panels to xarray here. pandas gets a lot of things right, but many science, engineering and complex analytics use cases need fully multi-dimensional data structures.

How do xarray data structures differ from those found in pandas?#

The main distinguishing feature of xarray’s DataArray over labeled arrays in pandas is that dimensions can have names (e.g., “time”, “latitude”, “longitude”). Names are much easier to keep track of than axis numbers, and xarray uses dimension names for indexing, aggregation and broadcasting. Not only can you write x.sel(time='2000-01-01') and x.mean(dim='time'), but operations like x - x.mean(dim='time') always work, no matter the order of the “time” dimension. You never need to reshape arrays (e.g., with np.newaxis) to align them for arithmetic operations in xarray.

Why don’t aggregations return Python scalars?#

Xarray tries hard to be self-consistent: operations on a DataArray (resp. Dataset) return another DataArray (resp. Dataset) object. In particular, operations returning scalar values (e.g. indexing or aggregations like mean or sum applied to all axes) will also return xarray objects.

Unfortunately, this means we sometimes have to explicitly cast our results from xarray when using them in other libraries. As an illustration, the following code fragment

In [1]: arr = xr.DataArray([1, 2, 3])

In [2]: pd.Series({"x": arr[0], "mean": arr.mean(), "std": arr.std()})
x                <xarray.DataArray ()> Size: 8B\narray(1)
mean            <xarray.DataArray ()> Size: 8B\narray(2.)
std     <xarray.DataArray ()> Size: 8B\narray(0.81649658)
dtype: object

does not yield the pandas DataFrame we expected. We need to specify the type conversion ourselves:

In [3]: pd.Series({"x": arr[0], "mean": arr.mean(), "std": arr.std()}, dtype=float)
x       1.000000
mean    2.000000
std     0.816497
dtype: float64

Alternatively, we could use the item method or the float constructor to convert values one at a time

In [4]: pd.Series({"x": arr[0].item(), "mean": float(arr.mean())})
x       1.0
mean    2.0
dtype: float64

What is your approach to metadata?#

We are firm believers in the power of labeled data! In addition to dimensions and coordinates, xarray supports arbitrary metadata in the form of global (Dataset) and variable specific (DataArray) attributes (attrs).

Automatic interpretation of labels is powerful but also reduces flexibility. With xarray, we draw a firm line between labels that the library understands (dims and coords) and labels for users and user code (attrs). For example, we do not automatically interpret and enforce units or CF conventions. (An exception is serialization to and from netCDF files.)

An implication of this choice is that we do not propagate attrs through most operations unless explicitly flagged (some methods have a keep_attrs option, and there is a global flag, accessible with xarray.set_options(), for setting this to be always True or False). Similarly, xarray does not check for conflicts between attrs when combining arrays and datasets, unless explicitly requested with the option compat='identical'. The guiding principle is that metadata should not be allowed to get in the way.

What other projects leverage xarray?#

See section Xarray related projects.

How do I open format X file as an xarray dataset?#

To open format X file in xarray, you need to know the format of the data you want to read. If the format is supported, you can use the appropriate function provided by xarray. The following table provides functions used for different file formats in xarray, as well as links to other packages that can be used:

File Format

Open via

Related Packages

NetCDF (.nc, .nc4, .cdf)

open_dataset() OR open_mfdataset()

netCDF4, netcdf , cdms2

HDF5 (.h5, .hdf5)

open_dataset() OR open_mfdataset()

h5py, pytables

GRIB (.grb, .grib)


cfgrib, pygrib

CSV (.csv)


pandas , dask

Zarr (.zarr)

open_dataset() OR open_mfdataset()

zarr , dask

If you are unable to open a file in xarray:

  • You should check that you are having all necessary dependencies installed, including any optional dependencies (like scipy, h5netcdf, cfgrib etc as mentioned below) that may be required for the specific use case.

  • If all necessary dependencies are installed but the file still cannot be opened, you must check if there are any specialized backends available for the specific file format you are working with. You can consult the xarray documentation or the documentation for the file format to determine if a specialized backend is required, and if so, how to install and use it with xarray.

  • If the file format is not supported by xarray or any of its available backends, the user may need to use a different library or tool to work with the file. You can consult the documentation for the file format to determine which tools are recommended for working with it.

Xarray provides a default engine to read files, which is usually determined by the file extension or type. If you don’t specify the engine, xarray will try to guess it based on the file extension or type, and may fall back to a different engine if it cannot determine the correct one.

Therefore, it’s good practice to always specify the engine explicitly, to ensure that the correct backend is used and especially when working with complex data formats or non-standard file extensions.

xarray.backends.list_engines() is a function in xarray that returns a dictionary of available engines and their BackendEntrypoint objects.

You can use the engine argument to specify the backend when calling open_dataset() or other reading functions in xarray, as shown below:


If you are reading a netCDF file with a “.nc” extension, the default engine is netcdf4. However if you have files with non-standard extensions or if the file format is ambiguous. Specify the engine explicitly, to ensure that the correct backend is used.

Use open_dataset() to open a NetCDF file and return an xarray Dataset object.

import xarray as xr

# use xarray to open the file and return an xarray.Dataset object using netcdf4 engine

ds = xr.open_dataset("/path/to/my/", engine="netcdf4")

# Print Dataset object


# use xarray to open the file and return an xarray.Dataset object using scipy engine

ds = xr.open_dataset("/path/to/my/", engine="scipy")

We recommend installing scipy via conda using the below given code:

conda install scipy


Use open_dataset() to open an HDF5 file and return an xarray Dataset object.

You should specify the engine keyword argument when reading HDF5 files with xarray, as there are multiple backends that can be used to read HDF5 files, and xarray may not always be able to automatically detect the correct one based on the file extension or file format.

To read HDF5 files with xarray, you can use the open_dataset() function from the h5netcdf backend, as follows:

import xarray as xr

# Open HDF5 file as an xarray Dataset

ds = xr.open_dataset("path/to/hdf5/file.hdf5", engine="h5netcdf")

# Print Dataset object


We recommend you to install h5netcdf library using the below given code:

conda install -c conda-forge h5netcdf

If you want to use the netCDF4 backend to read a file with a “.h5” extension (which is typically associated with HDF5 file format), you can specify the engine argument as follows:

ds = xr.open_dataset("path/to/file.h5", engine="netcdf4")


You should specify the engine keyword argument when reading GRIB files with xarray, as there are multiple backends that can be used to read GRIB files, and xarray may not always be able to automatically detect the correct one based on the file extension or file format.

Use the open_dataset() function from the cfgrib package to open a GRIB file as an xarray Dataset.

import xarray as xr

# define the path to your GRIB file and the engine you want to use to open the file
# use ``open_dataset()`` to open the file with the specified engine and return an xarray.Dataset object

ds = xr.open_dataset("path/to/your/file.grib", engine="cfgrib")

# Print Dataset object


We recommend installing cfgrib via conda using the below given code:

conda install -c conda-forge cfgrib


By default, xarray uses the built-in pandas library to read CSV files. In general, you don’t need to specify the engine keyword argument when reading CSV files with xarray, as the default pandas engine is usually sufficient for most use cases. If you are working with very large CSV files or if you need to perform certain types of data processing that are not supported by the default pandas engine, you may want to use a different backend. In such cases, you can specify the engine argument when reading the CSV file with xarray.

To read CSV files with xarray, use the open_dataset() function and specify the path to the CSV file as follows:

import xarray as xr
import pandas as pd

# Load CSV file into pandas DataFrame using the "c" engine

df = pd.read_csv("your_file.csv", engine="c")

# Convert `:py:func:pandas` DataFrame to xarray.Dataset

ds = xr.Dataset.from_dataframe(df)

# Prints the resulting xarray dataset



When opening a Zarr dataset with xarray, the engine is automatically detected based on the file extension or the type of input provided. If the dataset is stored in a directory with a “.zarr” extension, xarray will automatically use the “zarr” engine.

To read zarr files with xarray, use the open_dataset() function and specify the path to the zarr file as follows:

import xarray as xr

# use xarray to open the file and return an xarray.Dataset object using zarr engine

ds = xr.open_dataset("path/to/your/file.zarr", engine="zarr")

# Print Dataset object


We recommend installing zarr via conda using the below given code:

conda install -c conda-forge zarr

There may be situations where you need to specify the engine manually using the engine keyword argument. For example, if you have a Zarr dataset stored in a file with a different extension (e.g., “.npy”), you will need to specify the engine as “zarr” explicitly when opening the dataset.

Some packages may have additional functionality beyond what is shown here. You can refer to the documentation for each package for more information.

How does xarray handle missing values?#

xarray can handle missing values using ``np.NaN``

  • np.NaN is used to represent missing values in labeled arrays and datasets. It is a commonly used standard for representing missing or undefined numerical data in scientific computing. np.NaN is a constant value in NumPy that represents “Not a Number” or missing values.

  • Most of xarray’s computation methods are designed to automatically handle missing values appropriately.

    For example, when performing operations like addition or multiplication on arrays that contain missing values, xarray will automatically ignore the missing values and only perform the operation on the valid data. This makes it easy to work with data that may contain missing or undefined values without having to worry about handling them explicitly.

  • Many of xarray’s aggregation methods, such as sum(), mean(), min(), max(), and others, have a skipna argument that controls whether missing values (represented by NaN) should be skipped (True) or treated as NaN (False) when performing the calculation.

    By default, skipna is set to True, so missing values are ignored when computing the result. However, you can set skipna to False if you want missing values to be treated as NaN and included in the calculation.

  • On plotting an xarray dataset or array that contains missing values, xarray will simply leave the missing values as blank spaces in the plot.

  • We have a set of methods for manipulating missing and filling values.

How should I cite xarray?#

If you are using xarray and would like to cite it in academic publication, we would certainly appreciate it. We recommend two citations.

  1. At a minimum, we recommend citing the xarray overview journal article, published in the Journal of Open Research Software.

    • Hoyer, S. & Hamman, J., (2017). xarray: N-D labeled Arrays and Datasets in Python. Journal of Open Research Software. 5(1), p.10. DOI:

      Here’s an example of a BibTeX entry:

        title     = {xarray: {N-D} labeled arrays and datasets in {Python}},
        author    = {Hoyer, S. and J. Hamman},
        journal   = {Journal of Open Research Software},
        volume    = {5},
        number    = {1},
        year      = {2017},
        publisher = {Ubiquity Press},
        doi       = {10.5334/jors.148},
        url       = {}
  2. You may also want to cite a specific version of the xarray package. We provide a Zenodo citation and DOI for this purpose:

    An example BibTeX entry:

          author = {Stephan Hoyer and Clark Fitzgerald and Joe Hamman and others},
          title  = {xarray: v0.8.0},
          month  = aug,
          year   = 2016,
          doi    = {10.5281/zenodo.59499},
          url    = {}

What parts of xarray are considered public API?#

As a rule, only functions/methods documented in our API reference are considered part of xarray’s public API. Everything else (in particular, everything in xarray.core that is not also exposed in the top level xarray namespace) is considered a private implementation detail that may change at any time.

Objects that exist to facilitate xarray’s fluent interface on DataArray and Dataset objects are a special case. For convenience, we document them in the API docs, but only their methods and the DataArray/Dataset methods/properties to construct them (e.g., .plot(), .groupby(), .str) are considered public API. Constructors and other details of the internal classes used to implemented them (i.e., xarray.plot.plotting._PlotMethods, xarray.core.groupby.DataArrayGroupBy, xarray.core.accessor_str.StringAccessor) are not.