Colombia Hydrodata

Python client for accessing hydrological and meteorological data from Colombia.

The library integrates official IDEAM data sources and exposes them through a clean, Pythonic API for station discovery, dataset retrieval, and built-in time-series plotting.

Data is fetched from:

• Catalogo Nacional de Estaciones - Datos Abiertos Colombia
• AQUARIUS WebPortal - IDEAM

---

Architecture

The library follows a client -> station -> dataset hierarchy.

Client
  └── Station
        └── Dataset

• Client loads the station catalog and exposes methods to query and filter stations.
• Station represents an IDEAM monitoring station with its full metadata and available variables.
• Dataset represents a time series retrieved from the Aquarius WebPortal.

---

Installation

pip install colombia-hydrodata

Or with Poetry:

poetry add colombia-hydrodata

---

Quick Start

Interactive notebook: Open fetch-data.ipynb in Colab

from colombia_hydrodata import Client

client = Client()

# Fetch a single station
station = client.fetch_station("29037020")
print(station.name)

# Fetch its discharge time series
dataset = station["CAUDAL@HIS_Q_MEDIA_D"]
print(dataset.data.head())

---

Plotting

Interactive notebook: Open plot-data.ipynb in Colab

Datasets expose built-in plotting helpers through the plot property.

import matplotlib.pyplot as plt

dataset.plot.time_series(title=station.name)
plt.show()

For decomposed series, you can generate the built-in analysis layouts:

dataset = (
    station["NIVEL@NV_MEDIA_D"]
    .sight_level(-0.367)
    .rescale(1 / 100)
    .interpolate()
    .deconstruction()
)

fig, axs = dataset.plot.time_series_analysis(figsize=(10, 6), tight_layout=True)
plt.show()

You can also inspect the annual envelope and highlight specific years:

fig, axs = dataset.plot.daily_series_analysis(
    years=[2024, 2025],
    figsize=(10, 4),
    tight_layout=True,
)
axs[0].legend()
plt.show()

---

Client

Client is the main entry point. On initialization it downloads the full IDEAM station catalog from Datos Abiertos Colombia and stores it internally as a GeoDataFrame.

from colombia_hydrodata import Client

client = Client()

---

Fetching Stations

Single station

station = client.fetch_station("29037020")

Returns a Station object with full metadata.

Multiple stations by ID

stations = client.fetch_stations(["29037020", "29037021"])

Returns a list of Station objects.

---

Spatial Queries

Bounding box

stations = client.fetch_bbox(
    xmin=-75.0,
    ymin=9.5,
    xmax=-74.0,
    ymax=10.5,
)

Shapely geometry

from shapely.geometry import Polygon

region = Polygon(...)
stations = client.fetch_region(region)

Catalog only

To get a lightweight GeoDataFrame of station metadata without instantiating Station objects:

gdf = client.stations_in_region(region)
gdf = client.stations_in_list(["29037020", "29037021"])

---

Filtering Stations

Use the Filters dataclass to narrow results by metadata fields. All filters are optional and combined with logical AND.

from colombia_hydrodata import Client
from colombia_hydrodata.filters import Filters

client = Client()

filters = Filters(
    category="Limnimetrica",
    department="Bolivar",
    status="Activa",
)

stations = client.fetch_region(region, filters=filters)

Available filter fields:

Field	Description
category	Station category
department	Department
municipality	Municipality
status	Operational status
owner	Owning institution
hydrographic_area	Hydrographic area
hydrographic_zone	Hydrographic zone
hydrographic_subzone	Hydrographic subzone

---

Station

A Station is a frozen dataclass with full metadata about an IDEAM monitoring station.

station = client.fetch_station("29037020")
print(station)

It also provides the dataset access points:

dataset = station.fetch("CAUDAL@HIS_Q_MEDIA_D")
dataset = station["CAUDAL@HIS_Q_MEDIA_D"]

Both return a Dataset object.

---

Dataset

A Dataset represents a time series associated with a station and a variable.

dataset = station["NIVEL@NV_MEDIA_D"]

print(dataset.station)
print(dataset.variable)
print(dataset.data.head())

The data DataFrame has two core columns:

Column	Description
timestamp	Datetime
value	Numeric measurement

Datasets also expose transformation helpers such as:

• sight_level()
• rescale()
• interpolate()
• detrend()
• seasonal()
• anomalies()
• deconstruction()

And plotting helpers through:

dataset.plot

---

Documentation

Full documentation is available at:

https://sebnarvaez19.github.io/colombia-hydrodata/