Skip to main content
ARS Home » Midwest Area » Morris, Minnesota » Soil Management Research » Research » Publications at this Location » Publication #216727
Title: A hydrothermal seedling emergence model for giant ragweed (Ambrosia trifida)

Author
item SCHUTTE, BRIAN - OHIO STATE UNIV.
item REGNIER, EMILIE - OHIO STATE UNIV.
item HARRISON, S - OHIO STATE UNIV.
item SCHMOLL, JERRON - OHIO STATE UNIV.
item Spokas, Kurt
item Forcella, Frank

Submitted to: Weed Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/26/2008
Publication Date: 7/1/2008
Citation: Schutte, B.J., Regnier, E.E., Harrison, S.K., Schmoll, J.T., Spokas, K.A., Forcella, F. 2008. A hydrothermal seedling emergence model for giant ragweed (Ambrosia trifida). Weed Science. 56:555-560.

Interpretive Summary: Late-season emergence of giant ragweed in agricultural fields can greatly hamper the decision in timing of control measures in order to achieve maximum weed control and reduce crop yield loss. Timing of field weed-control measures can be aided by computer models, provided that the model can be calibrated for the particular weed species coupled with minimal site specific data. This research examined the potential of developing a model for the emergence of giant ragweed, accounting for the multiple flushes of emergence observed particularly in Ohio agricultural fields. Weather data, soil characteristics and geographic location were used to predict the soil thermal and moisture environment. These predictions were combined with data on the base temperature and moisture values (temperature and moisture level required for germination) to create a mathematical model for the emergence of giant ragweed. The bimodal pattern of emergence was modeled with two successive mathematical (Weibull function) models. The predictions were typically within half of a day of field observations. This experiment demonstrated an approach to weed seedling emergence modeling that can be used to forecast weed emergence on a local basis from easily obtainable soil and weather data. This information will assist scientists and engineers in developing improved weed prediction models to maximize control and minimize crop yield losses due to weed pressure. This information will be of direct benefit to the farmers to enable them to cope with planning the timing of their weed-control measures.

Technical Abstract: Late-season seedling emergence of giant ragweed in Ohio crop fields complicates efforts for predicting the optimum time to implement control measures for minimizing crop-yield losses. Our objectives were to develop a hydrothermal seedling emergence model for a late-emerging biotype in Ohio and validate the model in a variety of locations and burial environments. To develop the model, giant ragweed seedlings were counted and removed weekly each growing season from 2000 to 2003 in a fallow field located in west-central Ohio. Weather data, soil characteristics and geographic location were used to predict the soil thermal and moisture environment with the Soil Temperature and Moisture (STM2) model; hydrothermal time base values were extrapolated from the literature (Tb = 2 °C, 'b = -10 MPa). Cumulative percent seedling emergence plotted as a function of calendar and hydrothermal time (HT) revealed rapid emergence in late March and April, plateauing to 60% of maximum by 400 HT; then a two to three week, 200 to 300 HT lag in emergence in May; followed by a resumption of emergence at a lower rate. The biphasic pattern of emergence was modeled with two successive Weibull models. Emergence models were validated in 2005 in a tilled and no-tillage environment and in 2006 at a separate location in a no-tillage environment. Root mean square values for comparing actual and model predicted values ranged from 8.0 to 9.9, indicating a high degree of accuracy. The model predicted 90% cumulative emergence at 1110 HT in 2005, and 1158 HT in 2006, compared to actual cumulative 90% emergence at 978 HT, and 1138 HT in 2005 tilled and no-tillage environments, and 932 HT in 2006 no-tillage environments. This experiment demonstrated an approach to weed seedling emergence modeling that can be used to forecast weed seedling emergence on a local basis according to weed biotype and easily obtainable soil and weather data.