Assessing climate vulnerability of historical wheat yield in south-eastern Australia's wheat belt.

Agricultural vulnerability assessment is a comprehensive and powerful analytical tool to locate hotspots with states of susceptibility to harm and powerlessness of agricultural system. It plays an important role in guiding policy makers to plan and implement adaptation practices to mitigate potential climate risks to crop. However, due to the diversity in the methodology of vulnerability assessment, there are still knowledge gaps in assessing and comparing crop vulnerability to climate in different regions of the world, including Australia. Our main objectives were to: (1) present a vulnerability analytical method for wheat yield, which can be applied to different areas where long-term crop yield and climate data are available. (2) quantify temporal changes of the vulnerability of wheat yield to historical climate. (3) identify the most vulnerable region in study area to provide guidance for climate mitigation. Our study developed an indicator-based method using exposure, sensitivity, and adaptive capacity to assess the vulnerability of crop yield. We used the long-term recorded wheat yield data, combining with comprehensive exposure index to assess climate vulnerability of historical yield with a case study area of south-eastern Australia's wheat belt. The results showed that from the 1930s to the 1990s, both climate exposure and sensitivity had large inter-annual variations with no significant trends detected. However, adaptive capacity increased by 34% from 1930s to 1950s, 54% from 1950s to 1970s, and 54% from 1970s to 1990s. By contrast, climate vulnerability across the wheat belt decreased by 13% from 1930s to 1950s, 15% from 1950s to 1970s, and 33% from 1970s to 1990s. This is mainly due to increased adaptive capacity with the improvement of agronomic management practices, technological and socio-economic progress. We identified the areas with the highest vulnerability were in the northwestern parts of wheat belt while the least vulnerable areas located in the southeast. We expect that these identified vulnerable hotspots can be used by different landholders to allocate natural resources and policymakers to plan the priority mitigation to adapt to climate change in the local scale. Moreover, the method of vulnerability assessment used in this study can be applied to other regions around the world where long-term crop yield and climate data are available. [Display omitted] • A modified indicator-based method was developed to assess wheat yield vulnerability. • Machine learning was used to obtain weights of climate indices when calculating exposure. • Historical vulnerability of wheat yield decreased as the improved adaptive capacity. • The most vulnerable areas were the north-western parts of the wheat belt. • This vulnerability assessment method can be applied to other regions with long-term data of crop yield and climate. [ABSTRACT FROM AUTHOR]