Temperature response of litter decomposition in streams of eastern Canada depends upon the thermal tolerance of a leaf-shredding insect

Barry R. Taylor, Erin E. MacDonald and Irene Andrushchenko

Poster presented at the Annual General Assembly of the European Geosciences Union, Vienna, Austria, 21-27 April 2007.

Most research was carried out in two sections of a river system in Nova Scotia, Canada. The upstream site was Polson’s Brook, a cool, fast-flowing, woodland stream, heavily shaded by bank-side vegetation. The downstream site was South River, a warm, wide (30 m) unshaded river. Decomposition rates of leaf litter were measured throughout two summers and in autumn.

Although it was expected that the higher temperature downstream (Figure 3) would result in a faster decomposition rate, the results showed the opposite. For both red maple (Figure 1) and speckled alder leaf litter, (Figure 2), the cooler upstream reach had a significantly faster decomposition rate than downstream. This difference may be attributed to the stonefly Leuctra, a leaf-feeding insect restricted to cool, shaded streams. Although Leuctra is known to be an efficient shredder of leaf litter, its role in leaf decomposition may be more significant than previously thought. Other leaf-shredding species are either less numerous or less effective than Leuctra.

The following summer, the experiment was repeated and the same results were obtained. Polson’s Brook supported a faster decomposition rate than South River (Figures 5, 6), despite the lower temperature in the brook (Figure 4).

Fig 5 Mass remaining ± SD of maple leaves after 2 months in Polson's Brook and South River, summer 2006.

Fig 6. Mass remaining ± SD of alder leaves after 2 months in Polson's Brook and South River, summer 2006.

Figure above right shows the most recent version of the model, based on data from sites with and without Leuctra, irrespective of temperature.

The proposed model for decomposition (Figure 7) accounts for this difference by showing an increasing rate of decomposition within the cool temperature range of Leuctra. Above 14°C, Leuctra populations decline quickly, leading to an abrupt decline in decomposition rate between 15°C and 20°C. As temperature increases further, microbial metabolism again accelerates, resulting in an second rise in decomposition rate. Polson’s Brook falls on the peak of this graph, while South River is in the dip. Therefore, decomposition in Polson’s Brook is faster than in South River due to the abundance of Leuctra upstream.

The model is consistent with counts of Leuctra on litterbags retrieved from the field. On both red maple (Figure 8) and speckled alder leaves (Figure 9) the number of Leuctra in each bag was greater in Polson's brook after only seven days in the water, and remained greater for the rest of the experiment. In fact, the number of Leuctra in litterbags from South River was usually zero. Significantly, the cool-water, upstream Plecopteran shredder Leuctra is not replaced by a warm-water shredder downstream. Shredders expected in other warm rivers in temperate regions (Tipulid fly larvae, large Limniphilid caddisflies, crustaceans such as Gammarus) are absent in rivers of northern Nova Scotia.

This model was further confirmed by field and laboratory experiments, where shredders were removed from the environment, while keeping the temperature and the current of both systems constant. Data in Figure 11 are from a recirculating system in which decomposition was very slow without shredders. Figure 12 shows the effect of double-mesh bags, which enclose the leaf litter in a fine-mesh bag (0.25 mm) that excludes Leuctra and other shredders but allows fungi to colonize.

As shown in Figures 11 and 12, under these conditions the warmer South River environment showed a higher decomposition rate than Polson’s Brook. A parallel field experiment and data from other river systems (West River, St. Mary's River) agree with these results. The presence of Leuctra (Figure 10) is therefore instrumental in the efficient decomposition of leaf litter in cool streams, where microbial metabolism is otherwise limited by low temperature.

Figure 10. Late-instar larva of Leuctra sp. Note the relatively small size of this influential organism. Photo by Erin MacDonald

Fig 11. Mass remaining and SD in Maple and Alder leaf packs in laboratory environments mimicking Polson's Brook and South River.

Fig 12. Comparison of mass remaining and SD between single and double meshed bags of alder in Polson's Brook and South River, summer 2005.

This model has clear implications for global warming, because the mean temperature of shaded, groundwater-fed streams may be expected to increase in the near future. In South River, that increase may stimulate microbial metabolism, resulting in an increased rate of decomposition. In Polson’s Brook, however, a temperature increase would be detrimental to Leuctra (optimal temperature ~14oC), and thus depress the rate of decomposition. Slower leaf decomposition in such small streams will reduce the energy input and dramatically reduce their capacity to sustain healthy populations of organisms.

Acknowledgements

Thanks to Jennifer Cairns and Meghan Hines for field and laboratory assistance. This research was supported by a grant from the Natural Sciences and Engineering Research Council of Canada (NSERC) to Barry Taylor.