CEAL - Computational Epidemiology and Aerobiology Lab.

Research Themes
	Feature Project - Soybean Rust
		IAMS
		Ensemble Forecasting
		Historical Risk Assessment
		Epidemiology
		Spore Exposure
		Canopy Escape
		Spore Deposition
		Infection Timeline
		Spore Trapping
		PIPE
	Ragweed
		International Collaboration
		Forecasting Model
		Discussion Group
		Bibliography
	UG99 - TKKS Wheat Stem Rust
	Pest Detection
		Citrus Greening
		Grasshopper
Facilities
Personnel
Publications
Links
Home



Contact: Annalisa Ariatti

International Collaboration for Forecasting ragweed Pollen

The forecasting model

Using information from the scientific literature (Deen et al. 1998a, 1998b, 2001; Myers et al. 2004) and a general weed development algorithm, we have developed a model for predicting ragweed phenological progression including flowering. We used ragweed pollen counts from a number of stations in Europe over a period of a few years to calibrate the model. Thermal units in the soil are accumulated to predict ragweed germination and then air thermal units are accumulated to predict the other phenological stages, ragweed growth, and well as pollen release. An example of the model output is provided below. Not all our simulations of pollen production for sites and years match the observations as well as this one. There are many reasons for discrepancies between pollen production predictions and pollen trap observations that could potentially be related to model configuration, temperature data quality, and the contribution to the observation record of pollen blown from distant locations.

The chart shows simulated cumulative percent ragweed emergence (dark blue line), simulated ragweed percent growth completed (yellow line), cumulative percent pollen trapped (magenta line), and simulated cumulative percent pollen release (light blue line) for Lyon in 2000. Percent is given on the y-axis. On the x-axis (top line) is the weed stage simulation (stages 1-6) and the date (bottom lines). Weed stages are: 1-juvenile, 2-bud, 3-pistillate flower, 4-anthesis, 5-physiological maturity, and 6-post maturity.

Not surprisingly, we have found that model predictions are very sensitive to ragweed emergence date. Consequently, we would now like to establish an informal network of collaborators to provide observations of ragweed emergence (and other phenological stages) from locations in Europe and North America. In return for these data, we would like to provide participants with access to the restricted ragweed website. This site, currently under construction, will contain a map of Europe and North America with small buttons on locations where emergence data have been provided through the volunteer network. Users can view model output derived from local temperature data in a figure similar to the one above by clicking on these locations. The output would be displayed as four curves that extend to the current date, calibrated to the emergence observation from the location. Other phenological observations from the site provided by collaborators would also be superimposed on the plots.

We want to evaluate the model and gain greater confidence in its predictions of ragweed phenological stages, especially the start and end of pollination and the timing of peak pollen production before adding an aerial movement component. We hope to achieve the former over the 2007 season and include a pollen transport component the following winter.

Literature
Deen W., Hunt T., Swanton C.J. (1998a). Influence of temperature, photoperiod, and irradiance on the phenological development of common ragweed (Ambrosia artemisiifolia). Weed Science, 46:555-560.

Deen W., Hunt T., Swanton C.J. (1998b). Photothermal time describes common ragweed (Ambrosia artemisiifolia L.) phenological development and growth. Weed Science 46:561-568.

Deen W., Swanton C.J., Hunt L.A. (2001): A mechanistic growth and development model of common ragweed. Weed Science, 49:723-732.

Myers M.W., W.S. Curran, M.J. VanGessel, D.D. Calvin, D.A. Mortensen, B.A. Majek, H.D. Karsten, and G.W. Roth (2004): Predicting weed emergence for eight annual species in the northeastern United States. Weed Science 52: 913-919