Developing a Predictive Model for Spring Germination of Smooth Crabgrass (Digitaria ischaemum) and Annual Bluegrass (Poa annua) in Michigan.

Accurate prediction of the germination time of annual grasses is critical to maximize performance of preemergence herbicides. Soil temperature optima for germination of smooth crabgrass (Digitaria ischaemum (Schreb.) Schreb. ex Muhl.) and annual bluegrass (Poa annua L.) have been reported; for smooth crabgrass the time of germination has been closely linked with growing degree-days (GDD). This study was conducted to determine if GDD alone or in combination with soil temperature, soil moisture, daily air temperature cycling (max-min>10°C), or freezing events (air temp≤0°C) could be used to accurately predict germination time. The spring germination time for smooth crabgrass and annual bluegrass were monitored on an Owosso Sandy Loam (Fine-loamy, mixed, semiactive, mesic Typic Hapludalfs) in 2002, 2003, and 2008 at two locations in East Lansing, MI. Forward parameter selection methods were applied to determine the final model for each species. The final model for smooth crabgrass germination included soil temperature and a Boolean argument that reset the model if a freezing event occurred. The final model for annual bluegrass included soil temperature, daily air temperature cycling, and freezing events. GDD was not the most accurate method of predicting germination time. The final models were more accurate than soil temperature alone for both species.