index

What is Yield Prophet®?
Take the shortcut to the Yield Prophet website

Yield Prophet® is a web interface for the crop production model APSIM
(www.apsim.info). It simulates crop growth based on paddock-specific inputs of
soil type, pre-sowing soil water and nitrogen, rainfall, irrigation and nitrogen fertiliser applications, and climate data.

Yield Prophet® was developed by BCG (Birchip Cropping Group) in collaboration
with CSIRO as a risk management tool for dry-land farming systems in the
Victorian Wimmera and Mallee, with an emphasis on decision support for
nitrogen fertiliser inputs. It was first used for wheat at BCG trial sites
in 2002, and its early predictions of the failure of that season generated
sufficient interest and credibility to allow a commercial release to BCG
members in 2003 as a monthly fax-out service. Continuing demand resulted in the development of the
Yield Prophet® web-interface, which allowed a larger
number of subscribers to receive up-to-date crop information and forecasts
at will in 2004. The end of season report detailing model performance can
be downloaded as a PDF.

2005 was the first year of general commercial release of the service, and 338
paddocks were subscribed to the service from all over Australia . Over
6800 reports were generated during the season, and the end of season
report detailing model performance will be released online shortly.

How does Yield Prophet® work?
Subscription
Farmers or consultants subscribe to the service in late summer and autumn and
provide the Yield Prophet® team with their paddock names, locations (used
to determine soil type and closest Bureau of Meteorology weather stations)
and planned crops and varieties.

Subscribers are then given a user name and password with which they can log
onto the Yield Prophet® website. Growers are also able to nominate a
consultant with whom they wish to access Yield Prophet®, and this
consultant is also given access to data on that grower’s paddocks.

During autumn, subscribers sample their Yield Prophet® paddocks’ soil at
different depth intervals down to the maximum rooting depth of their crop
(e.g. 0-10, 10-40, 40-70, 70-100 cm). These samples are analysed for water
content, nitrate concentration, organic carbon, electrical conductivity,
chloride concentration and pH. These data are entered by growers into the
Yield Prophet® web interface, and are also used by the grower and Yield
Prophet® team to select a suitable soil characterisation.

An appropriately measured soil characterisation is an essential
input for
Yield Prophet® to simulate crop growth, yield and protein accurately.
The plant available water capacity (PAWC) and bulk density of a specific soil
type 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). The Yield Prophet®
team have a ‘library’ of soil characterisations measured for many of the
major cropping soil types found within BCG’s membership catchment and other
regions Australia-wide. However, many subscribers have soil types for
which there are no available measured characterisation data. In these
circumstances, a soil characterisation is estimated by the Yield Prophet® team
based on soil type and previous rainfall and crop yields provided by
the growers, and any information available from existing soil surveys. An
estimated characterisation is less likely to produce accurate results in
comparison to a measured characterisation, and it is recommended that
potential subscribers to Yield Prophet® consider characterising their soil if no appropriate data exist.
During the season, subscribers enter paddock management details (sowing date,
crop type, variety, nitrogen fertiliser and irrigation) and
rainfall. When growers wish to find out how much water and nitrogen is
currently available to a crop, the likely yield of their crop, or what the
likely impact of management events will be, they generate a report.

When a report is generated, Yield Prophet® simulates daily crop growth from
sowing up to the present using the paddock specific rainfall and
management data entered by the subscriber, and climate data (maximum and
minimum temperature, radiation, evaporation and vapour pressure) from the
nominated BOM weather station.

At every daily time step Yield Prophet® calculates the amount of water and
nitrogen available to the crop, and the water and nitrogen demand of the
crop. This is used to determine if the crop is suffering stress from lack
of either of these resources, and any subsequent reduction in growth and
yield potential. This information is then presented to subscribers in
reports returned to the subscribers’ account (Figure 1).

Figure 1. Output from Yield Prophet® indicating the amounts of water and
nitrogen available to the crop during the season. The stress graphs
indicate loss of potential growth and carbon fixation, i.e. on a day when
the graph is at 0.5, the crop is growing and photosynthesising at half its
potential rate.

Yield prediction
In order to make predictions about crop yield, Yield Prophet® uses the last
one hundred years of climate data taken from the nearest Bureau of
Meteorology weather station to continue the simulation from the date of
report generation to the end of the season. The model simulates one
hundred different crop yields and proteins, based on the current season up
until the day the report is generated, and on the season finishes of the
past one hundred years. These yields are then plotted as a probability
curve (Figure 2), which provides growers with an estimate of the
probabilities of obtaining different yields. This range of probabilities
narrows as the season progresses and components of yield become more certain.

Figure 2. A yield probability curve, the main output from Yield Prophet®.
This is the main output of Yield Prophet®, and its value is increased by
incorporating seasonal forecasts, such as the Southern Oscillation Index
(SOI) phase system. In this case, instead of using season finishes for the
last one hundred years, Yield Prophet® selects the years in which the SOI
phase was the same as in the current year, and runs the future part of
the simulation using only the finishes from those years. This creates
another probability curve which growers can use if the SOI phase is
strongly indicating wet or dry conditions (Figure 3).

Figure 3. Yield probability curve generated using season finishes for the last
hundred years of climate data (solid blue line), and only those years
in which the SOI phase was the same as the current phase at the time
the report was generated. In the above example, this is the years with a
negative SOI phase in June-July; the report was generated in early August 2004.

Scenario predictions
The likely impact of different sowing dates, varieties and irrigation and
nitrogen applications can then be determined by simulating
different ‘scenarios’. Yield Prophet® calculates a probability curve for
each scenario, and subscribers use this to determine the probability of
achieving a yield or protein response from the addition or water or
nitrogen (Figures 8 and 9), or from different sowing dates and varieties
(Figure 10). Yield Prophet® can also calculate a nitrogen gross margin
based on likely grain quality and price (Figure 11).

Figure 8. Yield probability curves for three different nitrogen top-dressing
scenarios generated for a dry land wheat crop on 1 August 2005. Scenario 1
(pink line) is the yield probability adding no further nitrogen, Scenario
2 (blue line) is the yield probability with 35 kg/ha of nitrogen top-
dressed on 15 August, Scenario 3 is the yield probability with 70 kg/ha of
nitrogen top-dressed on 15 August 2005. There is an 80% chance of
achieving a yield response with topdressing, and about a 50% chance of
achieving a 1 t/ha yield response from 35 kg/ha of nitrogen

Figure 9. Yield probability curves for three different nitrogen and irrigation
scenarios generated for an irrigated wheat crop on 3 October 2005.
Scenario 1 (pink line) is the yield probability adding no further water or
nitrogen, Scenario 2 (blue line) is the yield probability with an
additional 50 kg/ha of nitrogen top-dressed on 3 October, Scenario 3 is
the yield probability with 50 kg/ha of nitrogen top-dressed on 3 October
and two additional 25 mm irrigations on 3 and 17 October.

Figure 10. Yield probability curves for three different sowing date scenarios
(sowing dates are shown above the graph) generated for a wheat at Birchip
crop on 21 June 2005.

Figure 11. Nitrogen profit curves for the same two nitrogen application
scenarios shown in Figure 8. Each line is calculated as the return from
grain (determined by yield and protein minus cost of fertiliser and
spreading) for Scenarios 2 (solid red line, 35 kg/ha of nitrogen) and 3
(solid blue line, 70 kg/ha of nitrogen) from Figure 8, minus the return
from Scenario 1 (adding no further nitrogen). This shows the difference in
return between applying nitrogen of specified amounts, and not applying
nitrogen. In this case it assumed the cost of fertiliser as $0.95 per kg
of nitrogen, cost of spreading as $5 per ha and that the wheat price would
be APW at $160 per tonne, with a $2 per 0.5% protein bonus.

Irrigation scheduling
Because Yield Prophet® calculates the amount of water available to a crop, and
average evaporation and transpiration based on 100 years of data, it has
the potential to be a very effective tool for irrigation scheduling.
Figure 12 shows the Irrigation Scheduling report from Yield Prophet®. The
graph shows the PAWC of the soil that is being accessed by the crop as
roots grow, and the amount of PAW calculated from initial measured soil
water plus rainfall and irrigation, subtract evaporation and transpiration.
The red section of the line is a projection of PAW over the future
two weeks assuming no rain, and growers can use this to determine when to water,
and how much water to apply. The impact of any irrigation can be
calculated from the probability curves in the irrigation comparison report
described above (Figure 9).
This output is also very useful for dry-land crop management, as it provides
an indication of the amount of water that is available to a crop at any
point in the season.

Figure 12. The graphic display of soil water from the Irrigation Scheduling
Report in Yield Prophet®.

Yield Prophet® in 2006
Yield Prophet® in 2006 features many improvements on the service in 2005.
These include;
- New look user-friendly website
- More efficient website structure for consultants
- ‘Batch’ and ‘favourite’ report generation capabilities, as well as
streamlined report structures, for quicker and easier report generation -
Long waits for reports eliminated by installation of a 26 dual
processor computer cluster for simulation runs
- Improved barley module
- User memory meaning users don’t need to provide password during
each session - User selection of customised home page upon log-in
- Automatically updated crop growth stage readouts
- Additional agronomic information, including rooting depth and
forecast timing of growth stages relevant to management (end of tillering,
flag leaf emergence, flowering See it for yourself!

You can visit the Yield Prophet® website to view previously generated reports
for the BCG trial site and access help files which will tell you more
about operating Yield Prophet® and how to interpret reports. Simply
respond to Username and Password by typing in Visitor.

Pricing & Subscription
Yield Prophet® subscription types and prices are listed here. Prices do not
include sampling and analysis for initial soil conditions. Information on
the requirements and costs of initial soil data are available here.

Counter