Summary

This project used Mesonet weather data to validate APSIM-predicted flowering times for wheat and barley in South Australia’s Mid North and Mallee. Field trials showed flowering was more variable in the Mid North than the Mallee, largely due to elevation. APSIM simulations using both Mesonet and SILO PPD temperature data matched observed flowering dates, confirming the value of these data sources. Elevation was a key driver, with flowering delayed 3-4 days per 100m increase.

Background

Flowering time is a key driver of yield potential in wheat and barley. South Australia’s Mid North and Mallee regions present diverse growing environments with variations in elevation, temperature, and seasonal conditions. Understanding how these environmental factors influence flowering timing is essential for selecting suitable varieties and determining optimal sowing dates. Tools such as APSIM are widely used to simulate crop development stages, but their accuracy depends on reliable environmental data inputs. Localised weather networks such as the Mesonet system, and interpolated data sources like SILO’s patch point data (PPD), can provide the temperature and rainfall information needed to improve decision-making and long-term modelling. However, a robust understanding of regional environmental variation and how it affects cereal phenology is necessary to support accurate simulation and inform crop management strategies.

Research Aims

The core objectives of the project were to:

• Characterise the environmental differences across the Mid North and SA Mallee to inform the planting time and variety selection of cereal crops.
• Validate real-time thermal and vernalisations data available on Mesonet and APSIM simulation data with on-ground truthing.

In The Field

To explore regional variation in flowering time and assess the accuracy of simulation tools, small plot observation trials were established across South Australia’s Mid North and Mallee regions. A network of 24 trial sites was selected to capture a broad range of environments, including different elevations and climatic zones. Trials were aligned with 71 existing Mesonet automated weather stations to allow comparison with real-time temperature and rainfall data. Each site included two sowing times to evaluate development across early and later planting windows (26-28 April and 1-3 June). Consistent seeding practices were used, with irrigation applied at sowing to support even germination.

Three commonly grown cereal varieties – Scepter (wheat), RGT Planet (barley), and Longsword (winter wheat) – were sown at all sites and monitored through key development stages, including heading and flowering.

Alongside field monitoring, crop simulation models (APSIM) were run using different weather data sources to compare predicted flowering dates with field observations. This setup provided a framework to evaluate environmental influences and test modelling accuracy in real-world conditions.

Results

The project found that flowering time varied more across the Mid North than in the Mallee, with observed and simulated data showing a 20-25 day range in the Mid North compared to 10-12 days in the Mallee. APSIM simulations closely matched observed flowering dates, particularly for early sowing, and performed similarly when using Mesonet weather data or interpolated Patch Point Data (PPD).

PPD simulations tended to predict earlier flowering dates than Mesonet, likely due to slightly warmer temperature inputs. Despite minor differences, both data sources correlated strongly for temperature. Elevation proved to be a reliable predictor of flowering delay, with every 100m increase resulting in a 3-4 day delay. This relationship held across long-term simulations and offers a useful way to explain regional variation in flowering.