MONITORING TEMPORAL CHANGES IN THE BIOLOGY OF ACID WATERS

1. Assemblages of epilithic diatoms, macrophytes, benthic macroinverte brates and fish were monitored over a 5-year period (1988-92) in ten l akes and nine streams in the U.K., as part of the U.K. Acid Waters Mon itoring Network. All organisms were categorized according to their fun ctional or morphological characteristics and integrated to describe th e food web at each site. In general, the flora and fauna of all sites were typical of oligotrophic, softwater systems subject to various deg rees of acidification. 2. Salmonids were the only fish caught at any s ite and brown trout were the dominant species. With only 5 years of da ta it was not possible to test for directional changes in fish populat ions at each site. Among sites, fish density was positively associated with pH, and this did not vary between lake outflows and streams or b etween fish of different age class (0 +, greater than or equal to 1 +) . Condition factor, reflecting fish health, was not associated with pH among sites, but was negatively associated with fish density and, on average, was higher for fish in streams than those in lake outflows. 3 . Variability in the diatom, macrophyte and invertebrate data sets wer e quantified in three ways using multivariate techniques: species turn over or replacement (temporal variation due to directional change), pe rsistence (the reciprocal of between-year variability) and within-year variability (heterogeneity between sample replicates). For all groups , turnover was relatively low and persistence was high. The least pers istent macrophyte assemblages occurred in stream sites and this may re flect high inter-annual variation in the cover of filamentous algae wh ich are prone to scouring. Within-year variability was higher than tur nover or between-year variability for the diatoms and invertebrates, a nd highest values were recorded for lake invertebrates. 4. Redundancy analysis, RDA was used to test the diatom, macrophyte, invertebrate an d food web data for evidence of directional changes over time and its significance was assessed using Monte Carlo permutation tests. These t ests appeared robust to temporal and spatial variability in the data s et. Significant trends could be identified in some data sets despite c onsiderable between-replicate and non-linear between-year variability. 5. Significant linear trends in at least one biological group were fo und at eight lake and seven stream sites. Only one lake and one stream had significant trends in all four groups. These trends represent cha nges in the flora and/or fauna, but they can be interpreted in several different ways. Only six sites showed trends that were consistent wit h our knowledge of species' responses to water chemistry: three indica ted increased acidity and three indicated decreased acidity. At only o ne site were the biological results consistent with observed chemical changes and there was disagreement at the other five. Of the other nin e sites that showed biological changes, two appeared to reflect known physical habitat disturbances; the other seven remain unexplained.