ENSO climate risk: predicting crop yield variability and coherence using cluster-based PCA
Lu, W., Atkinson, D.E., Newlands, N.K. (2017). ENSO climate risk: predicting crop yield variability and coherence using cluster-based PCA. Modeling Earth Systems and Environment, [online] 3(4), 1343-1359. http://dx.doi.org/10.1007/s40808-017-0382-0
Plain language summary
The El Nino Southern Oscillation (ENSO) event of 2015-16 was one the strongest of the last 30 years and it impacted agricultural production, decreasing global wine production, while increasing yields for major crops such as wheat. Climate variation globally accounts for up to a third of the observed variability in the yield of major crops. In this study, the researchers explore how ENSO climate variability may impact wheat and barley crop yield across the Canadian Prairies. A novel mathematical model is constructed to predict crop yield under ENSO forcing and high risk areas. The model is able to identify different zones within large production regions like the Canadian Prairies, comprising a collection or cluster of census of agricultural regions. The model also predicts different levels of yield response to inter-annual climate variability based on historical yield and climate information. This work enhances our ability to predict changes in crop yield across large production areas under seasonal-to-subseasonal climate variability thereby improving accuracy and reliability of yield forecasts over the growing season. In this way, crop outlooks that incorporate such model-based forecasts of crop yield can assist farmers in making better, more informed crop management decisions, especially during strong El Nino and La Nina years where extreme drought and flooding conditions have a higher likelihood of occurrence.
The El Niño–Southern Oscillation (ENSO) has, in recent years, contributed to increases in the yields of major agricultural (annual) crops like wheat and barley in Canada. How such forcing alters the pattern of yield variation across different geographic scales and across large agricultural landscapes like the Canadian Prairies is less understood. Yet, such questions are of major importance in forecasting future cereal crop production. We explore the potential impact of ENSO on wheat and barley across the Canadian Prairies/Western Canada using a multi-scale, cluster-based principal component analysis (PCA) model that integrates machine-learning (K-means clustering) to predict areas of high climate risk. These risk areas are separable clusters of subregions that show similar ENSO-yield correlation response (spatial coherency). Benchmarking this spatial model to non-spatial models indicates that spatial coherency leads to gains in prediction skill. Incorporating spatial coherency increased the skill in predicting crop yield; reducing RMSE error by up to 26–34% (spring wheat) and 2–4% (barley). We infer that accounting for spatial coherency improves the accuracy and reliability of crop yield forecasts.