The utility of solaria (1 by 1-m plastic sheets) to predict densities of a few weed species in summer crops has been demonstrated previously, but needed further research to be adopted by farmers and advisors. We tested the method to detect important weeds in Argentina and Minnesota, and determined the minimum number of solaria required to predict the presence of emerged weed seedlings in the forthcoming growing season. Three experiments were performed in Buenos Aires Province, Argentina, and one in Minnesota. Solaria were placed in fields with different previous crops and soil management: no tillage (two fields) and conventional tillage (two fields). Preceding crops were corn (one field), wheat (one field), and double-cropped wheat/soybean (two fields). After weeds were enumerated, solaria were removed, sunflower (one field) and soybean (three fields) were planted, and weeds later assessed in each crop. Results indicate that one solarium per 1.9 ha can detect common lambsquarters with 95% confidence within the next summer crop. For other species, one solarium per 4.2, 1.2, 1.0, and 1.8 to 2.7 ha (depending upon field site) for large crabgrass, prostrate knotweed, wild buckwheat, and green foxtail, respectively, was required. The low cost and simplicity of assessment make this technique more suitable than that of soil seed-bank samples to predict weed emergence. The number of solaria required to forecast weed infestation levels confidently is sufficiently low that their use may be justified, especially in small fields of high-value crops.

Nomenclature: Common lambsquarters, Chenopodium album L. CHEAL; green foxtail, Setaria viridis (L.) Beauv. SETVI; large crabgrass, Digitaria sanguinalis (L.) Scop. DIGSA; prostrate knotweed, Polygonum aviculare L. POLAV; wild buckwheat, Polygonum convolvulus L. POLCO; corn, Zea mays L.; soybean, Glycine max (L.) Merr.; sunflower, Helianthus annus L.; wheat, Triticum aestivum L