Agricultural Production Systems Simulator APSIM Simulates v yield

Agricultural Production Systems Simulator (APSIM) Simulates: v yield of crops, pastures, trees, weeds. . . v key soil processes (water, N, P, carbon, p. H) v surface residue dynamics & erosion v range of management options v crop rotations + fallowing + mixtures v short or long term effects v one or two dimensions v high software engineering standards v BUT, not yet pests nor diseases

APSIM - developmental goals v Production u v Fate u sought to retain yield prediction in relation to management options and environment (c/f - CERES, CROPGRO models) of the soil resource sought valid long-term simulation of key soil processes (c/f CENTURY, EPIC) v Impacts u and profit off-farm predict loss of soil, water, nutrients off-site (c/f - EPIC)

APSIM - some statistics v Development team 7 programmers / model support staff u 12 scientist / modellers u v User base 180 licensed users u 9 countries, 4 continents u v Product Suite ca. 450, 000 lines of code u 4 languages u 38 modules u 12 interfaces or major tools u

Developing our knowledge & capability - APSIM modules Crop/pasture/tree wheat sorghum sugarcane chickpea mungbean soybean barley groundnut maize sunflower hemplucerne fababean canola lupin mucuna cowpea Pinus radiata Eucalyptus sp. cotton - CSIRO PI pearl millet - ICRISAT pigeonpea - ICRISAT Soil. Wat SWIM Soil. N Soil. P Soilp. H Solute Residue Manure - ICRISAT Management Sowing Tillage Irrigate Fertilize Intercrop/mixture competition

Multiple user interfaces – e. g. APSFront interface

APSIM has been used to simulate … Some examples

…physiological processes Pigeonpea qualitative photoperiod response

…plant organs Tiller leaf area in millet

…crop growth & development Growth & development of pigeonpea

…yield of experimental crops Cowpea Chickpea 5000 1200 yields 4500 2500 600 1: 1 line 300 Grain (g/m 2) 0 300 600 900 1200 Observed Prediction wheat grain maize grain chickpea grain mungbean grain cowpea grain stylo biomass 3500 3000 2500 2000 1500 y = 0. 87 x + 221. 44 R 2 = 0. 77 1000 Biomass (g/m 2) 0 Predicted 4000 900 500 0 0 1000 2000 3000 4000 5000 43 111 60 47 15 63 regression line slope 1. 07 0. 98 ( 0. 04) 0. 90 ( 0. 07) 1. 07 ( 0. 10) 0. 93 ( 0. 08) 0. 84 ( 0. 06) 1: 1 line regression y = 1. 0631 x - 70. 964 2000 R 2 = 0. 7924 1500 1000 500 0 0. 0 500. 0 1000. 0 1500. 0 2000. 0 2500. 0 Observed n Mungbean 3000 R 2 intercept -13. 0 -5. 5 ( 240) 163 ( 172) -27. 2 ( 128) -31. 6 ( 34. 6) -131. 7 ( 171) 0. 79 0. 85 0. 76 0. 72 0. 91 0. 78 3000. 0

…yield of commercial crops v APSIM tested against data from commercial farms v Crops include cotton, sorghum, mungbean, wheat, chickpea

… yield of smallholder crops Maize response to N in Malawi Maize response to N & manure in Kenya Maize response to N at Makoholi

… N response in smallholder crops Testing simulation of maize response to N at Makoholi over 7 seasons 1991 -1997

… seasonal perspectives How representative were the seasons 91 -98 at Makoholi?

… yield of crops in rotation Lines = predicted Symbols = observed Wheat-Sorghum Long Fallow rotation

… soil water of crops in rotation Wheat Sorgham Wheat-Sorghum Long Fallow rotation

… ET of crops in rotation 93 Wheat, 94 -97 Lucerne measured in lysimeter

… legume rotation effects Maize response (TBM) to fertiliser N following pigeonpea, India

… consequence of crop rotations $GM drainage wheat-mungbeansorghum-chickpea rotation

… soil organic matter changes Farming systems on a vertisol at Dalby, Qld.

…crop-weed competition Maize – volunteer stylo

…response to manure application High & low quality manure applied to maize

… response to N, P fertilizer & manure Maize response to P rates in Kenya Response to N, P and manure, India

… “on-farm” constraints Response to 36 kg N/ha

… agroforestry systems Enabling landholder assessment of the productivity and risk of commercial agroforestry investment on grain farms in Australia’s medium to low rainfall regions

… change in wheat production under climate change

… but can you use such technical information with farmers?

YES…but the information needs to be made relevant to farmers’ realities

Source: Peter Carberry CSIRO, Australia Click the back button on your browser to return to the main menu