Farming systems adaptations to climate change and extreme climatic events

 Securing global food supply under increasingly variable climates will be  one of the grandest challenges facing humanity in the 21st century.  Increased frequencies and intensities of extreme temperature and  rainfall will have ramifications for the production environments of most  crops, including barley, which globally is the fourth most prevalent  cereal crop. There is thus an urgent need to develop adaptations that  reverse adverse climate change effects on the productivity of barley  crops. This is particularly so in Australia, where the rate of genetic  yield gain of cereals is amongst the lowest in the world, primarily due  to challenging climatic conditions.
This thesis investigated  strategies for mitigating detrimental effects of current and future  climate change on barley production at a global scale, with special  attention on Australia. Adaptation strategies included systemic changes  in agronomy (e.g. altering flowering time through sowing time or  genotypic duration) and improving genetic tolerance by breeding to allow  successful cultivation under likely future environmental conditions.  Field experiments were conducted at representative sites in Western  Australia and Tasmania, where summer conditions are forecast to become  hotter during the 21st century.
A modelling study (Chapter 3)  identified optimal sowing and flowering times for rainfed barley in  water limited environments throughout the Australian barley growing  regions. A genotype × environment × management (G×E×M) factorial  analysis showed that optimal flowering periods (OFP) were driven by  environmental conditions more so than the genotype such that the  relative importance of solar radiation, frost, heat and water stresses  varied significantly with location. The OFP was earlier (mid-August to  late-September) in Western Australia and South Australia, while OFPs  were generally later (mid-October to mid-November) in Tasmania and  Victoria. Knowledge of OFPs based on long-term abiotic stresses will  allow breeders to develop genotypes with phenological durations that are  pertinent to each location. Better knowledge of OFPs will also allow  growers to match genotype with sowing time for their location/s,  minimising the combined risk of frost, heat and water stresses.  Collectively, these adaptations would be expected to be conducive to  increased maximum yield potential.
A little explored area associated  with impacts of extreme climatic events on cropping systems is that  related to waterlogging. To address this knowledge deficit while also  quantifying OFPs in waterlogging-prone regions, controlled-environment  experiments (Chapter 4) were conducted with a range of modern barley  genotypes differing in their waterlogging tolerance. In these  experiments, the extent to which yield was dictated by the timing and  duration of waterlogging stress relative to crop phenology was examined.  It was found that crop heading was the most susceptible period to  waterlogging, with yield losses during this period primarily attributed  to reductions in spikelet fertility and grain weight. At earlier stages,  yield loss caused by waterlogging were primarily caused by reduced  spike number and to a lesser extent kernels per spike. Phenology was  delayed 1-8 Zadok stages at the end of waterlogging treatments, with the  waterlogging-susceptible cultivar Franklin showing the greatest delays  in crop development.
Chapter 4 provided a solid foundation for the  derivation and improvement of the internationally-renowned systems model  APSIM in Chapter 5. To account for physiological effects of  waterlogging on phenology and photosynthesis, new algorithms were  developed for commercial Australian barley genotypes and coded within  APSIM. The improved version of the model was used to conduct a genotype  by environment by management analysis (G×E×M) that was used to examine  the effects of soil type, phenology and genotypic tolerance to  waterlogging on crop development and yield under current and future  climates. It was shown that climate change will reduce waterlogging  stress and shift forward OFPs significantly (26 days earlier on average  across locations) under the emissions scenario RCP8.5 at 2090. While  waterlogging stresses diminished, alleviation of this stress was unable  to prevent yield reduction due to severe high temperature stress  exposure (−35% average reduction in yield across locations, genotypes  and sowing dates). Chapter 5 also showed that seasonal waterlogging  stress patterns under future conditions will be similar to those  occurring historically. Earlier sowing and adoption of waterlogging  tolerant genotypes alleviated yield penalty caused by waterlogging by up  to 26% and 24% under historical and future climates.
To understand  how climate changes affect global barley production and the timing and  magnitude of waterlogging stresses (Chapter 6), rainfed barley cropping  systems were simulated on a global scale using the improved version of  APSIM to investigate the risk posed by soil waterlogging on crop  production. A new paradigm was used to classify the typology and  frequency of waterlogging-stress across environments using statistical  clustering. This process was conducted using 4,104 global simulations  from 27 global climate models, 38 sites, 2 genetic traits, and 2  management inputs of sowing date for RCP8.5 at mid- and end-century.  Using the new paradigm it was shown that extreme waterlogging stress  will cause large reductions in barley yields at the global scale, even  though typologies of waterlogging stress under future climates were  relatively similar to those occurring historically. Application of our  new paradigm reveals that averaged yield penalty caused by waterlogging  ranged by up to 3-11% across GCMs, sites and climate horizons. Holistic  systems adaptations such as altering sowing time and planting  waterlogging tolerant genotypes mitigate yield penalty caused by future  waterlogging up to 14%. We uncover a dangerous shift of waterlogging  typologies under future climates towards flowering time. This may have  serious implications, because cereals are highly sensitive to  waterlogging during their reproductive phase. We highlight the  serendipitous outcome wherein future waterlogging typologies will be  similar to those occurring in the recent past, suggesting that  adaptations designed today will also suited future agricultural systems,  provided adaptations are applied within the same typology.