Flow Estimation

This tool represents the final step of the hydrological workflow, where the actual flow values are estimated. The flow estimation is based on a machine-learning method known as Random Forest (source). In this step, the tool combines the outputs generated by Fix River Network and Calculate Geofactors. Using these inputs, the model is trained and validated on gauged subcatchments, allowing it to learn the relationship between geofactors and flow characteristics. Once trained, the model is applied to ungauged subcatchments to predict their flow behavior.

The resulting outputs include annual mean flow and annual mean low flow, calculated for each subcatchment or along the river network, depending on the selected configuration. Fig. 12 shows a simplified scheme of the model flow.

Fig. 12 Simplified scheme of flow estimation process.

Input data

• gauged_subcatch_geofactors.shp (from Calculate Geofactors)

• ungauged_subcatch_geofactors.shp (from Calculate Geofactors)

• fixed_river_network.shp (from Fix River Network)

Workflow

1. Add all the input data to the project by clicking on “Layer –> Add Layer –> Add Vector Layer”

2. Go in the Processing Toolbox and look for the APRIORA plugin. Click on Hydro-Module and open 4 - Flow Estimation

3. Choose gauged_subcatch_geofactors.shp as input for Gauged subcatchments with geofactors

4. Choose ungauged_subcatch_geofactors.shp as input for Ungauged subcatchments with geofactors

5. Choose fixed_river_network.shp as input for River network

6. Select your preferred option for Model parameter estimation source. If your catchment has no or few gauging stations, a pre-trained model for different regions can be used. Alternatively, you can estimate the paramteres from the gauged_subcatch_geofactors.shp previously derived.

7. Click on Run

Important

Video tutorial will follow soon.

Fig. 13 Interface of the “Flow estimation” window.

Output data:

• subcatchment_level.shp

• river_level.shp

The tool generates two output datasets: one at the subcatchment level and one at the river-section level. Both contain a similar set of attributes. The newly added flow-related fields are displayed in Table 4.

Table 4 New flow-related fields added to output.

Column ID	Description	Unit
Mean_Flow	Estimated annual mean flow for the individual river section or subcatchment	m³/s
M_Low_Flow	Estimated annual mean low flow for the individual river section or subcatchment	m³/s
acc_Mean	Accumulated mean flow, representing the total upstream contribution of mean flow up to the given river section	m³/s
acc_M_Low	Accumulated mean low flow, representing the total upstream contribution of mean low flow up to the given river section	m³/s

These outputs can be used for regional water balance assessments, hydrological model initialization, environmental flow studies and downstream analysis requiring spatially distributed flow estimates.