Exploring the green break of season and green date over northern Australia

Abstract

Context: In northern Australia, livestock production relies heavily on dryland pastures, whose growth is strongly controlled by wet season rainfall. Knowledge of the likely timing of the first productive pasture after the long dry season – marked by the green cover onset (GCO) - can help graziers establish an appropriate stocking rate based on the available fodder at the end of the previous growing season. Objective: This study focuses on the ‘green break of season' date (GBOS), defined as the first day after 1 September when a threshold amount of rainfall (e.g., 50 mm) is accumulated over a 3-day period. This rainfall-based metric aims to coincide with the annual onset of effective pasture growth (GCO) in northern Australia. Methods: Using robust model-derived estimates of green pasture cover at a representative location in northeastern Queensland we compute the Green Cover Onset (GCO), defined as the first day after 1 October on which modelled green cover reaches or exceeds a specific threshold. This study explores the relationship between GCO and the GBOS for different 3-day accumulated rainfall thresholds (10–80 mm, in increments of 10 mm). We further explore the ‘green date’ (GD), defined as the 70th percentile of the distribution of GBOS dates, calculated over a long historical period (1900–2023) in northern Australia using daily rainfall observations from the Scientific Information for Land Owners (SILO). We then analyse, how the phase of the El Niño-Southern Oscillation (ENSO) influences the GBOS distribution. Results and conclusions: The strongest relationship between GBOS and GCO is found defining the GBOS as the first occurrence of 50 mm of rainfall accumulated over 3 days (R2 > 0.94). This correlation is stronger than that between the commonly used Northern Rainfall Onset (50 mm accumulated after 1 September) and GCO (R2 = 0.62), with a regression slope closer to 1 and a y-intercept closer to zero, indicating a better one-to-one relationship with the GCO. This suggests that the GBOS is a more effective indicator for estimating the onset of productive pastures. Additionally, our analysis reveals that during El Niño years, the reliable GBOS (the 70th percentile of GBOS for ENSO-influenced years) occurs slightly later than the GD for all years, with no significant difference. In contrast, during La Niña years, it occurs significantly earlier. This pattern is consistent across regions in northern Australia, showing that El Niño delays and La Niña advances the northern wet season. Significance: This study paves the way for the development of a seasonal GBOS prediction product, which will help livestock producers in making more informed and effective management decisions.