We analysed long-term monitoring data on submerged macrophytes and water chemistry from 666 Danish lakes > 1 hectare and mean depth < 3 m, encompassing a total of 1447 lake years. Our aim was to describe how plant cover (COV), plant volume inhabited (PVI) and species richness related to physical and chemical and environmental variables. Boosted regression tree (BRT) analyses revealed that chlorophyll a, Secchi depth and depth were the strongest predictors of COV and PVI. Chlorophyll had a strong negative effect up to 50 mu g/l, whereas the changes related to Secchi depth and depth were more gradual and covered more of the gradient. Macrophyte species richness was best predicted by lake area and alkalinity, with chlorophyll a, nutrients and colour having significant but less marked effects. For chlorophyll a, 78% of the observed variance could be explained by the BRT model, with the most powerful predictors being both phosphorus and nitrogen, but with significant additional effects of plant cover and alkalinity. Our analyses revealed limited direct effect of nutrients on macrophyte abundance, but an indirect hierarchical effect of nutrients mediated through chlorophyll a with additional interactive effects by plant cover itself, alkalinity, mean depth and colour.