Page 2 – Digiscape Future Science Platform

Science Wednesday: Big Data in farming – have we worked it out yet?

With the Australian AgriFood Data Exchange implementing initial use cases, it's timely to revisit the 2018 Digiscape paper "Is big data for big farming or for everyone? Perceptions in the Australian grains industry." (5 Jan 2022)

Science Wednesday: soils information for “somewhere” as well as for “anywhere”

In this 10-minute video, Kirsten Verburg takes us through the science behind an app that helps farmers to predict plant-available water content for specific paddocks, by combining their own local observations with landscape-level information provided centrally. Deeper questions about digitally-enabled extension arise... (22 Dec 2021)

Science Wednesday: the shapes of soils

Our recent paper in Geoderma uncovers the simplicity beyond the complexity of soil attribute profiles, and points a way toward "profile-centric" digital soil maps (15 Dec 2021)

Science Wednesday: detect, consolidate, delineate = field boundaries

In today's video, Digiscape Early Career Researcher Franz Waldner takes us through the complexity lying behind the simplicity of CSIRO's ePaddocks field boundaries product (8 December 2021)

Science Wednesday: visualizing uncertainties across space

User demand to understand the uncertainty in agritech predictions can only increase. This Wednesday's's paper describes Vizumap, an R package for displaying spatial predictions in tandem with their uncertainties (1 Dec 2021)

Science Wednesday: how to buy your carbon

CSIRO economist David Evans explores how to run auctions to purchase greenhouse gas mitigation from landholders (24 Nov 2021)

Science Wednesday: no need to go off-road

How to monitor the landscape to get the most out of machine learning-derived maps? Franz Waldner & colleagues showed that roadside sampling works fine (17 Nov 2021)

Science Wednesday: user experience meets spatial statistics

What happened when we tried to put sophisticated statistical techniques in the hands of on-farm experimenters? We found we needed to combine user-experience and statistical expertise (10 Nov 2021)

Science Wednesday: ML unpacks water quality for prawns

A paper in Biosystems Engineering has used machine learning techniques to examine not just which water quality parameters are most important to the growth and yield of prawns in a commercial setting, but also when in the season each water-quality parameter is important (3 Nov 2021)

Science Wednesday: provenance for scientific workflows

"Explainable" is an adjective that needs to apply to all our predictions, not just to AI. In this 9-minute talk, CSIRO Early Career Fellow Anila Butt describes a system for capturing the provenance of computations (27 Oct 2021)