How it works

Yield Prophet uses the world-class APSIM model, developed and maintained by the CSIRO and the APSIM Initiative, to simulate the effects of environmental variables and management decisions on crop yields.

Simulation Inputs

The crop simulations are created by combining the essential components of growing a crop including

a soil test sampled prior to planting
a soil characterisation selected from the APSoil library of ~1,000 soils selected as representative of the production area
historical and current climate data taken from the nearest Bureau of Meteorology (BOM) weather station or current climate data from your privately owned weather station (optional)
paddock specific rainfall data recorded by the user (optional) or your privately owned weather station (optional)
individual crop details
fertiliser and irrigation applications during the growing season.

Soil Test Sampling

In late summer and autumn, subscribers sample the soil in their Yield Prophet paddocks/zones down to the maximum rooting depth of their crop at different intervals (e.g. 0-10, 10-40, 40-70, 70-100cm). These samples are analysed for water content, organic carbon, nitrate and chloride concentration, electrical conductivity and pH. This information is entered into the Yield Prophet web interface. Alternatively, Yield Prophet can directly import existing soil data from Back Paddock’s SoilMate, allowing SoilMate users to avoid additional data entry. The soil test information is then used by the grower/consultant and the Yield Prophet team to select a suitable soil characterisation. Information on the requirements and costs of initial soil data are outlined in Yield Prophet® Soil Sampling Information below.

Soil Characterisation

An appropriately measured soil characterization is essential for Yield Prophet to accurately simulate crop growth and yield. The Plant Available Water Capacity (PAWC) and bulk density of a specific soil type are used to determine how much of the measured water and nitrogen is available to the crop for growth during the season. PAWC is determined by a soil’s ‘Drained Upper Limit’ (DUL, or field capacity) and its ‘Crop Lower Limit’ (CLL, similar to permanent wilting point). Yield Prophet has a catalogue of soil characterisations for many of the major cropping soil types found throughout Australia. More information about the catalogue of soils can be found on the APSOIL website. Yield Prophet users must select a soil characterisation from the catalogue that is most representative of the area soil sampled. In some circumstances there are soil types for which there is no available characterisation data. In these circumstances, it is recommended that potential subscribers to Yield Prophet consider characterising their soil if no appropriate data exists. If you are interested in having your soils characterised or would like more information please contact Tim McClelland. Alternatively, if you would like to understand more about the soil characterisation process, please visit the APSOIL website for a methodology.

Climate Data & Rainfall

Yield Prophet accesses climate information sourced directly over the web from the SILO Patched Point Dataset (PPD). This is a catalogue of climate information for 4600 weather stations across Australia. If you wish to know more about the SILO PPD please visit the SILO website. Currently users select a weather station from the database that is representative of the location of their paddock/zone. The weather station has two distinct purposes in the simulations process:

1. It provides climate data from the current season which is used to simulate crop growth and the soil water and nitrogen processes in the paddock/zone from the time of soil sampling to the time of the report. During the season, Yield Prophet subscribers may enter rainfall for each paddock/zone into the web interface. Alternatively, they can opt to utilise the seasonal rainfall data recorded at their nominated PPD station.

2. It provides historic data which is used to simulate crop growth and resource availability from the day on which the report was generated to the end of the season. This process is repeated once for each year of climate record (~130 years) providing 130 separate yield outcomes.

Now users may opt to utilize private live weather stations located at the paddock/zone being simulated. However, it should be noted that this improvement will only be realised in the simulation of crops in the current season. Yield Prophet will still need to utilise the historic records from the PPD as the basis for simulations from the day on which the report was generated to the end of the season.

Crop Details

Yield Prophet requires subscribers to enter individual crop details into the website for inclusion in the simulations. Specific details required are the type and amount of starting stubble, sowing date, crop type, cultivar and sowing density.

Fertiliser and Irrigation Applications

During the season, subscribers update paddock/zone management details (cultivations and nitrogen fertiliser and irrigation applications) as they occur.

Crop Growth Simulation

When a report is requested, the user paddock-specific information is emailed from the Yield Prophet database to a computer cluster where the simulation is processed. Using climate data for the current season, Yield Prophet simulates the soil water and nitrogen processes in the paddock/zone from the soil sampling date, together with the crop growth from the user-nominated sowing date, to the present. Yield Prophet calculates the amount of water and nitrogen available to the crop and the water and nitrogen demands of the crop. This is used to determine whether the crop is suffering stress from lack of either of these resources and to assess any resultant reduction in growth and yield potential.

Using historic climate data, Yield Prophet then simulates crop growth and resource availability from the day on which the report was generated to the end of the season. This process is repeated once for each year of climate record (~130 years) providing 130 separate yield outcomes (Figure 1). These yields are then plotted as a probability curve (Figure 2), providing growers with an estimate of the probabilities of obtaining a range of yields. This output is then adapted to the requested report type and emailed to the website or the subscriber, where it can be viewed by the subscriber. Report generation takes from five to fifteen minutes.

Figure 2 is the main output of Yield Prophet which is a core component of all the reports generated in Yield Prophet. Figure 2 is showing that this crop has:

a yield potential of 4.8t/ha given the best season finish on record with available nitrogen
a yield potential of 7.0t/ha given the best season finish on record with unlimited nitrogen from today forward
a yield potential of 7.9t/ha given the best season finish on record with unlimited nitrogen for the whole season
a yield potential of 2.3t/ha given the worst season finish on record with available and unlimited nitrogen
a 50% chance of achieving a yield of at least 3.8t/ha with available nitrogen
a 50% chance of achieving a yield of at least 4.5t/ha with unlimited nitrogen and unlimited nitrogen from today forward
a 60% chance of benefitting from the application of nitrogen (i.e. the nitrogen limited curve separates from the nitrogen unlimited curves at the 60% line)

Figure 1: Visual representation of the Yield Prophet simulation process

Figure 2: A yield probability curve, the principal output of Yield Prophet.

Scenario predictions

The probable impact of different sowing dates, varieties and irrigation and nitrogen applications can be determined by simulating different ‘scenarios’. Yield Prophet generates a probability curve for each scenario, and subscribers use this to determine the probability of achieving a yield response from the addition of water or nitrogen (Figure 3 and Figure 4) or from different sowing dates and varieties (Figure 5). Yield Prophet can also calculate a nitrogen gross margin based on the predicted grain quality and price (Figure 6).

Figure 3: Yield probability curves for three different nitrogen top-dressing scenarios generated for a dryland wheat crop. Scenario 1 (pink line) represents the yield probability adding no further nitrogen, Scenario 2 (blue line) represents the yield probability with 20 kg/ha of nitrogen top-dressed and Scenario 3 represents the yield probability with 40 kg/ha of nitrogen top-dressed. There is an 80% chance of achieving a yield response with topdressing, and about a 30% chance of achieving a 1 t/ha yield response from 40 kg/ha of nitrogen.

Figure 4: Yield probability curves for three different nitrogen and irrigation scenarios generated for an irrigated wheat crop. Scenario 1 (pink line) represents the yield probability adding no further water or nitrogen, Scenario 2 (blue line) represents the yield probability with an additional 50 kg/ha of nitrogen top-dressed, Scenario 3 represents the yield probability with 50 kg/ha of nitrogen top-dressed and two additional 25 mm irrigations.

Figure 5: Yield probability curves for three different varieties and sowing date scenarios (shown above the graph) generated for a wheat crop at Birchip.

Figure 6: Nitrogen profit curves for the same nitrogen application scenarios shown in Figure 3. Each line is calculated as the return from grain (determined by yield and protein minus cost of fertiliser and spreading) for Scenario 1 (solid pink line 0kg/ha of nitrogen) Scenario 2 (solid blue line, 20 kg/ha of nitrogen) and 3 (solid red line, 40 kg/ha of nitrogen). This shows the difference in return between applying nitrogen at specified amounts, and not applying nitrogen. In this case, the cost of fertiliser is assumed to be $2.00 per kg of nitrogen, cost of spreading $10 per ha and the wheat price (AH) $350/t, with a $2.50 per 0.5% protein bonus.