Weed population sequential sampling plan and weed seedling emergence pattern prediction
Date of Completion
Agriculture, Agronomy|Biology, Botany|Biology, Ecology|Agriculture, General|Environmental Sciences
Weed integrated pest management (WIPM) is becoming more and more important. Intimate understanding of weed spatial distribution and weed emergence pattern is the starting point of developing any successful WIPM program. As an applied activity, weed research results should be readily applicable by farmers and weed managers. The objectives of this study were to model weed seedbank and seedling spatial distributions, and weed emergence patterns, and consequently develop weed sampling plans and predict weed emergence patterns. Study was conducted to show that the weed seedbank and weed seedlings possess spatial distribution. Negative binomial (NB) distribution fitted spatial distribution well. The sequential sampling plans based on NB are easy to use by weed scouts using only a hand calculator. The research also explored the possibility of incorporating spatial distribution into linear regression models which predict weed infestation from soil nutrient analysis. Research was also conducted to formulate a temperature-dependent population level model for predicting pigweed (Amaranthus spp.), lambsquarters (Chenopodium album L.), and crabgrass (Digitaria sanguinalis (L) Scop.) emergence patterns. The predictive program was developed by combining the poikilotherm equation with the Weibull function. The parameters were estimated using laboratory datasets. The program was validated against independent field observations and results were accurate. The unique feature of this program is that farm managers with a PC can access the program and make weed control plans. ^
Zou, Guangyong, "Weed population sequential sampling plan and weed seedling emergence pattern prediction" (1999). Doctoral Dissertations. AAI9918106.