Aquatic plant diversity in the irrigation ponds of a terraced paddy field area in western Japan: The relationship between distribution and environment

We investigated aquatic plant distribution and environmental factors in 38 irrigation ponds of four valleys of a terraced paddy field area in Awaji City, Hyogo Prefecture, western Japan. Sixteen species including two threatened and two exotic species were recorded in 24 ponds, and no species in the other ponds. The average number of species per pond was 1.7 (range, 0 to 8). There were few ponds with more species. Although species number was not related to location of ponds within each valley, it was significantly different among valleys. The results of detrended correspondence analysis (DCA) also showed no trends in species composition of ponds between and within four valleys. However, significant nestedness of aquatic plant communities was detected using 'BINMATNEST', and the threatened species occurred in ponds with more species. The results of Moran's I statistics of the present/absent data of Lemna aoukikusa and Potamogeton oxyphyllus showed the significant positive spatial autocorrelation within valleys for only the former species, of which populations were observed in surrounding paddy fields. We analyzed the relationships of species number with environmental factors, including conductivity, NO2sup(-), NO3sup(-), NH4sup(+), PO4sup(3-) and mean water level, using generalized linear models (GLMs). The model with the lowest AIC included conductivity, NO3sup(-) and mean water level, and the latter two variables had significant effects on species number. Adding autocovariate explaining spatial autocorrelation slightly improved model fit, but its significant effect was not found. Both NO3sup(-) and mean water level varied among valleys in different way from species number, indicating that only each of the variables do not explain the difference in species number among valleys. The final GLM models explained 27.5% of the variation in species number, suggesting other environmental factors may also be important. The results of this study suggest that active conservation efforts and water management are necessary in order to maintain the species richness of ponds in our study area. Considering the nestedness of aquatic plant communities and occurrence of threatened species in ponds with more species, conservation of those ponds should have priority. Controlling the exotic species should also be important and indispensable.