Search

Performing large scale simulation analyses using complex process-driven models can be very time consuming and incur significant computational expense. These analyses involve generating synthetic datasets and include processes such as impacts analysis (IA) and variance-based sensitivity analysis (SA). Machine learning (ML) provides a potential alternative path to reduce computational costs incurred when generating output from large simulation experiments. We assessed the accuracy and computational efficiency of three ML-based emulators (MLEs): artificial neural networks, multivariate adaptive regression splines, and random forest algorithms, to replicate the outputs of the APSIM-NextGen chickpea crop model. The MLEs were trained to predict seven outputs of the process-driven model. All the MLEs performed well (R2 > 0.95) for predicting outputs for the training data set locations but did not perform well for previously unseen test locations. These findings indicate that modellers using process-driven models can benefit from using MLEs for efficient data generation, provided suitable training data is provided. • Machine learning emulators of process-driven models produce accurate predictions. • Some emulator types produce more consistently accurate predictions than others. • Comprehensive training data is essential to ensure emulators can predict accurately. [ABSTRACT FROM AUTHOR]

There is interest in the reincorporation of legumes and forbs into pasture-based dairy production systems as a means of increasing milk production through addressing the nutritive value limitations of grass pastures. The experiments reported in this paper were undertaken to evaluate milk production, blood metabolite concentrations, and forage intake levels of cows grazing either pasture mixtures or spatially adjacent monocultures containing perennial ryegrass (Lolium perenne), white clover (Trifolium repens), and plantain (Plantago lanceolata) compared with cows grazing monocultures of perennial ryegrass. Four replicate herds, each containing 4 spring-calving, cross-bred dairy cows, grazed 4 different forage treatments over the periods of early, mid, and late lactation. Forage treatments were perennial ryegrass monoculture (PRG), a mixture of white clover and plantain (CPM), a mixture of perennial ryegrass, white clover, and plantain (RCPM), and spatially adjacent monocultures (SAM) of perennial ryegrass, white clover, and plantain. Milk volume, milk composition, blood fatty acids, blood β-hydroxybutyrate, blood urea N concentrations, live weight change, and estimated forage intake were monitored over a 5-d response period occurring after acclimation to each of the forage treatments. The acclimation period for the early, mid, and late lactation experiments were 13, 13, and 10 d, respectively. Milk yield (volume and milk protein) increased for cows grazing the RCPM and SAM in the early lactation experiment compared with cows grazing the PRG, whereas in the mid lactation experiment, milk fat increased for the cows grazing the RCPM and SAM when compared with the PRG treatments. Improvements in milk production from grazing the RCPM and SAM treatments are attributed to improved nutritive value (particularly lower neutral detergent fiber concentrations) and a potential increase in forage intake. Pasture mixtures or SAM containing plantain and white clover could be a strategy for alleviating the nutritive limitations of perennial ryegrass monocultures, leading to an increase in milk production for spring calving dairy cows during early and mid lactation. [ABSTRACT FROM AUTHOR]

Abstract: Considerable variability in the yield response of temperate dairy pastures to the strategic use of nitrogen (N) fertiliser has been observed previously. Some of this variation could potentially be attributed to environmental effects on the available N (soil mineral N plus fertiliser N) concentration required to optimise pasture growth. Five nitrogen application rate response field experiments were undertaken in an attempt to elucidate the environmental influence on pasture mineral N requirements. A significant response to N fertiliser application rate was detected in four out of the five experiments. The absence of a significant response in one experiment can be attributed to limited water availability. Across the experiments where a significant response to N fertiliser rate was detected the available N that was required to achieve 90% of maximum yield ranged between 60 and 81kgN/ha. It is concluded that when water availability, soil temperatures and air temperatures are conducive to pasture growth an available N level of between 60 and 81kgN/ha should be targeted when N fertiliser is used strategically to optimise pasture growth. Soil N testing results should be interpreted in light of the prevailing climatic and edaphic conditions if they are to be used to guide N fertiliser rate decisions within intensively managed temperate pastures. [Copyright &y& Elsevier]

Forage brassicas have historically been used in high rainfall/irrigated temperate livestock systems, but there is increasing interest in diverse forage brassicas in drier mixed crop-livestock farming systems. Computer-based modelling is an important decision support tool used in agriculture to explore the adaptability of crops to different climates and agronomic management practices, but existing modelling tools for forage brassicas are limited to temperate environments. We parameterised the APSIM (Agricultural Production Systems Simulator) model for four forage brassica genotypes, including three diverse forage rape cultivars and a raphanobrassica. The model was calibrated using two experiments with repeated measures of biomass components, nutritive value, and leaf and canopy development. We then tested the model extensively using data from a diverse set of environments within Australian and New Zealand (23 sites across four agro-climatic zones). Model predictions of biomass were good for all the genotypes (NSE > 0.60, Nash-Sutcliffe efficiency; RMSE ~1.5 t DM/ha, root mean square error). Predictions of metabolisable energy yield were satisfactory for all genotypes (NSE 0.43–0.73; RMSE ~17.8 GJ ME/ha) but forage dry matter digestibility (DMD) were poorly predicted due to the small variation in observed data. Our robust and widely tested model can be confidently used to predict forage productivity of common and new forage brassicas across a wide range of production environments and agronomic management practices. This model will enable future work to develop a better understanding of the potential value of these important forage crops for livestock production systems. • A forage brassica model, including three diverse forage rape cultivars and a raphanobrassica were parameterised in APSIM. • Forage brassica productivity during vegetative growth can be modelled in different environments and agronomic practices. • These new capabilities enable the value of forage brassicas to farming systems to be explored in different environments. [ABSTRACT FROM AUTHOR]