Mitigation or disturbance? Effects of liming on macroinvertebrate assemblage structure and leaf-litter decomposition in the humic streams of northern Sweden

Summary

Stream liming can alleviate the effects of anthropogenic acidification but may itself lead to substantial ecosystem perturbations. Acidity in the humic streams of northern Sweden largely arises from natural causes but liming is still extensively practised, with uncertain ecological outcomes. Macroinvertebrate assemblages and leaf-litter decomposition in response to liming was investigated in seven such humic streams.

Study sites: Seven streams (Bladtjärnsbäcken, Hörnån, Oxbäcken, Pålböleån, Stridbäcken, Tvärån, Vadbäcken), each limed at a single point using a dosing tower, were studied in the province of Västerbotten, northern Swedish. Here liming is undertaken to mitigate flood-associated acid episodes, with release of lime powder triggered by rising water levels. Accordingly, liming commences each year during snowmelt in spring, usually stopping over summer then recommencing in response to autumn precipitation, and ceasing when streams freeze in winter.

Two litter decomposition studies were undertaken during periods of liming, in autumn 2003 and spring 2004. Within each stream, three study sites, each comprising a 50 m stretch of riffle, were established: i) in an unlimed upstream site (US) less than 100 m above the dosing tower; (ii) a mixing zone (MZ) within 100 m downstream of the tower, where the powder is incompletely dissolved; and iii) a section further from the tower at the goal site (GS) i.e. the point where the liming is anticipated to affect a pH change. Streambeds were predominantly rocky, and study sites similar within streams (except at Tvärån where the MZ substrate was sandy). Stream width did not vary significantly among sites within streams. Streams flowed through boreal forest or a mixture of forest and agricultural land. All US and GS and most MZ sites were shaded by riparian strips, comprising conifers (especially Norway spruce Picea abies), grey alder Alnus incana and birch Betula spp.

Litter decomposition: Litterbags, containing 3 g of air-dried grey alder leaves picked prior to leaf-fall, were arranged in six evenly spaced pairs at each study site. A pair consisted of one fine (0.5 mm) mesh bag designed to assess the microbial component of decomposition through exclusion of most macroinvertebrate leaf shredders, and one coarse (10 mm) mesh bag, open to shredders. Bags were left longer in autumn (c. 40 days) to compensate for slower decomposition, in spring they were left for around 30 days. After retrieval, leaves were washed in the laboratory, dried, weighed and then ashed in a muffle furnace. The mineral ash remaining was used to determine leaf ash-free dry mass (AFDM), thereby correcting for residual sediment. Insects found in the coarse bags were preserved in 70% ethanol, with shredders categorized and identified.

Benthic samples: Five replicate Surber net samples were collected from each site during 20–21 May 2004 to characterize benthic macroinvertebrate assemblages. Captured insects were identified to the lowest level possible (generally species) and allocated to a functional feeding group.

Stream abiotic characteristics: Stream temperature, pH and conductivity were measured in the field at the beginning, midpoint and conclusion of both studies. Current velocity was measured at the midpoint and conclusion upstream of each litterbag pair. Water samples were collected for analysis at the conclusion of the autumn study and at the beginning, midpoint and conclusion of the spring study. Alkalinity was assessed and disolved organic carbon was quantified. Water samples were analysed once each season for common cations.

Abiotic variables: In the spring, pH, alkalinity, calcium and silica were elevated by liming, as was conductivity (mean ± SE; US 23.6 ± 2.1, MZ 30.5 ± 1.6, GS 34.3 ± 3.0 µS/cm). Concentrations of aluminium, other metal cations and DOC were reduced. Similar effects were apparent in autumn, but significant only for pH, alkalinity and calcium (autumn conductivity 28.1 ± 1.8 µS/cm). Mean water temperature (autumn 7.1°C, spring 6.7°C) nor mean current velocity (spring 16 cm/s) varied among sites within streams.

Upstream acid status varied. During autumn, US alkalinity was greater for Bladtjärnsbäcken and Hörnån (146.2 ± 20.0 µ/L, pH 5.3) than the other streams (29.8 ± 10.2 µ/ L, pH 5.1). In spring, Bladtjärnsbäcken US alkalinity (73.3 ± 1.0 µ/L) was higher than other streams (7.3 ± 3.2 µq/L), with pH also relatively high (5.8 compared with 5.1).

Leaf decomposition and colonizing shredders: Leaf decomposition data from those bags that were retrieved intact, were similar. Decomposition in the fine bags was enhanced at the GS, whereas the C–F fraction was reduced. Decomposition was faster in spring than autumn, and the effect of liming on the C–F fraction appeared stronger in spring. Decomposition in the coarse bags appeared unaffected by liming in either season.

In autumn, neither shredder abundance, evenness, species density nor richness varied with liming. In spring, liming had no effect on shredder abundance or species density but evenness was decreased and species richness declined following liming, but not significantly so. GS shredder richness was lowered relative to the US for Bladtjärnsbäcken (5-times lower than upstream), Tvärån and Pålböleån (3-times lower) and Stridbäcken (1.3 times lower). However, richness was unaffected in Oxbäcken and was 1.3 times greater at the Vadbäcken GS.

In autumn, calcium, aluminium, pH and shredder species density, abundance and evenness were negatively associated with the C–F fraction, while temperature and shredder richness were positively associated. DOC, pH, temperature and calcium (VIP = 0.69) were positively associated with the F fraction, with conductivity and aluminium negatively associated. In the spring C–F analysis, temperature, alkalinity and shredder species richness and density were positively related to decomposition, while silica, calcium, DOC and shredder abundance and evenness were negatively related. Only aluminium was negatively related to the F fraction in the spring analysis, with all other important variables positively related.

Shredder assemblages differed according to liming in spring but not autumn. Species contributing most to liming-associated dissimilarity in spring included three Amphinemura stonefly species, which increased in abundance at the GS, and four limnephilid caddisfly species, which, with the stonefly Nemoura cinerea, decreased in abundance following liming. Similar patterns were apparent in autumn, with four stonefly species increasing in abundance following liming, while Potamophylax sp. caddisflies and Protonemura meyeri stoneflies decreased.

Spring benthic samples: Spring macroinvertebrate assemblages differed according to liming and among streams. Taxa contributing most to assemblage dissimilarity between US and GS which increased in abundance following liming included: Simuliidae, two species of leaf-shredding Amphinemura stoneflies, two Baetis mayfly species, larval Elmis aenea beetles, two chironomid subfamilies and the predaceous caddisfly Rhyacophila nubila. Species that decreased included: the shredders Potamophylax cingulatus (Trichoptera), Nemoura cinerea (Plecoptera) and water louse Asellus aquaticus (Isopoda), along with the mayfly Leptophlebia marginata, tanypod chironomids and the caddis Polycentropus flavomaculatus.

Liming did not affect benthic macroinvertebrate abundance, species richness, species density or evenness. Among functional groups, collector-gatherers were impacted by liming, with abundance reduced in the mixing zone (mean abundance US 8.3 ± 1.7, MZ 2.9 ± 0.7, GS 10.8 ± 2.5), but shredder, predator, scraper and filterer abundances were unaffected overall.

Conclusions: Impacts of liming on northern Sweden's humic streams which are little affected by anthropogenic acid deposition, are contrary to the goals of the liming programme, with evidence for a decrease in the diversity of functionally important leaf-shredding macroinvertebrates coupled with a decline in shredder-mediated leaf decomposition. Species characteristic of such systems have substantial conservation value in their own right. Accordingly, any potential positive effects of stream liming (e.g. on recreational fish populations) in regions characterized by extensive natural acidity need to be balanced against its impacts on the ecology of these distinctive ecosystems.


Note: If using or referring to this published study, please read and quote the original paper. Please do not quote as a www.conservationevidence.com case as this is for previously unpublished work only. The original paper can be viewed at: http://blackwellpublishing.com/submit.asp?ref=0021-8901