A Chilling Year: Was 2013 a low winter chill year?

Rebecca Darbyshire

Rebecca Darbyshire
Research Fellow
University of Melbourne
03 5833 5909

Light flowering and reduced yields observed by growers in the 2013/14 season were not the result of insufficient accumulation of ‘winter chill’.   

In spring 2013, apple and pear growers in the Applethorpe region of Queensland noted that flowering appeared to be light. Meanwhile, further south in Victoria, cherry growers experienced a low yield season for 2013/14 across many different growing regions. Both groups queried if these problems were the result of inadequate ‘winter chill’, or in other words – a warm winter.

Winter chill is a cumulative measure of ‘cold’ temperatures over winter. Accumulation of sufficient winter chill breaks bud dormancy and allows normal flowering to proceed. In seasons where sufficient winter chill is not accumulated, light flowering and low yields can occur (Oukabli et al., 2003; Voller, 1986). So, questioning whether winter chill was inadequate in 2013 and potentially driving these observations, is fitting.

To take a closer look at winter chill for 2013, we used and assessed temperature data collected for the Australian Government funded project Crossing the threshold: Adaptation tipping points for Australian fruit trees and a HAL sponsored project Understanding apple and pear production systems in a changing climate.

Approach and analysis

To assess winter chill we used the Dynamic Chill Model (Erez et al., 1990), which converts temperature into units called ‘chill portions’. The Dynamic Chill Model incorporates many of the observed relationships between temperature and dormancy breaking when allocating chill portions. These include the positive influence of optimum chilling temperatures (6-8°C), negative effects of high temperatures (dependent on the value and length of exposure), nil contribution of freezing temperatures to breaking dormancy, chill enhancement from moderate temperatures when cycled with cooler temperatures and provisions to account for the sequence of temperature over time. Using this model, a higher total chill portion figure indicates that winter temperatures were favourable for dormancy breaking.

The Dynamic Chill Model better represents winter chill than other chill models that growers may be more familiar with, including the ‘Chill Hours Model’ and the ‘Utah’ (or ‘Richardson’) model. For further information see the paper by Luedeling (2012) who conducted a review of chill models and recommended the Dynamic Chill Model for use in chill assessments.

To consider whether winter chill conditions in 2013 were unusual, we compared total accumulated winter chill for 2012, a ‘normal’ year, and 2013, an ‘abnormal’ year at four locations including Applethorpe (QLD) and Shepparton (VIC) (Table 1). To further assist in interpretation, we also compared these with previously determined long-term averages (Darbyshire et al., 2011). For a more detailed comparison between 2012 and 2013, Figure 1 illustrates chill accumulation over the chill period.

Table 1

Chill accumulation

Figure 1 Chill accumulation for 2012 and 2013 at Plenty (TAS), Shepparton (VIC), Applethorpe (QLD) and Manjimup (WA).


The primary aim of our investigation was to assess if 2013 was a low chill year. This is not supported by the data at Plenty (TAS) or Applethorpe (QLD). Both locations accumulated similar chill portions in 2012 and 2013 and in 2013 chill accrual was close to or above the long-term averages at locations near these sites (Table 1).

In 2013, Shepparton accumulated approximately 14% less chill than 2012 and chill accrual in 2012 tracked consistently higher than in 2013 over the whole chill period (Figure 1). However, total accumulated chill in 2013 was close to the long-term average at nearby Tatura. Thus the results are indicative of 2012 being a high chill year rather than 2013 being a low chill year.

Manjimup (WA) was the only site with data that supports that 2013 may have been a low chill year. Accumulated chill was approximately 20% down on the long-term average. 2012 was also lower than the long-term average but only marginally (<5%). If winter chill accrual in 2013 dropped below varietal thresholds it is likely that flowering timing, strength and length would have differed between the 2012/13 and 2013/14 seasons. If growers observed such a difference in Manjimup this would add support to 2013 being a low chill year.

In summary, there is little evidence in the temperature data to support that 2013 was a low chill year, except at Manjimup. The observations of light pome fruit flowering in the Applethorpe region and low cherry yields in Victoria are unlikely to be the result of insufficient chill accrual and additional investigations are required to better understand the unusual conditions that were present in the 2013/14 season.

This investigation has provided some information about winter chill accrual in recent years, however much is still unknown about the process. Better comprehension of the chill mechanism and modelling of this mechanism is fast becoming critical with warmer winters likely, due to future global warming. If winter chill accrual is reduced below varietal thresholds, productivity will likely reduce if adaptive actions are not taken. Many important research questions remain to be answered to help growers better prepare for future conditions.

For instance, when does chill accumulation begin? Does this differ with species/ variety? How can we measure this in the field? Can current chill models account for this? What is the chilling requirement of important crops (e.g. Pink Lady™ apple)? Is there a rootstock influence on chill accumulation? Do varietal chill requirements differ with location? If so, why? How can we exploit species characteristics and/or site conditions to adapt to future lower chill conditions? Two related and current projects hope to shine some light on these questions: Crossing the threshold: Adaptation tipping points for Australian fruit trees” led by the University of Melbourne and “Understanding apple and pear production systems in a changing climate” led by Department of Agriculture Fisheries and Forestry, Queensland.


  • Darbyshire, R., Webb, L., Goodwin, I. and Barlow, S., 2011. Winter chilling trends for deciduous fruit trees in Australia. Agricultural and Forest Meteorology, 151: 1074-1085.
  • Erez, A., Fishman, S., Linsley-Noakes, G.C. and Allan, P., 1990. The dynamic model for rest completion in peach buds. Acta Horticulturae, 279: 165-174.
  • Luedeling, E., 2012. Climate change impacts on winter chill for temperate fruit and nut production: A review. Scientia Horticulturae, 144: 218-229.
  • Oukabli, A., Bartolin, S. and Viti, R., 2003. Anatomical and morphological study of apple (Malus X domestica Borkh.) flower buds growing under inadequate winter chilling. Journal of Horticultural Science and Biotechnology, 78(4): 580-585.
  • Voller, C.F.P., 1986. Predicting rest-breaking: Principles and problems. Deciduous Fruit Grower, 36(8): 302-308.
By |March 28th, 2014|Climate change|

About the Author:

Research Fellow, University of Melbourne