English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Pre-season prediction of crop production outcomes such as grain yields and Nlosses can provide insights to stakeholders when making decisions. Simulationmodels can assist in scenario planning, but their use is limited because ofdata requirements and long run times. Thus, there is a need for morecomputationally expedient approaches to scale up predictions. We evaluated thepotential of five machine learning (ML) algorithms as meta-models for acropping systems simulator (APSIM) to inform future decision-support tooldevelopment. We asked: 1) How well do ML meta-models predict maize yield and Nlosses using pre-season information? 2) How many data are needed to train MLalgorithms to achieve acceptable predictions?; 3) Which input data variablesare most important for accurate prediction?; and 4) Do ensembles of MLmeta-models improve prediction? The simulated dataset included more than 3million genotype, environment and management scenarios. Random forests mostaccurately predicted maize yield and N loss at planting time, with a RRMSE of14% and 55%, respectively. ML meta-models reasonably reproduced simulated maizeyields but not N loss. They also differed in their sensitivities to the size ofthe training dataset. Across all ML models, yield prediction error decreased by10-40% as the training dataset increased from 0.5 to 1.8 million data points,whereas N loss prediction error showed no consistent pattern. ML models alsodiffered in their sensitivities to input variables. Averaged across all MLmodels, weather conditions, soil properties, management information and initialconditions were roughly equally important when predicting yields. Modestprediction improvements resulted from ML ensembles. These results can helpaccelerate progress in coupling simulation models and ML toward developingdynamic decision support tools for pre-season management.

Maize Yield and Nitrate Loss Prediction with Machine Learning Algorithms