Skip to main content
ARS Home » Pacific West Area » Corvallis, Oregon » Forage Seed and Cereal Research Unit » Research » Publications at this Location » Publication #221855

Title: Spatial Clustering of Grass Seed Weeds

Author
item Mueller Warrant, George
item Whittaker, Gerald
item YOUNG, WILLIAM - OREGON STATE UNIVERSITY

Submitted to: Seed Production Research at Oregon State University
Publication Type: Experiment Station
Publication Acceptance Date: 3/31/2007
Publication Date: 4/30/2007
Citation: Mueller Warrant, G.W., Whittaker, G.W., Young, W.C. 2007. Spatial Clustering of Grass Seed Weeds. Seed Production Research at Oregon State University. Department of Crop & Soil Science. Ext /CrS 126 P. 53-59.

Interpretive Summary: Statistical analysis of spatially-referenced data can provide us with far better understanding/appreciation of complex phenomena than more traditional, nonspatial approaches normally achieve. Numerous technical hurdles were overcome in the process of transforming an extremely large, nonspatial database of 10 years of OSU Seed Certification field inspection reports into visual displays of weed severity without compromising the confidentiality of individual growers. Weed species varied greatly in their tendency to appear in clusters as opposed to random patterns, with German velvetgrass, field bindweed, roughstalk bluegrass, and annual bluegrass clustering most strongly. Weeds tended to appear in groups based on crop species for the major crops grown in the area, and were also linked to soil type and soil chemical and physical properties. Distances over which individual weed species clustered varied from a 5-kilometer scale for bentgrass to a 59-kilometer scale for mayweed chamomile. Crop management implications include the possibility/desirability of tailoring weed control practices by geographic location based on likeliness of individual weed species occurring and a better understanding of reasons behind recent increases in weeds such as wild carrot. The public release of weed hot spot maps should help growers, consultants, and other interested parties deal with these weeds.

Technical Abstract: Statistical analysis of spatially-referenced data can provide us with far better understanding/appreciation of complex phenomena than more traditional, nonspatial approaches normally achieve. Numerous technical hurdles were overcome in the process of transforming an extremely large, nonspatial database of 10 years of OSU Seed Certification field inspection reports into visual displays of weed severity without compromising the confidentiality of individual growers. Weed species varied greatly in their tendency to appear in clusters as opposed to random patterns, with German velvetgrass, field bindweed, roughstalk bluegrass, and annual bluegrass clustering most strongly. Weeds tended to appear in groups based on crop species for the major crops grown in the area, and were also linked to soil type and soil chemical and physical properties. Distances over which individual weed species clustered varied from a 5-kilometer scale for bentgrass to a 59-kilometer scale for mayweed chamomile. Crop management implications include the possibility/desirability of tailoring weed control practices by geographic location based on likeliness of individual weed species occurring and a better understanding of reasons behind recent increases in weeds such as wild carrot. The public release of weed hot spot maps should help growers, consultants, and other interested parties deal with these weeds.