forked from https://github.com/jmunroe/intake-erddap.

Intake-ERDDAP

Copyright 2022 Axiom Data Science

See LICENSE

Copyright 2022 James Munroe

For changes prior to 2022-10-19, all contributions are Copyright James Munroe, see PREV-LICENSE.

Check out our Read The Docs page for additional documentation

Intake is a lightweight set of tools for loading and sharing data in data science projects. Intake ERDDAP provides a set of integrations for ERDDAP.

• Quickly identify all datasets from an ERDDAP service in a geographic region, or containing certain variables.
• Produce a pandas DataFrame for a given dataset or query.
• Get an xarray Dataset for the Gridded datasets.

The Key features are:

• Pandas DataFrames for any TableDAP dataset.
• xarray Datasets for any GridDAP datasets.
• Query by any or all:
  • bounding box
  • time
  • CF standard_name
  • variable name
  • Plaintext Search term
• Save catalogs locally for future use.

User Installation

In the very near future, we will be offering the project on conda. Currently the project is available on PyPI, so it can be installed using pip

  pip install intake-erddap

Developer Installation

Prerequisites

The following are prerequisites for a developer environment for this project:

• conda
• (optional but highly recommended) mamba Hint: conda install -c conda-forge mamba

Note: if mamba isn't installed, replace all instances of mamba in the following instructions with conda.

1. Create the project environment with:

   mamba env update -f environment.yml
   

2. Install the development environment dependencies:

   mamba env update -f dev-environment.yml
   

3. Activate the new virtual environment:

   conda activate intake-erddap
   

4. Install the project to the virtual environment:

   pip install -e .
   

Examples

To create an intake catalog for all of the ERDDAP's TableDAP offerings use:

import intake
catalog = intake.open_erddap_cat(
    server="https://erddap.sensors.ioos.us/erddap"
)

The catalog objects behave like a dictionary with the keys representing the dataset's unique identifier within ERDDAP, and the values being the TableDAPSource objects. To access a source object:

source = catalog["datasetid"]

From the source object, a pandas DataFrame can be retrieved:

df = source.read()

Consider a case where you need to find all wind data near Florida:

import intake
from datetime import datetime
bbox = (-87.84, 24.05, -77.11, 31.27)
catalog = intake.open_erddap_cat(
   server="https://erddap.sensors.ioos.us/erddap",
   bbox=bbox,
   start_time=datetime(2022, 1, 1),
   end_time=datetime(2023, 1, 1),
   standard_names=["wind_speed", "wind_from_direction"],
)

df = next(catalog.values()).read()

	time (UTC)	wind_speed (m.s-1)	wind_from_direction (degrees)
0	2022-12-14T19:40:00Z	7.0	140.0
1	2022-12-14T19:20:00Z	7.0	120.0
2	2022-12-14T19:10:00Z	NaN	NaN
3	2022-12-14T19:00:00Z	9.0	130.0
4	2022-12-14T18:50:00Z	9.0	130.0
...	...	...	...
48296	2022-01-01T00:40:00Z	4.0	120.0
48297	2022-01-01T00:30:00Z	3.0	130.0
48298	2022-01-01T00:20:00Z	4.0	120.0
48299	2022-01-01T00:10:00Z	4.0	130.0
48300	2022-01-01T00:00:00Z	4.0	130.0