Skip to main content
ARS Home » Midwest Area » Morris, Minnesota » Soil Management Research » Research » Publications at this Location » Publication #317440

Title: Predicting field weed emergence with empirical models and soft computing techniques

item GONZALEZ-ANDUJAR, JOSE - Consejo Superior De Investigaciones Cientificas (CSIC)
item CHANTRE, G - Universidad Nacional Del Sur (UNS)
item MORVILLO, C - Universidad De Buenos Aires
item BLANCO, A - Universidad Nacional Del Sur (UNS)
item Forcella, Frank

Submitted to: Weed Research
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/21/2016
Publication Date: 12/1/2016
Publication URL:
Citation: Gonzalez-Andujar, J., Chantre, G.R., Morvillo, C., Blanco, A.M., Forcella, F. 2016. Predicting field weed emergence with empirical models and soft computing techniques. Weed Research. 56:415-423.

Interpretive Summary: The ability to predict when weed seedlings emerge is useful to farmers, crop advisors, crop scouts, and others who are involved with managing these plants. Researchers have strived for the past 25 years to develop easy-to-use models that predict weed emergence with varying levels of success. Variation in success was due, in part, to limitations of the modeling methodologies available and employed during this time period, which almost always employed various nonlinear regressions. Now, however, new methods, collectively known as "soft computing techniques" eliminate some of the limitations and restrictive assumptions of the older methods. The use of two of these new methods, Artificial Neural Networks and Genetic Algorithms, are highlighted in this review. Soft computing techniques have fewer conceptual and statistical limitations than nonlinear regression, and they appear to have greater universal application for widespread weeds. This review is expected to provide useful information to weed scientists who conduct research on predictive weed emergence models specifically and weed dynamics models generally.

Technical Abstract: Seedling emergence is the most important phenological process that influences the success of weed species; therefore, predicting weed emergence timing plays a critical role in scheduling weed management measures. Important efforts have been made in the attempt to develop models to predict seedling emergence patterns for weed species under field conditions. Empirical emergence models have been the most common tools used for such purpose. They are based mainly on the use of temperature, soil moisture and light. In this review, we present the more popular empirical models, highlight some statistical and biological limitations that could affect their predictive accuracy and, finally we present a new generation of modeling approaches to tackle the problems of conventional empirical models, focusing mainly on soft computing techniques. We hope that this review will inspire weed modelers and that it will serve as a basis for discussion and as a frame of reference when we proceed to advance the modelling of field weed emergence.