School of Agriculture, Food and Ecosystem Sciences - Theses
Permanent URI for this collection
Search Results
1 - 1 of 1
-
ItemWinter and spring phenology of Australian pome fruit in historical and future climates( 2013)Successful agricultural production relies on favourable environmental conditions in combination with sound management strategies and practices. The advent of anthropogenically induced climate change is likely to modify climate variables, potentially influencing the productivity of these systems. Pome fruit trees are vulnerable to future climate changes due to their perennial nature, extended juvenile phase and long productive lifespan (decades). Australian growers are further susceptible as Australia has been identified as particularly exposed to future climate change. To explore this potential exposure, interpretation of expected changes to the climate as applicable to Australian pome fruit trees was considered in this thesis. Specifically, two key phases in the growth cycle were evaluated, winter chilling and spring flowering phenology. These phases are both temperature dependent and therefore implications of changes to temperature conditions were the focus of the research. Prior to consideration of climate future impacts, assessment of plant responses under current conditions is required. For the two phenophases considered, substantial knowledge gaps persist in relation to physiological mechanisms as driven by environmental conditions. As a result, many methods have been developed to estimate these processes based on observed responses to temperature. In this study, several methods were applied in all the analyses to consider the implication of methodological selection on the results obtained. Winter chill was analysed at 13 locations across southern Australia representing the major pome fruit growing regions. Four different models were used to calculate chill historically and into the future (0-7.2°C, Positive Utah, Modified Utah and Dynamic models). Historically, many locations recorded notable declines in accumulated chill regardless of chill model selection (Orange, Lenswood, Tatura, Yarra Valley and Bacchus Marsh). Other locations have remained stable with no location exhibiting a consensus increase in chill across the tested models. Large season-to-season variability was observed at all locations and for all chill models. Comparison of output between the chill models indicated that the 0-7.2°C model frequently behaved differently to the other models. Projected changes in winter chill were assessed according to +1, 2 and 3°C increases to mean global temperatures. Reduced chill accumulation in the future was found for all locations with notable regional differences. The sites in Western Australia were found to bear the greatest impact with a substantial reduction in chill expected. The chill models tended to interpret changes differently with the 0-7.2°C model generally being most sensitive to changes in temperature followed by the Utah models and then the Dynamic model. The impact of climate on spring flowering was also considered. Collation of historical flowering phenological records of pome fruit in Australia was an important contribution of this research. No such information has been analysed for Australia and only one other study exists for the Southern Hemisphere. Observed relationships indicated flowering advancement of between 4.1 to 7.7 days per degree Celsius increase in mean spring temperature. Compared with similar phenological research in the Northern Hemisphere, the changes in timing in relation to temperature were often shallower in Australia, although the significance of observed hemispheric differences was not clear. To determine relationships to temperature, a sequential chill-growth model as well as correlation to mean springtime temperatures were used. The sequential chill-growth approach proved superior, with coefficients of determination between 0.49 and 0.85, indicating the inclusion of chill conditions are important for spring phenology modelling. Projections of future flowering timing were created using the relationships determined in the historical analysis combined with +1, 2 and 3°C increases to mean global temperatures. Both of the previous sequential chill-growth and the springtime temperature modelling approaches were used. The springtime temperature model predicted advancement in flowering timing for all datasets. With 1°C warming the sequential chill-growth model recorded similar flowering timing to current conditions for many of the species. Greater warming, however often tended towards later emergence. The timing of projected flowering was related the relative dominance of the protraction of the chill period versus the contraction of the growth phase. Finally, the implication of changes to flowering timing was considered in relation to frost risk. Results from projections of both phenology modelling approaches were used in combination with two assumptions about future frost conditions; current conditions or frost is modified with climate change. Depending on the combination of phenology method, frost assumption and level of warming, frost risk was predicted to notably increase, remain similar or decrease. Overall, chill accumulation in Australia is likely to decline. The changes reported are sensitive to chill model selection for Australian conditions, consistent with previous assessment made for other regions. Further research using the 0-7.2°C model is not recommended whilst greater uptake of the Dynamic model is encouraged. Modelling of flowering phenology, especially under climate perturbed conditions, illustrated flowering phenology model selection is important. Although the sequential chill-model model performed better than the springtime temperature approach, reservations still remain in employing this method for climate impact statements. Future research directed at greater understanding of physiological processes is required to better characterise current physiological responses and produce robust climate impact assessments.