Practice note: Estimation of Unmetered Irrigation Diversions (On- Farm Water Balance)

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 and Basin States.

Produced in collaboration with:

This practice note, 'Estimation of unmetered irrigation diversions', describes the general principles and a suggested method for estimating the total unmetered irrigation diversions from floodwaters. This practice note aims to detail how a modeller should estimate total irrigation take (and thus estimate unmetered take). This estimate can then be used to constrain the calibration of crop water models or, where crop water models are not used to estimate unmetered take, provide an estimate of unmetered diversions.

Background

Understanding the on-farm water balance is essential for predicting the demand for irrigation water within a river system model. Modelling the on-farm water balance provides an understanding of the total volume of water required to meet irrigation demands based on the area of crops planted. Depending on the location of the irrigator within the river system, the water sources available for irrigation may vary, and thus the complexity of the on-farm water balance may also vary.
When unmetered diversions are not a significant component of the on-farm water balance, metered diversions can be assumed to represent the surface water diversions for irrigation purposes. Where the unmetered diversions are not a significant component of the water balance, there is no need to estimate the on-farm water balance before the calibration of irrigation demand models. However, where crop models are being used to represent the demand for irrigation water, checks should be undertaken following calibration to ensure that the application rates in these demand models are within the expected range.
Where unmetered diversions represent a significant component of the on-farm water balance, modelling will be required to provide estimates of the unmetered components of the on-farm water balance. This practice note aims to detail how to estimate total irrigation diversions via an on-farm water balance in areas where water is taken from floodwaters to meet irrigation demands. This estimate of total irrigation diversions may come from crop water models or other methods. Where crop water models are used, this provides an estimation of the take from rainfall-runoff harvesting and floodplain harvesting.

General principles

An understanding of the different sources of water that can be used to meet irrigation demands should be developed during reach conceptualisation (Reach Conceptualisation - Identification of key fluxes).
Where unmetered diversions are considered to be a significant component of the on-farm water balance, periods of no unmetered water use (dry years) and obvious periods of unmetered water use (flood years) should be identified.
All available information that may assist in the estimation of total water use by crops should be collated. Data collection may include: metered diversions, groundwater usage, planted areas, crop type, on-farm storage information, other on-farm infrastructure, an understanding of access to floodwaters via breakouts, intermittent tributary inflows or rainfall-runoff harvesting.
The modeller should consider what secondary information or multiple lines of evidence are available to assist in either the estimation of unmetered irrigation diversions or to provide a check on the values obtained for total crop water use. Secondary information may include: remote sensing information, and information relating to irrigation application rates from alternative sources such as ABS data.
The modeller should document the method used to estimate the on-farm water balance and the estimate of unmetered diversions used during calibration of the reach water balance.
Where water requirements and diversions for irrigation will be estimated using a crop water model (aiming to maintain soil moisture at a nominated depletion value), this model should be used to underpin the estimation of total crop water requirements and on-farm water balance, and appropriate checks should be undertaken following the calibration of the crop water model to ensure that the diversions, planted area and crop water use (ML/ha), and the timing of floodplain harvesting diversions are within expected ranges and timeframes.

Recommended High-Level Method

Collate data on:
1. Areas developed for irrigation.
2. The capability to store water (permanent and temporary storages).
3. The capacity of on-farm storages (including stage-volume relationships).
4. Floodplain harvesting infrastructure.
5. Operation of on-farm storage (e.g. reserves maintained during the growing season, airspace retained, storage levels or storage volumes).
6. Pump capacities (particularly on-farm storage pump rates).
7. Observed data (time series) on planted crops (type and area).
8. Location of breakouts (the ability of floodwater to be used for irrigation), the relationship between river flow and breakout flows.
9. Average on-farm storage (OFS) losses to seepage.
10. Irrigation efficiency and irrigation application rates.
11. How planting and watering strategies change during floodplain harvesting events. Do application rates increase? Is a second crop planted? Or is there an increased area?
12. Development levels over time. This should include data points from multiple years if possible, or at least some estimates of when infrastructure was built or altered as this may change over time.
13. Irrigation practices (e.g. skip row practices).
Review data collected, which may include:
1. Using remote sensing data to confirm the capacity of on-farm storages and stage-volume relationships.
2. Confirming the date of OFS construction and history of temporary storage use.
3. Reviewing planted areas, taking into account information on known irrigation practices.
4. Providing an independent estimate of crop actual evapotranspiration (ET) (noting the uncertainty in the ET estimates).
5. Understanding the timing of overbank flow events that could provide unmetered irrigation diversions.
Estimate water use by irrigated crops. This estimation of water use is a key component of on-farm water balance and is required to produce realistic estimates of diversions from surface water via floodplain harvesting and rainfall-runoff harvesting. If a crop/soil water balance model is being used, then this should be calibrated according to the 'Calibration of Crop Water Models' practice note and should be underpinned by the best available industry data, rather than relying on calibration to match the metered diversions. The key steps in calibrating the crop water model are:
1. Develop a crop water balance for each crop using the method outlined in 'Crop evapotranspiration – Guidelines for computing crop water requirements– FAO Irrigation and drainage paper 56'. To do this, the modeller should:
  1. Determine crop factors from published data or using crop factors based on advice from local experts. Remote sensing to directly estimate crop ET may also be used if available, but it should be noted that uncertainty in ET estimates using remote sensing is ±20% (pers comm., Albert Van Dijk, 2018).
  2. For each crop type, use a simplified irrigator model with an efficiency estimate and unlimited water supply to determine the crop water use per ha for the available climate sequence. This step includes the impact of effective rainfall on irrigation requirements. The outputs from this model would represent an upper estimate for crop water usage.
  3. Check the values obtained for application rates, (ii) against other available information such as mm or ML/ha from WaterSched Pro (https://waterschedpro.net.au/), ABS data on irrigation application rates.
2. Using the time series of planted areas (observed/estimated and reviewed) and the ML/ha application rates (determined above), the total water use by irrigated crops for each year in the calibration period can be estimated.
3. Develop a water balance of the OFS, including representation of net evaporation and seepage losses.
4. Calculate the total water use as the sum of water use by irrigated crops, net evaporation, and seepage from the OFS and metered diversions.
5. Calculate the unmetered diversions as the difference between total water use and metered diversions. Where crop water models are being used to estimate the unmetered diversions a time series of unmetered diversion can be generated using this model.
6. Following calibration of the irrigation demand model, the outputs should be reviewed to ensure that:
  1. Planted areas simulated by the model are reasonable.
  2. Total water use is within the expected range.
  3. Crop water use in ML per hectare is reasonable.
  4. On-farm storage behaviour matches any recorded on-farm storage information and is consistent with known wet and dry events or periods.
  5. Floodplain harvesting extractions occur at appropriate times.
  6. Ensure that the runoff coefficients for rainfall-runoff harvesting are within the expected range. At the mean annual timescale, the runoff coefficients should be checked using the Budyko Framework as detailed in Neumann et al. (2017). However, if the best available local information on runoff from irrigated areas indicates a runoff coefficient from irrigated areas outside the expected bounds, this should be reported and explained in the calibration report.

Companion Practice Notes

References

Neumann, L.E., Brown, A., Zhang, L., Zheng, H.X., Davidson, A.J., Egan, C. and Korn, A. (2017). Assessing residual inflow and loss estimates methods in river reach calibration using the Budyko Framework. In Syme, G., Hatton MacDonald, D., Fulton, B. and Piantadosi, J. (eds) MODSIM2017, 22nd International Congress on Modelling and Simulation. Modelling and Simulation Society of Australia and New Zealand, December 2017.
https://www.mssanz.org.au/modsim2017/L6/neumann.pdf