Project Title
A Regional Commodity Forecasting System for major Australian crops
Project Leader
Graeme Hammer
Organisation
Department of Primary Industries
Funding Body
APSRU/QCCA
Administration Contact
Andries Potgieter
Department of Primary Industries
Ph: 07 4688 1417 Fax: 07 4688 1193
Commencement Date
1 February 1999
Completion Date
Ongoing
Research Proposal Summary
The logistics of handling and trading Australia’s grain commodities are confounded by huge swings in production associated with climate variability. Advance information on likely production and its geographical distribution is sought by many industries, particularly in the recently deregulated marketing environment.
Australia’s agricultural research scientists have over the last decade developed some of the most useful seasonal climate forecasting and cropping models systems in the world. By integrating simplified crop modelling systems within a Geographical Information System (GIS) environment a regional commodity forecasting system (RCFS) is created. According to the AACM (1991) existing commodity forecasts are based on a compilation from various sources of information and are unreliable and only broadly indicative. A more quantitative approach to RCFS has been initiated in previous work between DPI (Hammer and Butler) and the WA Dept of Agriculture (Stephens). Hammer (1999) has defined an effective application of a seasonal climate forecast system (SCFS) as use of forecast information leading to a change in a decision that generates improved outcomes in the system of interest.
The RCFS is a perfect example of an operational system, which can make use of SCFS and generate commodity forecasts, based on the available seasonal climate forecast. Such commodity forecasts are scientifically sound, objective, accurate and repeatable. De Jager et al. (1998) has shown that by implementing RCFS improved decision-making that generated improved outcomes resulted in the maize industry of South Africa. The objective of this study will be to implement a RCFS, in Queensland and nationally, for the forecasting of wheat yields (as a first) on a regional scale. The outputs will be updated on a monthly basis and the SOI phases will be used as an SCF indicator. Hammer et al. (1996) have compared the predictive capability of six yield forecasting models for wheat yield for the major cropping areas of Australia. Based on that study it was decided to use the simple stress index (SI) model as the agro-climatic model in the RCFS project for Australia.
Aims
§ To integrate existing crop modelling and climate forecasting systems on regional and national scales and create an operational commodity forecasting system
§ To quantify the skill levels of commodity forecasts on a regional spatial scale
Potential Outcomes
§ Improved decision making on regional commodity resources (eg. Logistics of handling, importing/exporting, fertilisers ect.) well before harvest via dissemination of the forecasts to potential clients like agribusiness, commodity traders as well as Co-ops and grain handling companies.
§ Improved commodity drought alert systems, which assists government in drought management decisions
§ Improved assessment of the spatial correlation of climate forecasting indicators and performance of cropping systems at regional scale.
Milestones
§ Meeting with relevant role players - ongoing
§ Updated historical wheat production data (83 years) – June 1999
§ Updated meteorological station data from QCCA in Indooroopilly - May 1999
§ Changed SI-model code to read APSIM climate data format – March 1999
§ Utility program which will read near real-time climate data and patch it forward with analogue years climate based on the SOI phases – April 1999
o Adapt SI-model to do forecasting based on analogue SOI years.
§ Calculate model error against previous outputs - June 1999
§ Simulated wheat stress index for each met-station for the 1999 wheat season
o Calculate weighted averages of model outputs for each shire within GIS - ongoing
o Calibrate SI’s against 1975-1993 ABS census data - once
§ Import all relevant GIS data from Arc Info to Arc View Spatial Analysis - ongoing
§ Regional wheat forecast for the 1999 wheat season started in May and updated monthly - ongoing
§ Operational wheat forecasting system – June 2000
§ Meeting with relevant researcher to assess the effect of changing certain model parameters like planting date, soil water etc – Feb 2000
§ Link model outputs to GIS database (Long-term & forecasted) - ongoing
§ Statistical analysis on 83 years of simulated wheat data for each shire – Dec 1999
o Regional cluster analysis (Stats & GIS)
o Assessing how good climate predictors can forecast the wheat clusters
o Published article on this research – June 2000
§ Make contact and explain the system to possible end users like and put communication links in place for further collaboration with:
o Grainco,
o Australian Wheat Board,
o Agri-Industry,
o State & National Government
§ Calculate accuracy and relative efficiency of RCFS
o Meeting with relevant statisticians on this methodology – Apr 2000
o Hindcast of 83 wheat yield years for each shire (284) – May 2000
o Published of article on this research – Dec 2000
§ Change or improve the GIS methodology regarding the selection of meteorological stations – Dec 2000 and ongoing
§ Commodity forecast for other Australian crops - ongoing
Budget
The project is currently part of core funded activity of QCCA and APSRU.
Resource Requirements & Contributions
Prior Provision of IP & Rights of Ownership
APSRU will retain the sole intellectual property rights to APSIM
Ownership, equity in and use of IP to be developed
A – 100
Rights of Publication
Joint
Strategic Plan Goal No.
3 & 4
Relevance to Strategic Plan
Core
§ maintain and foster alliances with policy makers and their advisers to identify issues relevant to APSRU involvement
§ apply simulation and other technical tools to policy issues on an opportunistic basis
§ collaborate in refining methods to integrate APSRU point modelling with spatial modelling
§ develop alliances (nationally and internationally) to enhance capacity and effectiveness in developing methodologies