xscape 0.2.0

A Python library for recovering the seascape around a set of geographical points, based on Xarray.

XScape

A Python library for recovering the seascape around a set of geographical points, based on Xarray.

Why XScape?

XScape can help you whenever you want to add environmental information to georeferenced events. The information may come from satellites, numerical models, or any field represented on a latitude/longitude grid.

The library was originally developed as part of the PPR Océan & Climat, a research project funded by the French government.

Installation

Installing XScape is as simple as running:

pip install xscape

You may also install the library directly by cloning the repository. Once you have it on your own machine, go into the XScape directory and run poetry install. The whole process can be done in three commands:

git clone https://github.com/Articoking/XScape.git
cd XScape
poetry install

If you want to contribute to the library, follow the specific installation instructions in the documentation.

Basic usage

To use XScape, you only need two things:

• A list of points in a pandas DataFrame with lat and lon as columns.
• An Xarray DataArray containing the "background" variable in a regular latitude/longitude grid.

Let's start by generating some points in the South Atlantic Ocean

import xarray as xr
import xscape as xscp

n_points = 10
points = xscp.generate_points(
    n_points,
    lon_range = (-50, 20),
    lat_range = (-60, 0)
    )

In this example, we'll use XScape functions to get Sea Surface Temperature data from GLORYS. To do that, we need to first define the size of the seascape we want to get around each point, in this case 1°. XScape uses this information to avoid requesting an area much larger than that covered by the points.

seascape_size = 1 # degrees

glorys_var = xscp.get_glorys_var(
    points,
    seascape_size,
    'thetao', # Temperature variable name
    start_datetime = "2000-01-01",
    end_datetime = "2000-12-31",
    )

glorys_var = glorys_var.isel(time=0, depth=0) # See note below

NOTE: XScape currently has no way of obtaining seascapes with time or depth as dimensions, but that will be added in the future. For now, only 2D horizontal maps can be used to create seascapes.

Finally, we create our "XScape DataArray" using the create_xscp_da() function:

xscp_da = xscp.create_xscp_da(
    points,
    seascape_size,
    glorys_var
)

This is a regular Xarray DataArray object with a special format that makes working with seascapes quite simple. For example, you can easily select the seascape that corresponds to the first point in your list:

first_point = points.iloc[0]
seascape = xscp_da.xscp.ss_sel(first_point)

For more information on the format of XScape DataArrays refer to the dedicated documentation page.

Related works

If XScape does not fit your needs, consider taking a look at geoenrich.