Presentation at RELU Farm Level Workshop 2009
Download as PPT, PDF0 likes219 views
1) The document describes a linear programming model that was developed to analyze the impact of biodiversity policy on farmers and the consequences of farming on biodiversity. 2) Key aspects of the model include quantifying measures of biodiversity, incorporating field boundary features and their effects on spatial geometry, and developing criteria to model farmer decision behavior. 3) Results showed that farmers on lighter soils can increase overwintering crop residues more readily than those on heavier soils, but doing so increases financial risks. Promoting spring crops alone does not necessarily provide more overwintering residues. Higher commodity prices tend to reduce residues by prioritizing timely harvesting.
Presentation at RELU Farm Level Workshop 2009
- 1. RELU FARM- level Modelling Land-use/ environmenta l interactions U. Stirling 30-6-08 Biodiversity and Agricultural Production Planning by LP Daniel L. Sandars & E. Audsley
- 2. Structure • Background • Methodological challenges • Results • Summary & (Discussion)
- 3. Declining farmland birds A political objective to halt the decline
- 4. Arable farming • Why a decline? – reduced winter food resources • Intensification leads large-scale homogenisation in the landscape • Herbicides lead to few weeds surviving to harvest • High capacity machinery leads to timely harvest and the swift removal of residues and stubble • Increased winter sown cropping leads to less over wintering stubbles
- 5. Decision variables • Crop choice • Rotational combination • Operational choice • Operational timing • Input choice • Input timing
- 6. Policy questions • How would farmers react, in the long term, to change? • Climatic • Technical • Financial • Regulatory • Social • How does the cropping, environmental emissions and biodiversity change? • What would make a particular management action appealing to farmers? • For example, how will farmers respond to increasing prices of biofuel crops. What will the unintended consequences be?
- 7. Model-based farm-level policy impact analysis • Linear Programming, such as Silsoe whole-FARM Model (SFARMMOD), is well established at predicting the optimizing behavioural response of farmers in response to choice and change in prices, technology and regulations. • Recently extensions include environmental pollution, such as nitrate leaching as multiple objectives to be constrained or minimised • We extend this modelling approach to predict the impact of biodiversity policy on farmers and the consequences of farming on biodiversity
- 8. Soils and Weather Workable hours Profitability (or loss) Crop and livestock outputs Environmental Impacts Possible crops, yields, maturity dates, sowing dates Silsoe Whole Farm Model Linear programme, important features timeliness penalties, rotational penalties, workability per task, uncertainty Machines and people Constraints and penalties
- 10. Structure • Background • Methodological challenges • Results • Summary & (Discussion)
- 11. Key tasks Three main types of model extension are envisaged 1) Quantified measures of biodiversity, which could include four mammal species, indicator bird species, and weed species. 2) Field boundary features and the effects of spatial geometry. These are habitats that support biodiversity. 3) Incorporate sets of criteria to explain and predict the decision behaviour of a population of land managers
- 12. Weeds, birds and mammals • A wide varied of detailed ecological models • Habitat association models of birds • Difference equation and Markov chain models of weed dynamics • Game theory models of bird populations and winter feed availability • Development of a single metric ‘biodiversity units’? • Fitting these to an LP requires meta-modelling to enable each to be quantified for the set of all farm plans
- 13. LP model of weeds, etc cR dijC iwQ wQ rRxCaQQ w dci w dji w i W dci dci w dcidji dji w dji i i w i W cropprevioustoduechange at timecroponoperationtoduechange cropforweedofpopulationdefault weed,ofpopulationpredictedtotal where ,, ,, ,, ,, ,,,, ,, ,, = = = = ++= ∧ ∧ ∑∑∑
- 14. boundary features Spatial geometry effects • The length and depth of field boundary per cropped hectare effects field shape which effects the efficiency of field work • A model of field work efficiency is being developed to quantify the effects and determine significant non-linear behaviour • At a larger scale the increase of contract farming operations can mean entire farms are in a single crop in a given year
- 15. Non linearity! Can we maintain linearity and model the effects of promoting an increase in hedges and probable reduction in field size
- 16. Decision Making Behaviour 1 • Profit maximising (long-term net farm profit) accounts well for the aggregate production behaviour of farmers, but what about conservation behaviour? • At farm level decision making behaviour may differ due personal values, views on future prices, risk, and the information available • Conservation behaviour may involve the understanding of objectives such as ‘stewardship of the land’, and ‘professional pride/identity’, etc • Aggregate behaviour can be built up from a distribution of farmer values. Is this a better decision model?
- 17. Decision Making Behaviour 2 • Multiple Objective Decision Making (MODM) can be used. It is based on Multiple Attribute Value Theory (MAVT) • The two common implementations are • Goal Programming (GP): Objectives are satisfied by obtaining a series of hierarchical goals • Multiple Objective Programming (MOP): Objectives are involved in a weighted trade-off • Which is better …both or ANP or Stated Choice or…?
- 18. Structure • Background • Methodological challenges • Results • Summary & (Discussion)
- 20. Sensitivity to commodity prices 0 50 100 150 200 250 80% 90% 100% 110% 120% Change in oilseed commodity price Averagecropping,ha/250hafarm Rotational setaside Dried Peas W.OSRape Spring Barley Winter Barley Spring Wheat Winter wheat Stubble Prices £/t: W Wheat £78, S Wheat £81, Barley £73, Peas £87, Rape seed £150 Sandy clay loam with 595 mm annual rainfall
- 21. Promoting spring crops v. stubbles R2 = 0.2824 0 10 20 30 40 50 60 70 80 90 0 50 100 150 200 250 Spring crops, ha/ 250 ha farm StubblesoverwinteringtomidFeb., ha/250ha
- 22. wintering stubbles are one measure of ‘stewardship’ 0 50 100 150 200 250 £- £10,000 £20,000 £30,000 £40,000 £50,000 £60,000 Net farm profit, £/250 ha Stubblearea@14thFeb.,ha £- £5,000 £10,000 £15,000 £20,000 £25,000 Risk,£Totalabsolutedeviation/ 250ha Clay 700mm rainfall, stubble area Sand 500mm rainfall, stubble area Clay 700mm rainfall, risk Sand 500mm rainfall, risk
- 23. Structure • Background • Methodological challenges • Results • Summary & (Discussion)
- 24. Summary • Farmers on lighter and dryer soils can increase the amount of stubble available more readily than those on heavier wetter soils. • However, in doing so the risks rise sharply • Promoting spring crops does not in itself provide more stubble. • Raise farm incomes do to higher prices tends to reduce winter stubble availability because the benefits of timeliness progressively outweigh machinery costs
- 26. Discussion • Can we maintain linearity and its high utility • Can we identify the ‘missing’ attributes? Do they exist? Would we be better quantifying the farmers true full economic costs? • Can we quantify and model them for all farm plans? • Can we elicit preferences and value functions? • Can we generalise for all farmers for some farmers? • Can readily evaluate future, as yet unspecified choices by estimating their attributes only?