Alpine streams are important aquatic systems, hosting an unique set of ecosystems and species, representing early and delicate phases of the freshwater cycle (Ward, 1994), and constitute interesting environments for scientific and recreational purposes. Environmental characteristics of Alpine lotic systems differ from low-land habitats: they generally have cold, highly oxygenated and turbulent water, steep gradients, coarse substrata, low channel stability, limited nutrients and organic matter availability, and a shorter growing season (Mani, 1990; Hieber et al., 2002).
Various typologies of streams can be identified in the Alps. Kryal streams, of glacial origin, have low water temperatures, high concentrations of suspended particles, and higher channel instability (Milner et al., 2010). Because of their uniqueness, their extreme sensitivity to climate change, and their affinities to arctic streams, study efforts on Alpine kryal systems have recently increased (Maiolini and Lencioni, 2001; Jacobsen and Dangles, 2012). However, in the Alps these systems are less common than other stream types, which still remain little studied, such as krenal (groundwater-fed) and rhithral (snow-fed) streams (Ward, 1994).
Improving our knowledge on the ecology of Alpine streams is important, because these environments are subject to an increasing human pressure (Maiolini and Bruno, 2007; Bizzotto et al., 2009), such as widespread distribution of hydroelectric facilities, water diversion for irrigation, drinking, snow making, and other uses (Bruno et al., 2010; Zolezzi et al., 2011) and channel alterations (Bona et al., 2008). Moreover, these direct human impacts should be considered in the broader context of global climate change. Alpine aquatic environments are extremely sensitive to climate change; warming events reduce snow-pack and ice cover and increase water temperatures (Brown et al., 2007), with important hydrological, morphological and ecological consequences, such as biodiversity loss and decreased functional efficiency (Acuña et al., 2008; Milner et al., 2009; Finn et al., 2012).
Since the formulation of the River Continuum Concept (RCC; Vannote et al., 1980) it is well known that the main energy source of small, low-order lotic systems is constituted by allochthonous inputs of organic matter, mainly terrestrial leaves (Gessner et al., 1999; Tank et al., 2010). Streams in mountainous regions could represent an exception: above the tree line, a ragged line that separates areas with different growth limiting factors for trees (Körner, 1998), stream catchments have scarce terrestrial vegetation, and consequently reduced input of allochthonous organic matter. Although Alpine streams are important ecosystems, threated by human impacts and climate change, little information is available on the balance of allochthonous and autochthonous energy inputs. The main objectives of this study were: i) to describe seasonal variation of in-stream benthic chlorophyll a and coarse particulate organic matter amounts over an annual cycle; ii) to analyse the relationships between these energy inputs and the characteristics of the macroinvertebrate community in an Alpine non-glacial stream.
The Po River is the longest Italian lotic system (652 km). It rises from a spring below the northwest side of the Monviso Mountain, in the Cottian Alps of north-western Italy. The study site was located at Pian della Regina (1750 m asl, UTM 357750-4951547, Crissolo, NW Italy - Fig. 1), 2.5 kilometres downstream from the spring. Here, the Po River is a typical second order mountain stream, with a catchment area of approximately 16.0 km2 that ranges from an altitude of 1750 to 3841 m asl. Many lakes of glacial origin are present in the highest area of the catchment but have no direct surface connections to the Po River. At the study site, the stream is an open system flowing across a plain of glacial origin, characterized by extensive Alpine meadows pointed by large erratic boulders and scattered Larix decidua Mill., 1768. Riparian vegetation is composed almost exclusively by Poaeceae and Ericaceae.
The study was performed realising monthly samplings in a homogeneous 100 m stream reach. Substrate in the study site was homogeneous and composed mainly by coarse elements (approximately boulders 50%, cobbles 40%, gravel 10%).
Water temperature was measured hourly with a data-logger (HOBO® Water Temp Pro, 0.01°C accuracy). Conductivity, dissolved oxygen, pH, chemical oxygen demand (COD), biochemical oxygen demand (BOD), total phosphorus, discharge, streambed width and water column height were measured on each sampling date (Tab. 1).
Benthic samples were collected monthly from January to December 2010 using a Surber net (250 µm mesh size; 0.062 m2 area). Eight samples were collected randomly during each sampling date, accounting for a total area of 0.50 m2. The substrate collected, including macroinvertebrates, was mixed and stored in plastic jars containing 4.0% formalin. The location of each Surber sample was recorded and a recolonisation period of a minimum of two months was allowed between collection events. Organisms were counted in the laboratory, using a Nikon SMZ 1500 stereomicroscope, and identified to species or genus level, except for some Trichoptera, Coleoptera and Diptera, which were identified to the family level using taxonomic keys (Ruffo, 1977-1985, Campaioli et al., 1994, 1999). To assess the trophic characteristic of the community, each taxon was assigned to a Functional Feeding Group (FFG: scrapers, shredders, collector-gatherers, filterers and predators) according to Merritt and Cummins (1996).
Coarse particulate organic matter (CPOM) collected in the Surber samples was hand-sorted and divided into two categories: larch (L. decidua) litter and other CPOM (grass fragments, leaf litter, and miscellaneous organic particles), to account for the different palatability of coniferous litter to macrobenthonic fauna (Campos and González, 2009; Collen et al., 2004). In the laboratory, larch litter and the other CPOM were air dried for 24 h, oven dried (105°C) for additional 24 h, and weighed with an electronic scale (accuracy±0.001 g).
Autochthonous inputs were assessed by measuring chlorophyll a, according to Uehlinger et al. (1998). During the monthly samplings, ten boulders were randomly chosen and algal film was collected from an area of 256 cm2 (16x16 cm) using a brush. The samples were transferred into opaque falcon tubes with 20 mL of deionised water, and transported in ice to the laboratory. Chlorophyll a was extracted in 90% acetone according to Steinman and Lamberti (1996) and its concentration measured using a DUSERIES 500 Beckman spectrophotometer.
During the study, mean stream width and water depth were 5.13 m and 24 cm, with a maximum of 7.5 m and 51 cm and a minimum of 3.75 m and 0.11 cm, respectively. Water temperature varied considerably (Fig. 2), with a mean annual temperature of 4.40°C, a minimum of 0.01°C in January, February and December and a maximum of 13.0°C in mid-August. Discharge was highest in spring and early summer following snowmelt; mean discharge was 0.49 m3/s, with a minimum in December (0.07 m3/s) and a maximum in May (1.78 m3/s - Fig. 2). Chemical characteristics are in the range of what reported for similar lotic environments in this Alpine area (Fenoglio et al., 2007).
A total of 120 benthic algae samples were examined (ten samples per month). Mean annual benthic chlorophyll a concentration was 0.25 (±0.11 SD) µg/cm2. Significant monthly differences were detected in the concentration of this pigment (Friedman ANOVA F=47.2, P<0.001; Fig. 3): chlorophyll a values were lowest in late spring (0.03 µg/cm2±0.02 SD in June) and winter (0.13 µg/cm2±0.08 SD) and increased significantly to a peak in the late summer (0.42 µg/cm2±0.10 SD in August and similar values in July).
The mean weight of benthic coarse particulate organic matter was 11.4 g/m2 (±15.6 SD dry mass). Total amount of CPOM in the streambed showed high seasonal variability, with a maximum in May (58.2 g/m2) and a minimum in December (1.40 g/m2). At the study site, CPOM was mostly constituted by grass fragments, vegetal particles and small roots. Larch litter constituted only the 6.6% of the total CPOM, with a minimum in July (0.04 g/m2) and a maximum in November (2.25 g/m2 - Fig. 4).
We collected 29,950 organisms belonging to 44 aquatic taxa, representing 13 families from 10 orders (Ephemeroptera, Plecoptera, Trichoptera, Diptera, Coleoptera, Tricladida, Haplotaxida, Lumbriculida, Heterostropha, Basommatophora). Macroinvertebrate density was 4317.5±617.7 ind m–2 (mean±se). Benthic communities were dominated by aquatic insects (96.3% of total abundance), mainly represented by Ephemeroptera (38.0%), Plecoptera (29.8%), and Diptera (22.2%). The most abundant FFG was scrapers (40.1%), followed by shredders (28.4%), collector-gatherers (16.8%), and predators (10%). Filterers accounted for only the 4.1% of total invertebrate number, and were almost exclusively represented by Diptera Simuliidae. The amount of CPOM was positively and significantly correlated with taxonomic richness (S: Spearman R=0.67, P<0.05), macroinvertebrate abundance (N: Spearman R=0.69, P<0.05), and shredder density (Spearman R=0.76; P<0.05). However, no significant correlations were found between chlorophyll a concentrations and the same parameters (S: Spearman R= -0.14; P>0.05; N: Spearman R=0.22; P>0.05; Sh: Spearman R=0.25; P>0.05), although we noticed a tendency of scrapers to increase with increased chlorophyll a availability, even if not statistically significant (Sc: Spearman R=0.31; P>0.05).
At the study site, the Po River was characterized by elevated variability in discharge, streambed width, and temperature during the study period. Mean chlorophyll a values were in the range of what reported in other studies regarding mountain lotic systems (Hauer et al., 2007), but slightly lower compared to values reported by Lods-Crozet et al. (2001) for a Swiss glacial stream at the same altitude. Seasonal patterns of chlorophyll a presented a time-lag compared to what reported for glacial-fed streams, where the lowest values of chlorophyll a are in the summer (July and August) when, despite elevated light availability, high temperatures favour ice melt which in turn increases fine particulate transport, shear stress, abrasion and turbidity (Uehlinger et al., 1998; Rott et al., 2006; Uehlinger et al., 2010, Bona et al., 2011). At our study site, chlorophyll a concentrations reached the lowest values in June and peaked in August. Because of the absence of glaciers, snowmelt and therefore discharge increase resulted more anticipated and concentrated than in kryal environments (Rott et al., 2006). A second negative peak occurred in the winter, when, due to reduced daylength and geographic position of the site, light availability was at its minimum. The coarse particulate organic matter availability and its variation in catchments lacking forest canopies are poorly studied compared to those of other regions (e.g., forested streams - see review in Tank et al., 2010). In deciduous forested temperate catchments, CPOM availability is highly seasonal with the highest amounts recorded during autumn and winter (Benfield, 1997; Fenoglio et al., 2005).
In our study site, CPOM amount was generally lower than in lowland, forested streams of the same area (Fenoglio et al., 2005), and showed high seasonal variability, with highest values in the spring. This pattern contrasted what observed in lowland streams, where the peak in abundance of CPOM was recorded in autumn (Allan and Castillo, 2007). In our study only larch litter, a small percentage of total CPOM, had an autumn peak, because larch is a deciduous conifer. We hypothesized that the observed CPOM trend was related to climatic and environmental factors: above the tree line, terrestrial vegetation is constituted mainly by alpine prairie and scattered small shrubs; in the cold season, decaying organic matter is trapped under the snow cover. During the spring thaw, meltwater collects and transports grass fragments and other coarse organic particles from throughout the catchment to the stream.
Benthic macroinvertebrate communities were dominated by orophilous, specialized taxa, with high diversity and abundance of Plecoptera, Ephemeroptera, and cold stenothermic Diptera. Scrapers were the most abundant Functional Feeding Group, suggesting that primary autochthonous production represented a major energetic input in this environment. Filterers are relatively poorly represented probably because, according to the general statements of the RCC (Vannote et al., 1980), fine particulate organic matter is expected to be scarce in mountain headwaters. Abundance and functional composition of macroinvertebrate communities were in the range of what observed in similar Alpine environments (Fenoglio et al., 2007; Lencioni et al., 2011). However, it is difficult to find direct and clear relationships between the variation of observed parameters (chlorophyll a and CPOM availability) and characteristics of macrobenthic communities, as the latters are influenced by a larger number of factors (Allan and Castillo, 2007). However, the lack of significant relationships between algal availability and macroinvertebrate community descriptors, in particular grazers abundance, could be due to a top-down regulation mechanism where scrapers reduced significantly the amounts of benthic chlorophyll a (Brown et al., 2000; Hillebrand, 2002). On the contrary, detritivorous macroinvertebrates are part of a typical donor type system and show an evident bottom-up regulation; i.e., the availability of the resource limits the abundance of consumers (Dobson and Hildrew, 1992).
This study reports one of the first descriptions of the seasonal variations of allochthonous and autochthonous energy inputs in a non-glacial, Alpine stream, with some notes about their relationships with macrobenthic communities. In this context, the most important factors are likely related to climate, especially snow accumulation and meltdown, which consequently promotes substantial discharge variation, abrasion of the substrate and collecting of huge amounts of terrestrial organic matter from the catchment. These factors influence both the autochthonous production and the amount of allochthonous organic input, which has a different trend to what usually happens in lower altitude, forested river systems. The macrobenthic community that characterized the Po River at the study site consisted of stenoecious, specialized taxa that are able to exploit the limited resources of this harsh environment. A better understanding of ecological dynamics and the effects of climate and hydrology on the biota of Alpine rivers is essential for management and planning, in an era when Alpine lotic systems are facing increasing direct anthropogenic impacts and the evolving meteorological patterns associated with the undergoing global climate change.