Practice note: Infilling and extension of streamflow data

This practice note is one of a set developed to provide consistency and transparency of river system models being used within the Murray-Darling Basin. The notes cover modelling practices, such as naming conventions for folder structures, to model methods, such as for flow routing and residual inflow estimation, and have been developed through a collaboration between the MDBA, Basin States and CSIRO.

Produced in collaboration with:



This practice note, 'Infilling and extension of streamflow data', documents the general principles to be followed when infilling and extending streamflow data to cover the period of a model run (for calibration or scenario modelling) in an eWater Source river system model.

Background

When running a scenario in eWater Source, all input data needs to cover the period of the simulation run (noting that different scenarios may have different run periods). Where streamflow is an input to your model (e.g. headwater catchments, drain flows), this may require infilling or extension of the streamflow data.

In the Murray Darling Basin, river system models are typically run from 1895 to current. This means that input streamflow data needs to be infilled and extended to cover this period of record.

Methods are summarised in the data section of the eWater Source User Guide Data preparation, and in Vaze et al (2012).

General principles

1. Understand which sites and time series are used as inputs for scenario modelling
2. Sites used as input for scenario modelling should be infilled and extended so they cover the period of the simulation
3. A review of data quality should be undertaken prior to infilling and extension.This may include checking quality codes and metadata, graphing to check for unusual patterns, and using double mass curves between a gauge and reference gauge (or total of a group of nearby gauges) to determine if there is an unusual trend in one of the sets (for both rainfall and flow).

Recommended practices

Model flow sequences can be infilled and extended using various data which may or may not need to be manipulated. In no particular order this includes but is not limited to:

• Observed data at the site or from other sites. Data may be at the required or different timesteps
• Estimates made using empirical and statistical techniques
• Rainfall runoff model estimates (various methods of application in situ or outside catchment).

In some cases, inflows to downstream models are provided from the upstream model and may be provided by external organisations. It is important to understand if these data are a combination of observed and modelled flow or only modelled flow data.

Methods vary depending on the length of the gaps. Interpolation is commonly used for small gaps, with longer gaps infilled and/or extended using a rainfall-runoff model. In Queensland and New South Wales, the Sacramento rainfall-runoff model is typically used for this purpose.

The methods used to infill and extend input data for the Source Model of the Murray and Lower Darling System are described in MDBA (2013a, 2013b).

For scenario flow data, a range of methods are available, including:

• Data infilled and extended using a rainfall-runoff model
• Monthly regression relationships, disaggregated using flow patterns
• Monthly flows disaggregated using flow patterns from nearby gauging stations
• Mean monthly patterns.

References

MDBA (2013a) River Murray and Lower Darling Routing and Loss Calibration Report. Technical Report 2013/25. Murray Darling Basin Authority, Canberra.

MDBA (2013b) Draft Review of Inflow Data for Source. Technical Report 2013/18. Murray Darling Basin Authority, Canberra.

Vaze J, P Jordan, R Beecham, A Frost, G Summerell (2012) Guidelines for rainfall-runoff modelling: Towards best practice model application. eWater Cooperative Research Centre. ISBN 978-1-921543-51-7 https://toolkit.ewater.org.au/Tools/Best%20Practice%20Modelling%20Guidelines/