Spatial and temporal definitions of weed patches using quantitative image analysis

Publikations-Art

Zeitschriftenbeitrag

Autoren

Krohmann, P. and R. Gerhards and W. Kühbauch

Erscheinungsjahr

2006

Veröffentlicht in

Journal of Agronomy And Crop Sciences

Band/Volume

192/1

Seite (von - bis)

72-78

Populations of Alopecurus myosuroides were studied in maize, sugar beet, winter wheat and winter barley over 5 years. Objective definitions for a 'weed aggregation' and a 'weed patch' within weed distribution maps were determined. Weedy areas in 1 year, whose distance to each other not exceeded a certain maximum value, are defined as 'weed aggregation'. Field areas, which were defined as a weed aggregation in at least 1 year, were investigated for weed occurrence in the other years. Therefore, the resulting field area was determined by the spatial expansion of the aggregations of all years. Image analysis was a helpful tool for further investigation of the visually obtained information. With this method, marginal changes within the weed aggregations could be determined. The analysed weed populations were largely stable in their existence throughout the period of the study. Still, the aggregations varied in their size depending on the crop. 'Weed patches' were defined as weeds that appear in 4 of the 5 years at the same location and were limited themselves to very small field segments. Identification of the weed patches was enabled by image analysis. It is obvious that the field segments become smaller by increasing requirements on the repetition rate of a weed population at the same location