The macro-scale influence of climate on crop production in the Fitzroy catchment of Central Queensland

Climate is the most significant factor in determining plant growth and productivity (Buchdahl, 1996). Recent studies with general circulation models suggest that one of the consequences of increase in greenhouse gases may be greater variability in the climate of a region. Given the possibility of increased variations in climatic conditions, it is deemed that the study of the relationship between climate and crop yield would be useful. For a system which is too complex to be explained by the laws of physics, typically, the determination of the relationship is done through multiple regression analysis. The latter approach is adopted in this study.

Yields of four crops - wheat, barley, sunflower, and cotton - are related to minimum, maximum, and average values of rainfall, temperature, and humidity at planting time, flowering time, and harvesting time for the Fitzroy catchment region of Central Queensland. The most significant climatic factors affecting crop yield are identified. Regression models have been developed which are capable of forecasting the variability of crop production in the region due to climatic variations. Consistent genetic trends and use of existing management practices are assumed in such forecasting.

Furthermore, the complex interaction of climate on crop growth is addressed through the application of the statistical tool of path analysis. From the application of stepwise multiple regression analysis techniques, three explanatory variables are identified as important for each of wheat and barley yields and only one variable is identified for each of sunflower and cotton yields. The regression models are discussed in Section 5.3 of Chapter V. By using path analysis it is found that additional variables have indirect influence on wheat and barley yields. Multiple regression analysis could not identify these variables, which are significant but in an indirect way. The path diagram models are discussed in Section 5.4 of Chapter V. Path analysis allowed us to explore the existence of important variables with indirect influence, and the technique provided us with greater information in predicting changes in yields from the vagaries of climate than is ordinarily available from econometric and regression models.