.png)
In April Farmanco Facts, we welcome Ben Percy to the Albury office and introduce the updated UI for CashPeek. We review Octopusbot out of the 22/23 season, look at Spatial Data Layers and Satellite Imagery to Predict Canola Yield, and discuss Analysing and Benchmarking GHG Emissions.
Subscribe to Farmanco Facts via the Farmanco Facts App (iOS devices), through our website store or contact the Mundaring office to set you up. Farmanco Facts is available printed, and electronically via PDF and devices.
Key point review:
Octopusbot Review & Forecast
Ryan Duane (Grain Marketing Consultant)
Machine learning is being used increasingly in grain and oilseed markets to assist with supply, demand and price forecasting.
In 2022, Farmanco subscribed to Octopusbot. Strange name, excellent at getting its tentacles into a huge amount of data and converting that data into insightful and visual outlooks to aid with grain marketing decisions.
Octopusbot had some very accurate predictions for both supply, demand and price in 2022/23.
We take an early look at some of Octopusbot’s 2023/24 forecasts
Machine learning is starting to be used more widely by many within the grain and oilseed industry to convert huge amounts of data into actionable insights.
Spatial Data Layers
Alice Butler (Precision Ag Consultant)
Using methodology defined by Oliver et al. 2019, EM and gamma surveys could be used to define sand/sandy earths and duplex soils.
Late season biomass imagery correlated well to wheat yield data for the 2020 and 2022 growing seasons.
It is always important to ground truth data layers to ensure accuracy and correlation with the paddock’s soil characteristics and grower observations.
Combining and comparing data layers over time should provide a more comprehensive understanding of the paddock, leading to more effective management practices and increased productivity.
Satellite Imagery to Predict Canola Yield
Derrick Chan (Student, UWA Ag Science Masters) reviewed by Alice Butler
By converting stacks of Sentinel-2 satellite images into spectral indices, we can better understand key canola phenological events in the season.
These events are indicators of crop condition and can be characterised as length of season, length of flowering period, intensity of flowering, and peak biomass.
Of particular interest here is the use of peak Normalised Difference Yellowness Index (NDYI) values.
Crop yield predictions are of particular importance for assessing food security. Accurate pre‑harvest predictions can assist farmers to tailor site specific crop management. By projecting end-of-season yield, it may be possible for farmers to undertake corrective measures to increase yield.
Analysing & Benchmarking GHG Emissions
Stacey Bell-Crookes (Farm Management Consultant)
There is little correlation between the volume of emissions produced on a per hectare basis, and the profitability of a business.
Lowering emissions does not necessarily mean a business will experience an economically negative impact.
The spread of businesses by rainfall zone shows that volume of emissions is based on management decisions rather than location of the business.
Emission intensity fluctuates annually as it is dependent on seasonal rainfall, temperature, crop rotation, yield, crop residues, disease, management approaches to nitrogen and fertiliser, lime, chemical and fuel use.
Rainfall does not directly affect the amount or intensity of emissions. Emissions are not solely determined by rainfall but can be managed and controlled by choices made on the farm.
It doesn’t matter what beliefs you hold about GHG emissions as an individual, the impact they have on the planet or Australian agriculture’s contribution. What matters to those of us in Australian agriculture is what the legislators and markets that we operate within believe about these topics.
