Accessing Argo data by float using Argopy

Accessing Argo data by float using Argopy#

Although the goal of the Argo Online school is to teach the basics of the Argo data, and therefore, we explain the data using the primary source, it is worth mentioning argopy:

argopy is a python library dedicated to Argo data access, manipulation and visualisation for standard users as well as Argo experts. the Argo dataset is very complex: with thousands of different variables, tens of reference tables and a user manual more than 100 pages long: argopy aims to help you navigate this complex realm.

In this notebook we show a few examples, but we refer to the argopy Gallery for a more detailled explanation:

Fist, as usual, import the libraries:

import numpy as np
import matplotlib as mpl
from matplotlib import pyplot as plt

import xarray as xr
xr.set_options(display_expand_attrs = False)

<xarray.core.options.set_options at 0x10914fc50>

Import argopy and set-up a data fetcher:

import argopy
argopy.reset_options()
argopy.clear_cache()

from argopy import DataFetcher  # This is the class to work with Argo data
from argopy.plot import scatter_map, scatter_plot  # Functions to easily make maps and plots

You can load one profile from a float:

ArgoSet = DataFetcher().profile(6901254,1)
apProfile = ArgoSet.load().data

/Users/pvb/miniconda3/envs/AOS_clean/lib/python3.12/site-packages/argopy/xarray.py:70: FutureWarning: The return type of `Dataset.dims` will be changed to return a set of dimension names in future, in order to be more consistent with `DataArray.dims`. To access a mapping from dimension names to lengths, please use `Dataset.sizes`.
  self._dims = list(xarray_obj.dims.keys())
/Users/pvb/miniconda3/envs/AOS_clean/lib/python3.12/site-packages/argopy/xarray.py:70: FutureWarning: The return type of `Dataset.dims` will be changed to return a set of dimension names in future, in order to be more consistent with `DataArray.dims`. To access a mapping from dimension names to lengths, please use `Dataset.sizes`.
  self._dims = list(xarray_obj.dims.keys())
/Users/pvb/miniconda3/envs/AOS_clean/lib/python3.12/site-packages/argopy/xarray.py:70: FutureWarning: The return type of `Dataset.dims` will be changed to return a set of dimension names in future, in order to be more consistent with `DataArray.dims`. To access a mapping from dimension names to lengths, please use `Dataset.sizes`.
  self._dims = list(xarray_obj.dims.keys())
/Users/pvb/miniconda3/envs/AOS_clean/lib/python3.12/site-packages/argopy/xarray.py:70: FutureWarning: The return type of `Dataset.dims` will be changed to return a set of dimension names in future, in order to be more consistent with `DataArray.dims`. To access a mapping from dimension names to lengths, please use `Dataset.sizes`.
  self._dims = list(xarray_obj.dims.keys())

or all the profiles of a given float

ArgoSet = DataFetcher().float([6901254])

apDS = ArgoSet.load().data

Argopy gets all the data, including the metadata in a xarray format: very handy

note that the data is organized in ‘points’, a 1D array collection of measurements:

However, and for the purpose of the Argo online school is easier to work with the data in profiles; argopy allows the transformation:

The core-Argo profile files contain the core parameters provided by a float: pressure, temperature, salinity, conductivity (PRES, TEMP, PSAL, CNDC).

Now, let’s see how this argopy data set works as a xarray, and it easy to plot any data. Let’s begin ploting the first two profiles:

fig, ax = plt.subplots(1, 2, figsize=(8,10), sharey=True)

#Temperature
ax[0].plot(data.TEMP[0],-data.PRES[0],'ro',label='N_PROF=0 D Fetcher')
ax[0].plot(data.TEMP[1],-data.PRES[1],'bo',label='N_PROF=1 A Fetcher')
ax[0].set_title(data['TEMP'].attrs['long_name'])
ax[0].set_ylabel(data['PRES'].attrs['long_name'])
ax[0].grid()
ax[0].legend();

ax[1].plot(data.PSAL[0],-data.PRES[0],'ro',label='N_PROF=0 D Fetcher')
ax[1].plot(data.PSAL[1],-data.PRES[1],'bo',label='N_PROF=1 A Fetcher')
ax[1].set_title(data['TEMP'].attrs['long_name'])
ax[1].grid()
ax[1].legend();

../../_images/8467ed9e647a3208e57869d2bc9b3a8c6f820d96b6256cb15fd8163cdb78e29d.png

and of course the locations of these two profiles and the rest of them:

fig, ax = plt.subplots(figsize=(8,8))

ax.plot(data.LONGITUDE[0],data.LATITUDE[0],'ks',markersize=10,label='001D N_PROF=0 Descending')
ax.plot(data.LONGITUDE[1],data.LATITUDE[1],'bs',markersize=10,label='001 N_PROF=0 Ascending')
ax.plot(data.LONGITUDE[2:],data.LATITUDE[2:],'bo',label='rest of the proiles')
ax.grid()
ax.legend();

../../_images/40834c9c910bcb76ff85ffa1b2704391b29981049f9970314e7ecf0cd7fc036f.png

or using the plotting routines of argopy, easyly we can plot the full trayectory

scatter_map(data, set_global=False);

/Users/pvb/miniconda3/envs/AOS/lib/python3.9/site-packages/argopy/plot/plot.py:409: UserWarning: More than one N_LEVELS found in this dataset, scatter_map will use the first level only
  warnings.warn("More than one N_LEVELS found in this dataset, scatter_map will use the first level only")

../../_images/9606676a0e57eb07539edaa74838ae1bac7867dca68614e67657d2844a85debf.png

argopy is powerfull with continous updates, an de refer to the argopy Documentation for a more detailled explanation of its capabilities.