Litter processing in diffuse and conduit springs

We used bedrock geology and prior water chemistry data to classify and choose 4 conduit and 4 diffuse flow springs in a karst region of southeast Minnesota. Decomposition processes and chemical/physical conditions were compared between spring classes during two seasons (May and September, 1987). Although large storms and conduit run-in events did not occur during our sampling, baseline water chemistry supported our a priori spring classification. Baseline nitrate and atrazine levels were significantly higher in conduit than in diffuse springs. During a minor run-in event, atrazine levels increased significantly to at least 1.2 μg L^-1 in all conduit systems, but remained unchanged in diffuse systems. Decay rates of the two predominant litter sources in the springs (watercress [Nasturtium officinale R.Br.] in May and box elder [Acer negundo L.] in September) were used to assess differences in biological activity between the spring classes. During May, watercress decayed (1 mm coarse mesh bags) at approximately the same rate in conduit and diffuse systems, k = 0.094 and 0.099 respectively. This result was unexpected since shredder colonization was much greater on litter bags in diffuse systems. In September, boxelder decay rate was significantly higher in diffuse (k = 0.018) than in conduit (k = 0.013) springs; and decay rates seemed to reflect significantly higher shredder colonization on bags in diffuse systems. Microbial activity on watercress and boxelder litter, measured as electron transport system (ETS) activity, was seldom significantly different between spring classes. Also, both watercress and boxelder litter decay rates for fine mesh bags (240μm) were similar between conduit and diffuse springs, suggesting that microbial processing did not differ greatly between spring classes. In conduit springs, low shredder colonization was apparently the result of low benthic shredder densities. Shredders and other macroinvertebrates may be adversely affected by discharge fluctuations from conduit run-in events. We suspect that, during times of low flow, watercress standing stock was also an important influence on shredder abundance.