Crop Residue Simulator

The AFO residue simulator estimates the proportion of the crop residue in each category using trial data of sequential animal liveweights grazing crop residue of a crop with known yield. There are 2 steps to using the residue simulator:

1. The discrete animal liveweights (from trial data) are converted to a continuous function by fitting a quadratic function to the liveweights and the number of grazing days since grazing started (LW = a GD2 + b GD + c). This approach is as developed by Thomas et al. 2021. Using grazing days as the independent variable allows trials that had varying stock numbers to be included.

   The fitted quadratic curves are a better comparison of the grazing achieved from each crop or site than liveweight versus date, because it includes the impact of stock numbers. However, it is not accounting for the differences in the yield of the crops between sites. Yield is accounted for in step 2.

2. Run the AFO residue simulator (CropResidueSim.py) which uses inputs from Property.xlsx. The inputs include

   1. Crop yield

   2. Coefficients from the quadratic LW function

   3. The commencement day of the trial after the harvest date, so that feed deterioration can be estimated.

   4. Description of the animals that were grazing in the trial, including genotype (SRW), age, fleece length (for insulation), the physiological state of ewes (pregnant or lactating with litter size)

   5. Description of the environment for calculation of chill effects (wind speed, rainfall and temperature).

   6. Level of supplement consumed.

As part of the Crop Residue Simulator, the stock generator (see Livestock for more information) is run for a range of feed qualities beginning at the week corresponding to the start of grazing in the trial. The liveweight change (LWC), metabolizable intake (MEI) & dry matter intake (DMI) is recorded for each feed quality assessed. The LWC of the animals in the trial, as estimated from the fitted quadratic, is compared to the range of feed qualities assessed to estimate the quality and quantity of the diet that was consumed by the animals in the trial. These calculations are carried out for a sufficient period that stubble (or fodder crop) from all the AFO stubble categories will have been grazed. The fitting of the quadratic to the measured LW means that the calculation of intake can be extended beyond the duration of the measurements made in the trial. The quantity of each stubble category is then calculated as a proportion of the total residue mass at harvest, and these values are the inputs stored in stubble.xlsx.

Multiple paddock trials that can be used to generate inputs for a single crop. If multiple trials are activated the results (proportion of residue in each category) is averaged. This is controlled in the inputs.

The module writes the answers to an excel book which is referenced by AFO - this means this module only needs to be run if you make changes to stub inputs or if you change the inputs/formulas that calculate lwc in the generator (it could be made so that the stubble generator is run for every trial but this will slow AFO and potentially confuse the interpretation of results).

27/2/2022: currently there is only liveweight data for GSW region. Thus CWW uses the same stubble. This would be improved by running a stubble trial in the wheatbelt and recording liveweight every week.

@author: young