Changes of Ecological Traits of Weed Species as Affected by Fallow Period in Paddy Fields

This study describes a pattern of weed species occurrence according to fallow period in paddy fields. The fields were abandoned in October 1995 and we investigated weed occurrence in September and December 1, 2, and 5 years after the abandonment of paddy fields. The weed species, which occurred in paddy fields 5 years after the fallow period were 14 families and 29 species. Among these weed species, 11 families and 25 species occurred in September, and 6 families and 7 species in December. The amount of weed occurrence and shoot dry weight decreased with the increasing length of the fallow period in September and December except for shoot dry weight of September. Generally troublesome weeds in paddy fields such as Echinochloa crus-galli, Aneilema keisak Hassk. Scirpus juncoides, and Cyperus difformis occurred within 1 year after the fallow period, and the weeds decreased gradually within 2 years after laying fallow. The weeds did not occur 5 years after the fallow period. The order of dominant summer weed species, ranked from highest to lowest was：Echinochloa crus-galli > Bidens frondosa > Leersia japonica 1 year after fallow; Leersia japonica > Bidens frondosa > Echinochloa crus-galli 2 years after fallow; and Leersia japonica > Miscanthus sinesis Anderss. > Eragrostis multicaulis Steud. 5 years after fallow. The dominant winter weed species were Juncus leschenaulti Gay and Conyza canadensis (L.) after 1 and 2 years of fallow period, and Alopecurus aequalis Ohwi after 5 years. When we classified these as life type, as we had for summer weed species, the annual weed species in the number of species, number of individuals, and shoot dry weight (except for 5 years after fallow) had a tendency to be more dominant than in perennial weeds. On the other hand, the number of weed species and the number of individual annual weeds in winter weeds were greater in quantity than that of perennial weeds. However, perennial weeds turned out to be higher in shoot dry weight after 2 fallow years than annual weeds. When identified according to leaf type, the number of broad-leaved weed species, which occurred in summer 1, 2, and 5 years after the fallow period increased 7-7-12 weed species respectively. Sedge weed species were 3-2-0, and grass weed species changed to 3-4-3 during the given period. The order of dominant weeds in the number of individual and shoot dry weight was grass weeds > broad-leaved weeds > sedge weeds. As for winter weed species, the number of weed species was greater in broad-leaved weeds, the number of individuals was greater in grass weeds, and broad-leaved weeds showed an increase from 96.8 to 97.4% in shoot dry weight after 1 and 2 fallow years. The ratios of broad-leaved and grass weeds showed a 47.1 and 52.9% change respectively after 5 fallow years