Australian
Agri-Food 2000 Research Forum
Melbourne August 17
Can agribusiness utilise better information on climate variability?
Brennan, L.1, Carberry, P2. and Hochman, Z2
CSIRO Tropical
Agriculture /
Agricultural Production Systems Research Unit (APSRU)
1120 Meiers Rd,
Indooroopilly, QLD, 4068.
2 PO Box 102 (203 Tor St) Toowoomba, QLD, 4350
| Abstract |
| Introduction |
| The Agricultural Production Systems Simulator (APSIM) |
| Engaging with agribusiness |
| Case studies |
Click
here for the PowerPoint Presentation  |
Climate variability impacts significantly on the agricultural service sector, affecting the operations and policies of agribusiness suppliers, banking and insurance companies. For example, bank lending policy and agribusiness advice was likely affected by recent El Niño drought events. Through consultation with agribusiness suppliers, banks and insurance companies, it is clear that their business operations and policies could benefit substantially from access to enhanced processes for dealing with climate variability.
The Agricultural Production Systems Research Unit (APSRU) has demonstrated that farmers can utilise information derived from climate forecasts and simulation models in interpreting past experience, planning and decision making. Variability in production also poses challenges for both input suppliers and firms involved in the value chain from farm to consumer.
Bank lending policies, crop insurance policies, product inventories and marketing advice may all be positively influenced through better dealing with climate variability. For example, insurance policies based on the Agricultural Production Systems Simulator (APSIM) model’s objective prediction of yields may potentially reduce claimant disputes and cut legal costs, representing significant savings to industry.
Likewise, better prediction of seasonal outlooks using the APSIM model, climate forecasts and fallow water reserves would allow farmers and lenders, such as produce suppliers and banks, to negotiate individually-tailored financial packages. APSRU has recently extended risk management tools developed primarily for farmers to agribusiness to determine whether better targeted and cost-effective agribusiness services can be provided for the benefit of agribusiness organisations and Australian farmers.
In this paper we report on our experiences and learnings from our action research approach, where APSRU researchers are working alongside agribusiness staff on relevant case studies to identify the opportunities for and to resolve constraints against implementation of improved agribusiness operations based on climate forecasts and use of the APSIM model.
Our collaborators represent a mix of agribusiness organisations, including input suppliers, marketers, banks and insurance companies. The case studies involving our collaborators are described and discussed in the paper, along with results of our initial evaluation.
The use of climate forecasts and APSIM has generated interest amongst the agribusiness sector. While the costs of conducting this research are high, we conclude that there are good opportunities for these tools to assist agribusiness operations in providing better services to Australian farmers.
Climate variability impacts significantly on the agricultural service sector, affecting the operations and policies of agribusiness suppliers, banking and insurance companies. For example, bank lending policy and agribusiness advice were affected by droughts concomitant with recent El Niño events. The Agricultural Production Systems Research Unit (APSRU) has consulted agribusiness suppliers, banks and insurance companies and clearly these agribusiness sectors could benefit substantially from access to better processes for dealing with climate variability.
APSRU is a joint R&D unit consisting of CSIRO Tropical Agriculture and the Queensland Departments of Primary Industry and Natural Resources. APSRU’s expertise is in the computer simulation of farming systems targeted at research that impacts on how agricultural production systems are managed. APSRU has developed a set of tools that could assist agribusiness improve the way they deal with climate variability. These include:
- tools to assist farmers and agribusiness to better understand the soil resource in relation to alternative agricultural enterprises (e.g. soil coring equipment, soil resource database);
- seasonal climate forecasts based on the Southern Oscillation Index (SOI) and Sea Surface Temperatures (SST);
- the APSIM cropping systems model.
APSRU’s interest in agribusiness stems from eight years of experience in helping farmers deal with climate variability. APSRU’s FARMSCAPE (Farmers Advisers Researchers Monitoring Simulation Communication And Performance Evaluation - is jointly supported by APSRU and the Grains Research and Development Corporation.) project has demonstrated that farmers can utilise information from seasonal climate forecasts and crop simulation models to interpret past experience for planning and decision making.
Can agribusiness operations and policies also benefit from access to better processes for dealing with climate variability?
Variability in production also poses challenges for both input suppliers and firms involved in the value chain from farm to consumer. In what represents an evolution of the APSRU’s capability in this area, APSRU has recently extended risk management tools developed primarily for farmers to agribusiness to determine whether better targeted and cost-effective agribusiness services can be provided for the benefit of agribusiness organisations and Australian farmers through the application of climate forecasts and APSIM simulations.
With support from LWRRDC's Climate Variability in Agriculture R&D Program, APSRU is exploring delivery of the FARMSCAPE approach for institutional decision making by designing case studies, in partnership with industry, to test the commercial application of the APSIM model.
The Agricultural Production Systems Simulator (APSIM)
What is APSIM?
The Agricultural Production Systems Simulator (APSIM) is a model that uses various component modules to simulate cropping systems (McCown R.L., Hammer G.L., Hargreaves J.N.G., Holzworth, D.P. and Freebairn D.M. (1996). APSIM: A novel software system for model development, model testing, and simulation in agricultural systems research. Agricultural Systems, 50: 255-271).
The simulation capacity of APSIM is limited only by the availability of modules to simulate the processes peculiar to the system of interest. APSIM has the ability to simulate the growth of a range of crops in response to a variety of management practices, crop mixtures and rotation sequences, including pastures and some aspects of livestock production. Importantly, this is accomplished in such a way that the soil accrues the effects of the different agricultural practices such as cropping, fallowing, residue management and tillage.
In this way, APSIM can simulate long-term trends in soil productivity due to fertility depletion and erosion. APSIM contains modules that permit the simulation of soil organic matter rundown, nutrient leaching, soil erosion, acidification and soil phosphorus. There is however no current capability to deal with effects of salinisation, insects, diseases or biodiversity loss. Simulated outputs can include yield and gross margin (returns, cash flow and risk probabilities) and other systems consequences such as soil erosion or acidification etc.
APSIM has been specified for, and its simulations tested against, a range of crop and farming systems. Over the past 6 years, a number of commercial crops have been monitored and used to test APSIM simulations. In most cases, these tests have confirmed that APSIM is able to simulate commercial crop production –accounting for over 80% of the observed variation in crop yields. For most of those crops where predictions were significantly different, we have been able to determine the reasons for the discrepancies – most are due to impacts of factors not accounted for in the models (eg. severe pest damage).
Who uses APSIM?
APSIM is currently being used and tested in all cropping regions of Australia, and in a number of international collaborations, and has been used in a range of research and pilot commercial activities. Formal evaluation of the current uses has demonstrated impacts on participating farmers and advisers. For many farmers and consultants, APSIM has proved credible enough to be relevant to commercial cropping practices and they now want to use it in benchmarking the performance of their own crops and in exploring alternative management strategies. The demand for simulations has increased rapidly to the point where APSRU can not meet that demand, nor justify providing a "commercial" delivery service.
APSRU is currently transferring APSIM to the commercial sector via an Accredited Adviser Network for delivering simulation and related products such as soil monitoring, seasonal climate forecasts, analysis of relevant management scenarios and "what-ifs, analysis and discussion" to farmer clients in the northern cropping region. A number of agribusiness and private consultants are being accredited and supported in implementing this approach within their business practices.
How is APSIM used?
Pre-plant forecast
Just as the general public are becoming increasingly aware of events such as El
Niño, managers of cropping systems are increasingly utilising climate forecast
information in making management decisions. The Southern Oscillation Index (SOI)
is widely promoted as a forecast of seasonal outlook, providing probabilities of
achieving above or below average rainfall in upcoming seasons.
APSIM can be used in planning for the current or upcoming crop. Decisions on crop management based on expected yield and risk can be assessed based on knowledge of pre-plant soil water, soil chemical analysis and seasonal climate outlook (based on the SOI). Using this information, APSIM can provide an assessment of expected crop performance in the upcoming season by simulating what would have happened under these same conditions in past years for which rainfall records exist. APSIM can assess yield probabilities for different crop options for a farm or region.
Post-crop yield assessment
Assessing whether a crop has performed to its potential or the yield impact of a particular event on crop performance are issues readily addressed by APSIM. Given actual seasonal climate and management inputs, APSIM can predict what a crop should have yielded in the absence of extraneous factors, thus providing a benchmark against which actual crop yield can be assessed.
As an initial stage of this project, Rabobank Australia (specifically, Rabobank Australia Ltd’s Agribusiness Consulting and Research Services) was engaged to identify potential applications for APSRU tools within the agricultural value chain and prospective companies that may benefit from access to these tools. The following have been identified as potential client segments:
- Suppliers of genetics
- Other input suppliers & retailers such as fertilizer, chemical and crop care companies
- Primary producers
- Corporate producers and marketers (e.g.: cotton companies, feedlotters)
- Agents and wholesalers who (may) interface between producers and early stage processors (e.g.: soft commodity merchants)
- Early stage processors (e.g.: abattoirs, millers)
- Late stage processors (e.g.: Kelloggs, Goodman Fielder)
- Retailers
- Service providers such as insurance, finance, and logistic companies
In Rabobank’s assessment, APSRU’s value to industry is largely driven by the challenge to many CEOs of not simply how to play the current competitive game better, but how to anticipate and prepare for the next competitive game. Improving current performance is only part of their management challenge; looking at ways to foster growth, understand the value chain in which they operate, and hence to change the rules of competition is their real strategic challenge. For agribusiness service companies, this is largely an issue about managing risk, especially understanding of supply and biological risk of the agricultural production base – which is APSRU’s core capability.
The Rabobank consultancy delivered on providing us with access to senior managers of a number of agribusiness companies with possible interests in APSRU tools and capabilities. Interest generated among agribusiness was high, with genuine interest in APSRU tools and capabilities expressed by all parties. We also independently approached a number of agribusiness firms including banks, insurance and input suppliers.
Our experience suggests that the APSRU capabilities that underpin the value of the APSIM model to agribusiness are:
- Recurrent annual production assessment
- Assessment of yield variability
- Simulating historical records
- Reconstruction of the past
- Assessment of new technology
- Assessment of new crop varieties
- Assessment of new (alternative) land uses
- Assessment of risk through the early stage production chain
- Synthesising the link between production and the environment
These capabilities could be converted to benefits to industry in the following ways:
Ability to forecast supply – quality and yield – for a given area
Ability to conduct crop assessment
- existing crop in a new region
- current crop with new technology in an existing region
- potential versus current performance in an existing region
- new crop in a new region
- Ability to reconstruct the past and determine what went wrong and why
- Ability to forecast environmental impacts
- Ability to assess water utilisation
- develop more efficient utilisation of water and
- assess impacts of increasing water price
- Ability to contribute towards the development of new risk management (insurance) and financial products.
After initial interactions with agribusiness we sought to develop case study concepts based on matching APSRU’s capabilities and proposed solutions with anticipated industry needs. Interest generated among agribusiness was high (all approached companies through Rabobank agreed to consider developing case studies as part of the CVAP project). However, there were differences in the level of enthusiasm and attractiveness of possible case studies. We have reflected on the reasons for this in the "Learnings" section of this paper.
In developing case studies, the focus was on agribusiness firms other than primary producers and input retail firms, with whom the APSRU team has interacted with in the past. Criteria guiding our selection of case studies also included:
- A range of firms at different levels of the value chain
- Firms whose activities provided a geographic and crop spread
- Applications that show present and future value to Australian industry
- Provides ability to extend APSRU’s capabilities
In exploring potential applications within industry, we have adopted an action research approach, where APSRU researchers work alongside agribusiness staff in order to test the feasibility of an agribusiness firm, if it did have the in-house capability, utilising climate forecasts and APSIM simulations in designing and implementing improved marketing, finance and insurance packages. Thus, potential applications within industry will emerge via interactions with managers working in this area.
A high emphasis has been placed on evaluating the success of and reporting of learning from innovative application of improved information and tools in agribusiness. Baseline interviews of key stakeholders were completed at the commencement of the research.
APSRU has been able to engage with several companies to explore the use of APSRU tools within their business operations via in-house case studies. Three areas of involvement are discussed below:
Crop insurance and Crop Loss Assessment
Baseline interviews revealed that insurance managers were attracted by any new technology which enables them to more objectively quantify crop-losses as part of a continual quest to find methods of quantifying claims objectively. They identified a need to replace subjective ideas by objective data collected in an independent way to stop litigation, a task requiring access to respected technology.
APSRU have worked with a crop loss assessment company to provide estimates of potential crop yield based on APSIM simulations for the purpose of crop insurance. In other words, APSIM was used to establish the crop yield loss relevant to an insurance claim either resulting from natural events or liability.
Such information has then been used to aid the resolution of disputation regarding the amount of compensation that should be made for crop losses. Initial demonstrations of the use of simulations for insurance assessment has been undertaken in five case studies over the past 12 months. It is relevant that the APSIM model contributed to a negotiated settlement claim in at least three of these cases.
As an outcome of such preliminary case studies, loss assessment firms have expressed interest in further exploring the application of APSIM and climate information in crop insurance claims. The current status of interactions is that several loss assessment cases studies have been successfully completed and there is expressed interest in pursuing commercial application of APSIM in loss assessment cases.
Prior to establishing a case study in the rural banking industry, a meeting with representatives of several banks was held to discuss the use of climate information and tools by the banking industry. There was general agreement among local and senior managers that impacts of recent El Niño events were undoubtedly factored into lending policy and that such events were having a huge influence.
Bank representatives agreed that banks are raising the demand for information from clients all the time, e.g. asking for cash flow information, and that information about monitoring and simulation related to individual clients applying for finance would be beneficial to the banking industry.
The same sentiments were expressed in the baseline interviews conducted at the start of the project. Linking APSIM physical forecasts with financial projections was seen as having real potential. As a tool for risk management and forward planning of yield, the following possible APSIM applications relevant to banking were suggested by banking representatives:
Assisting bank staff to ask more pertinent questions about the farm business when assessing for loans.
Assisting bank managers to determine which businesses are financially feasible and assess existing customers’ ability to service new liabilities.
Helping bank clients look at options other than more debt.
The highly positive reaction resulted in expressions of interest by one bank with support from National & State Managers. Participation was negotiated for four local Rural Finance Managers (RFMs) to collaborate in developing relevant case studies for the application of APSRU tools. The proposed research methodology was for APSRU researchers to work alongside the RFMs on several case studies in order to test the feasibility of the bank, if it did have the in-house capability, utilising seasonal climate forecasts and APSIM simulations in designing and implementing improved financial services. The bank’s contribution was the involvement of their staff and contact with their farmer clients.
Through this collaboration, forecasts of cropping options for 10 bank clients were undertaken prior to the 1999/2000 summer season and provided to the RFMs and the participating farmers. This activity was seen as an initial step in building confidence in the accuracy of APSIM among bank staff and clients while exploring its potential value to the bank’s business. Subsequently, one of these farmers has keenly sought further interactions. We have characterized his soil and are providing him with customized simulations for this season. He has suggested that we meet with his RFM to discuss how useful the information is viewed.
While there has been some effort placed in licensing and training bank staff in the use of APSIM, developing an in-house capacity to generate such information has proven difficult, the main reason being that the APSIM model is complex software and requires significant allocation of time of already busy people.
Our reflections on this case study to date have led us to conclude that the information on crop and seasonal climate forecasts would be of benefit to bank clients and could be valued by lending managers. However, while interested in the data on seasonal cropping outlooks for individual farmer clients, there currently does not seem to be a practical and facilitated way for the bank to generate such information for individual clients on their behalf. It would be much better for bank clients to come with this information to the bank.
The recently initiated FARMSCAPE Training and Accreditation program which has been established to train and support four agribusiness companies to use APSRU tools in their commercial advisory services, can be a way for bank clients to access these technologies. The participating bank viewed this most positively.
Value from access to APSRU tools at the farm manager level was recognised in our discussions with managers of two companies involved in large corporate farming operations with portfolio management demands across a range of districts and enterprises.
An agreed area of mutual interest was to explore whether APSRU tools could focus on the next level in management, ie. assist such companies to manage their extensive corporate farm portfolios. In the case of one company, a relevant question was "in an upcoming year of current low commodity prices, what mix of enterprises should the company consider across several properties and seasons?".
For the other company, there was interest in obtaining yield and probability predictions to assist in the management of input supplies in large-scale operations.
We have negotiated case studies with both companies to explore the use of APSRU tools (seasonal climate forecasts and APSIM) in a portfolio management role. This will involve the development of financial analysis frameworks for the assessment of portfolio management options generated using APSRU tools.
We define "Portfolio Management" in the context of a) an informed selection of enterprises and b) allocation of inputs across a multi-farm / multi-location business. Collaboration on these case studies has only been recently initiated.
Opportunities for engaging agribusiness
We believe that opportunities abound for engaging agribusiness in exploring the role for seasonal climate forecasts and simulation models in their business operations. There has not been one company that was approached that has not demonstrated at least some interest in becoming involved in this project.
All companies that have been approached for collaboration (subsequent to an initial contact) have agreed to invest their own resources (mainly in terms of staff time) in project activities. At least two companies have indicated willingness to contribute small amounts of funding to progress this research activity. The breadth of businesses within which there are possible uses for seasonal climate forecasts and simulation models covers the agricultural value chain.
We have found that the insurance industry has been most receptive to the opportunities presented by this project and is at a stage of wanting to adopt the new technologies. Simulation is seen as an improved method of loss assessment and seasonal climate forecasting combined with simulation is being assessed as a methodology for developing innovative insurance schemes. Primary interest at this stage is in loss assessment due to successful completion of a number of case studies.
Agribusinesses have existing ways of dealing with climate variability within their operational systems. Consequently, there is significant inertia against consideration of new systems. For instance, we found that companies with the need for commodity forecasting system showed a lack of enthusiasm for a new commodity forecasting system using such tools as SOI and simulation.
Companies with this need have existing systems that appear to be flexible, based on information networks, are fairly transparent and seem to work well enough for their needs. It is hard to see how a simulation forecast could do better than these systems which have such a strong field informant basis. One exception to the point above is an apparent need to forecast quality grades, an attribute missing from current systems. This area may be an opportunity for research intervention.
The engagement process
We were able to approach agribusiness companies with a proposal for collaborative research that may benefit the companies in areas of mutual interest. This was possible because the CVAP program covered APSRU costs. Significantly less interest would have been generated from approaches to companies where funding support was sought as part of the collaboration. While many of the companies may have been willing to contribute funds in addition to staff time to joint activities in the future, they do require an introduction to what’s possible before reaching this stage.
We found that engagement with agribusiness benefited from the credibility with agribusiness created by APSRU projects with farmers such as FARMSCAPE. Another important factor was being able to approach agribusiness companies at high levels within their management structure. In this regard, the Rabobank consultancy which targeted introductions to agribusiness companies was very successful – APSRU gained access to the senior executives of a number of companies.
Dealing with agribusiness requires new ways of research engagement that differ to typical researcher-funder relationships and we were faced with bridging differences between organisational cultures and protocols, ie. a bureaucratic versus a commercial culture.
These ways were unknown at the initiation of the project and have had to be learnt on the run. For example, frequent low level iterations of engagement are preferred to infrequent detailed meetings. Researchers need to be proactive in the collaboration and communication as the project is on the periphery of most agribusiness collaborators’ work and thus needs to be placed into view regularly. Operating in a learning mode versus operating in a commercial mode, both within a commercial environment creates challenges.
Participatory action research framework
The Participatory Action Research (PAR) framework used by the project has been an appropriate methodology in which to pilot interactions and the strong emphasis on evaluation has provided baselines of stakeholder views as well as feedback on progress. That said, working with agribusiness in a PAR mode is time-consuming, stressful, and confronting, (especially when activities don’t work as expected).
It is clearly easier for researchers to undertake research on the role for seasonal climate forecasts and simulation models within a policy / operations research framework – ie. one step divorced from real clients. As a consequence, there is a real lack of incentive for researchers to address this work area – evidenced by the relative numbers undertaking policy analyses versus engaging directly with clients.
Negotiations to gain approval for collaboration and to design operations have taken significantly more time than envisaged. This is often also a result of still being in an exploratory frame of mind, of not yet having set firm boundaries and not having reflected on what it is that is likely to be learned and thus needs to be. The time taken to learn about dealing with agribusiness has meant that some activities have fallen short of collaborator expectations. Perceived poor performance may spoil real opportunity and more time and investment is required to capture and respond to the major opportunities with agribusiness.
Commercialisation issues
There is a need to address legal issues, client confidentiality, and ownership of intellectual property developed throughout the case study process. Clarification of the internal and intra-organisational management procedures for developing long-term arrangements and agreements required for the use and further development of APSIM beyond the current project, will be required should results of our research warrant this. We also require a clear understanding with case study clients as to the anticipated boundaries about the point at which the case study would cease and the next stage of commercial transactions might be entered into.
Our experience to date suggests that bank lending policies, crop insurance policies, product inventories and marketing advice could all be positively influenced through better dealing with climate variability.
For example, insurance policies based on the Agricultural Production Systems Simulator (APSIM) model’s objective prediction of yields could potentially reduce claimant disputes and cut legal costs, representing significant savings to industry. Likewise, better prediction of seasonal outlooks using the APSIM model, climate forecasts and fallow water reserves would allow farmers and lenders, such as produce suppliers and banks, to negotiate individually-tailored financial packages.
While discussions with agribusiness indicate keen interest in such tools and information, they have yet to have impact on policy and operations. While clear outcomes have been seen in some areas of engagement (e.g. insurance/loss assessment), other collaborative efforts have only progressed some way to exploring the role for seasonal climate forecasts and simulation models in their business operations (e.g. banking, portfolio analysis).
More time and investment is required to capture and respond to the major opportunities with agribusiness.
The outcomes of this research will determine whether better targeted and cost-effective agribusiness services can be provided to Australian farmers through the application of climate forecasts and APSIM simulations. The two major beneficiaries of this project are companies involved in agribusiness and Australian farmers.
The major output of this project will be well documented, pertinent experiences, obtained under realistic institutional conditions, on whether better targeted and cost-effective agribusiness services can be provided to Australian farmers.
In doing this, we recognise the importance of resourcing the client interface during the life of the case studies ie. with a strong evaluation process and regular status reports. Ultimately, a successful project will see the adoption of climate forecasts and simulation within the agricultural service sector beyond the life of this project.
