Intermittency

intermittency loads flow-state observations such as ONDE-style dry or flowing states. It is useful for diagnosing active network behavior and for qualitative comparison with simulated drainage activation.

Accepted sources

Source	Use when	Source page
custom	Local flow-state observations are authoritative.	custom
hubeau	Hub’Eau/ONDE-style public intermittency observations should be retrieved.	hubeau

Minimal example

[data.intermittency]
date_start = "2010-01-01"
date_end = "2020-12-31"

[[data.intermittency.sources]]
source = "hubeau"
extent = "watershed"

Visual check

Intermittency is a state observation. The figure should make the state coding, dates, and stations readable before it is compared with a simulated active network.

Local deterministic run

The repository also includes a local intermittency-only case under hydromodpy/data/variables/intermittency/cases. It is useful for explaining the custom file convention without relying on an online Hub’Eau request.

Fig. 162 The timeline checks whether station identifiers, dates, and categorical state codes were loaded as intended. The histogram makes the balance between dry, transitional, and flowing states visible before the data are compared with a simulated active network.

Downstream uses

• active-network interpretation;

• qualitative validation of drainage/seepage behavior;

• seasonal low-flow diagnostics.

Intermittency Source: custom

Use source = "custom" when local flow-state observations should be ingested from project files.

Minimal example

[[data.intermittency.sources]]
source = "custom"
path = "data/intermittency"
col_id = "station"
col_datetime = "date"
col_value = "state"

Operational checks

• Document the state coding used in the local file.

• Confirm that station coordinates are available or discoverable.

• Do not compare state observations to simulated fluxes until the thresholding rule has been stated explicitly.

Intermittency Source: hubeau

Use source = "hubeau" to retrieve public flow-state observations.

Minimal example

[[data.intermittency.sources]]
source = "hubeau"
extent = "watershed"
code_departement = ["35", "53"]

Operational checks

• Department filters can keep discovery predictable.

• Check observation dates against the simulation or comparison window.

• The data are categorical observations, not discharge values.

Provider replay

Fig. 163 The intermittency part of the Hub’Eau replay shows categorical state values, not a continuous hydrological flux. That distinction should be visible before ONDE-style data are compared with a simulated active network.