Árni Einarsson

The ecology of Lake Myvatn and the River Laxá: Variation in space and time

Ecological features of Lake Myvatn and the outflowing River Laxá show a wide range of spatial and temporal variations. The physical division of the lake into three main basins and the variation in chemical composition and temperature of the artesian springs feeding this shallow productive lake have large spatial effects. Variation in groundwater characteristics depends on percolation time and proximity to geothermal sources. Variation in precipitation is evened out by the porous volcanic soil and bedrock and the spring-water discharge is therefore very stable. A pulse of volcanic activity in 1975–1984 (the Krafla Fires) heated the groundwater entering the North Basin of the lake and changed its chemistry. Although much reduced, these effects have not disappeared yet, but overall the impact of the volcanic activity on the biota seemed minimal. Recycling of nutrients through internal loading is important and occurs on various time scales. In winter, when the lake is ice-covered, the topmost 5-cm layer of sediment pore water has a hundredfold concentration of nutrients relative to the overlying lake water. The nutrients are released during the ice-free period by sediment resuspension, diffusion, bioturbation and recycling. In spring, resuspension events sometimes lead to spikes in dissolved phosphorus and nitrogen, but there is little evidence of any major desorption of nutrients from suspended particles during such events later in the summer. In contrast to the stable groundwater, the biota show more or less regular fluctuations with no straightforward correlation with external signals. The most prominent fluctuations, those of the chironomid Tanytarsus gracilentus seem to be driven by interactions between the species and its sediment resources. Fluctuations in other invertebrates could be a consequence of the Tanytarsus cycles due to the large impact this species has on the benthic environment of this detritus-driven ecosystem. Temporal variation in epibenthic chironomids and Cladocera translates into variable production of vertebrate predators (Arctic charr, Salvelinus alpinus, and ducks), body condition and mortality of fish and sometimes into return rates of migrating adult ducks. The waterfowl show large temporal variation on a centennial scale, e.g., the invasion of the tufted duck (Aythya fuligula) which arrived by the end of the 19th century and has by now outnumbered other species. Fluctuations of Cyanobacteria (Anabaena) and the fish Gasterosteus aculeatus (three-spined stickleback) harmonize with the cycles in the benthic community. Palaeolimnological studies indicate that primary production in the South Basin became increasingly benthic as the lake depth was reduced by sedimentation (around 2 mm year−1). Other trends include a decrease in Tanytarsus and Daphnia and an exponential increase in green algae (Cladophorales, Pediastrum) and associated organisms.

High-amplitude fluctuations and alternative dynamical states of midges in Lake Myvatn.

Complex dynamics are often shown by simple ecological models and have been clearly demonstrated in laboratory and natural systems. Yet many classes of theoretically possible dynamics are still poorly documented in nature. Here we study long-term time-series data of a midge, Tanytarsus gracilentus (Diptera: Chironomidae), in Lake Myvatn, Iceland. The midge undergoes density fluctuations of almost six orders of magnitude. Rather than regular cycles, however, these fluctuations have irregular periods of 4–7 years, indicating complex dynamics. We fit three consumer–resource models capable of qualitatively distinct dynamics to the data. Of these, the best-fitting model shows alternative dynamical states in the absence of environmental variability; depending on the initial midge densities, the model shows either fluctuations around a fixed point or high-amplitude cycles. This explains the observed complex population dynamics: high-amplitude but irregular fluctuations occur because stochastic variability causes the dynamics to switch between domains of attraction to the alternative states. In the model, the amplitude of fluctuations depends strongly on minute resource subsidies into the midge habitat. These resource subsidies may be sensitive to human-caused changes in the hydrology of the lake, with human impacts such as dredging leading to higher-amplitude fluctuations. Tanytarsus gracilentus is a key component of the Myvatn ecosystem, representing two-thirds of the secondary productivity of the lake and providing vital food resources to fish and to breeding bird populations. Therefore the high-amplitude, irregular fluctuations in midge densities generated by alternative dynamical states dominate much of the ecology of the lake.

Responses of breeding duck populations to changes in food supply

We describe the main results of a monitoring study at Lake Myvatn, northern Iceland, begun in 1975. The aims were to find factors that limit production of young and cause changes in density of breeding ducks of several species. We estimated numbers of ducks in spring, before nesting, numbers of ducklings produced, and numbers moulting. Chironomid and simuliid dipterans were monitored with window traps. In all duck species studied, production of young was correlated with food abundance. Reproductive performance determined subsequent changes in spring population density of Eurasian Wigeon, Tufted Duck, Greater Scaup, Common Scoter and Harlequin Duck. The spring population of Barrows Goldeneye apparently did not respond to variation in reproductive success. Moulting numbers of male Tufted Duck were related to chironomid abundance, but not those of Scaup and Barrows Goldeneye. Moulting numbers were not associated with previous reproductive output.

Temporal and Spatial Variations in Freshwater 14C Reservoir Effects: Lake Mývatn, Northern Iceland

Lake Mývatn is an interior highland lake in northern Iceland that forms a unique ecosystem of international scientific importance and is surrounded by a landscape rich in archaeological and paleoenvironmental sites. A significant freshwater reservoir effect (FRE) has been identified in carbon from the lake at some Viking (about AD 870?1000) archaeological sites in the wider region (Mývatnssveit). Previous accelerator mass spectrometry (AMS) measurements indicated this FRE was about 1500-1900 14C yr. Here, we present the results of a study using stable isotope and 14C measurements to quantify the Mývatn FRE for both the Viking and modern periods. This work has identified a temporally variable FRE that is greatly in excess of previous assessments. New, paired samples of contemporaneous bone from terrestrial herbivores and omnivores (including humans) from Viking sites demonstrate at least some omnivore diets incorporated sufficient freshwater resources to result in a herbivore-omnivore age offset of up to 400 14C yr. Modern samples of benthic detritus, aquatic plants, zooplankton, invertebrates, and freshwater fish indicate an FRE in excess of 5000 14C yr in some species. Likely geothermal mechanisms for this large FRE are discussed, along with implications for both chronological reconstruction and integrated investigation of stable and radioactive isotopes.

Reservoirs and radiocarbon: C-14 dating problems in Myvatnssveit, northern Iceland

This paper examines 2 potential sources of the radiocarbon offset between human and terrestrial mammal (horse) bones recovered from Norse (~AD 8701000) pagan graves in Mvatnssveit, north Iceland. These are the marine and freshwater 14C reservoir effects that may be incorporated into human bones from dietary sources. The size of the marine 14C reservoir effect (MRE) during the Norse period was investigated by measurement of multiple paired samples (terrestrial mammal and marine mollusk shell) at 2 archaeological sites in Mvatnssveit and 1 site on the north Icelandic coast. These produced 3 new Δ R values for the north coast of Iceland, indicating a Δ R of 106 10 14C yr at AD 868985, and of 144 28 14C yr at AD 12801400. These values are statistically comparable and give an overall weighted mean Δ R of 111 10 14C yr. The freshwater reservoir effect was similarly quantified using freshwater fish bones from a site in Mvatnssveit. These show an offset of between 1285 and 1830 14C yr, where the fish are depleted in 14C relative to the terrestrial mammals. This is attributed to the input of geothermally derived CO2 into the groundwater and subsequently into Lake Mvatn. We conclude the following: i) some of the Norse inhabitants of Mvatnssveit incorporated non-terrestrial resources into their diet that may be identified from the stable isotope composition of their bone collagen; ii) the MRE off the north Icelandic coast during the Norse period fits a spatial gradient of wider North Atlantic MRE values with increasing values to the northwest; and iii) it is important to consider the effect that geothermal activity could have on the 14C activity of samples influenced by groundwater at Icelandic archaeological sites.

Numbers and production of Eurasian wigeon in relation to conditions in a breeding area, Lake Myvatn, Iceland

1. A migratory breeding population of Eurasian wigeon Anas penelope L., was studied over the period 1975-95 on 54 km 2 of freshwater habitat at Lake Myvatn, northern Iceland. Spring numbers of males averaged 1076 (range 526-1887). The sex ratio was 0.97 females per male and the proportion of yearling males averaged 0.21. The mean brood was 3.93 and a mean of 2.83 young per female (range 0.09-5.14) was produced. Production of young was positively related to chironomid abundance and negatively to cold and wet weather. 2. Both sexes seem to be faithful to their winter range, but males are much less faithful to the summer range than are females. The apparent return rate of males is presumably an expression of the return rate of females, which had previously bred locally or been produced at Myvatn. 3. The apparent return rate of old males (2 years or older), mean 0.84 (range 0.58-l.22), was positively correlated with chironomid abundance in the previous summer, which varied greatly. It is suggested that abundant chironomids in late summer may provide a source of highly nutritious food which would increase the probability of wigeon returning to Myvatn or favour the survival of Myvatn wigeon on winter grounds shared with birds originating from other breeding areas. Chironomid numbers showed serial correlation, hence it may be advantageous for the birds to assess the food supply before departure and return accordingly. 4. The apparent return rate of yearling males, mean 0.22 (range 0.06-0.56), was negatively related to the number of young produced in the year before. 5. The density of breeding wigeon at Lake Myvatn was mainly determined by food conditions in the breeding area, as experienced by the adults 1 year previously, and events in the winter range appeared to have less influence. Flyway populations may be limited by a complex set of local and global fluctuations in resources in (i) breeding localities, determining density and production in each, coupled with (ii) staging and wintering localities in each of which local resources would determine density.

An Icelandic freshwater radiocarbon reservoir effect: Implications for lacustrine 14C chronologies

A freshwater radiocarbon (14C) reservoir effect (FRE) is a 14C age offset between the atmospheric and freshwater carbon reservoirs. FREs can be on the order of 10 000 14C yr in extreme examples and are a crucial consideration for 14C dating of palaeoenvironmental and archaeological samples. Correction for a FRE may be possible, provided the FRE and the proportion of FRE-affected carbon within a sample can be accurately quantified. However, although such correction is desirable for affected samples, it is essential that such correction is accurate in order to produce useful chronological information. Accuracy of FRE correction can be limited by spatial variation in FRE within a freshwater system, but despite this there is currently a paucity of information to identify and quantify such variability within affected systems. Here we present results of a study that investigates the effects of spatial FRE variation upon dating accuracy within the freshwater system of Lake Mývatn, northern Iceland. A substantial FRE (>10 000 14C yr) has previously been identified in archaeological and modern samples from the region, which shows the potential for considerable spatial variability. The study also assesses the use of δ13C and δ15N in age correction of affected samples. The results show that benthic detritus and organisms at primary trophic levels from locations within the lake are affected by a FRE of at least 3500 14C yr, with clear spatial variation resulting in 14C age differences of up to 7670 14C yr between samples. There is a broad correlation between stable isotope values and FRE within the data set. However, large associated uncertainties currently preclude highly accurate and precise stable isotope-based quantification of the proportion of FRE-affected carbon within archaeological and palaeoenvironmental samples from Mývatn and the surrounding region.

Long-term changes in benthic Cladocera populations in Lake Myvatn, Iceland

Benthic Cladocera were monitored at five sites in Lake Myvatn, Iceland, over a decade (1990–1999), as part of a programme documenting the population fluctuations of animals at different trophic levels in the lake. The species composition remained relatively stable over the first seven years, but in 1997 the population of Eurycercus lamellatus was greatly reduced at all sites. The following year saw a mass occurrence of Alona rectangula and Alonella nana that were previously abundant only locally and rare at most sites. Alona affinis, A. quadrangularis, Acroperus harpae and Chydorus sphaericus were not affected. In 1999 the Cladocera assemblages returned to the pre-1997 situation. The shift was from large-bodied epibenthic and planktonic species (Eurycercus, Daphnia) to small infaunal (Alona rectangula) and ubiquitous (Alonella nana) species. Medium sized (Alona, Acroperus) and some small cladocerans (Chydorus) were not affected. The course of events was reminiscent of a trophic cascade caused by a change in size-selective predation pressure. If so, the impact of a predatory fish population (three-spined stickleback, Gasterosteus aculeatus) depended on whether cyclic chironomid populations were in a high or a low phase. The change in the Cladocera coincided with profound changes in the sediment characteristics associated with low chironomid abundance. We suggest that the relative competitive ability of the Cladocera species is reversed when the sediment has become homogeneous and nutrient-poor after overexploitation by the dominant, tube building and detritivorous chironomid Tanytarsus gracilentus.

Populations of ducks and trout of the River LaxÁ, Iceland, in relation to variation in food resources

We examined annual variation in production, recruitment and density of the three most abundant vertebrate species of the River Laxa at Lake Mývatn, Iceland: Barrow’s goldeneye, Bucephala islandica, harlequin duck, Histrionicus histrionicus, and brown trout, Salmo trutta, in relation to food resources and other environmental variables. The study is largely based on correlations from long-term monitoring series in the period 1975–2002. Production of young in the harlequin duck was significantly correlated with food resources (the blackfly, Simulium vittatum) of the river, as was the recruitment of brown trout to the angling stock. In Barrow’s goldeneye, which uses both the lake and the river, dispersion of adults in spring and young in August was influenced by the availability of aquatic insects in each habitat. The dispersion of Barrow’s goldeneye tracks the availability of aquatic insects in each of these two main habitats. Introduced Amercian mink, Mustela vison, may have affected spring numbers and dispersion of harlequin ducks, but the evidence was not conclusive. Numbers of both duck species and the trout (as CPUE) were relatively stable, although a sharp drop in numbers followed by slow recovery was observed in Barrow’s goldeneye, and an increase was observed in harlequin ducks in the first year of study.

Islands of change vs. islands of disaster: Managing pigs and birds in the Anthropocene of the North Atlantic

The offshore islands of the North Atlantic were among some of the last settled places on earth, with humans reaching the Faroes and Iceland in the late Iron Age and Viking period. While older accounts emphasizing deforestation and soil erosion have presented this story of island colonization as yet another social–ecological disaster, recent archaeological and paleoenvironmental research combined with environmental history, environmental humanities, and bioscience is providing a more complex understanding of long-term human ecodynamics in these northern islands. An ongoing interdisciplinary investigation of the management of domestic pigs and wild bird populations in Faroes and Iceland is presented as an example of sustained resource management using local and traditional knowledge to create structures for successful wild fowl management on the millennial scale.