The distances travelled by drifting invertebrates in a Lake District stream

The distances travelled by drifting invertebrates of 18 taxa were investigated at sites 3 and 4 in the Wilfin Beck, a small stony stream in the English Lake District. Fifty invertebrates of the same taxon were released into the stream at increasing distances upstream from a large net which caught all invertebrates drifting downstream. The relationship between the catch in the net (Y _x ) and the distance from the release point to the net (X m) was well described by the regression equation: $$Y_x = Ae^{ - RX} $$ whereR is the constant relative rate of return of invertebrates to the bottom of the stream, andA=Y ₀=number of invertebrates released at each point=50. Values ofR, the mean drift distance ( \(\bar X\) m), and the distance (X _p m) travelled byP% (1%, 10%, 50%) of the drifting invertebrates were calculated for each taxon at different modal water velocities. Values ofR, \(\bar X\) andX _p were fairly constant for each taxon at a particular modal water velocity, and were not significantly affected by the source of the experimental animals (benthos or drift), by changes in illumination (daylight or darkness), or by seasonal changes, including water temperature. The experimental taxa at site 4 were divided into the following three groups according to their ability to return to the bottom: Polycelis felina, Ancylus fluviatilis, Chironomidae, Eliminthidae, andAmphinemura sulcicollis. Values ofR, \(\bar X\) ,X _p not significantly different from those obtained for dead invertebrates, which were removed from the drift by chance effects. Protonemura meyeri, Leuctra spp.,Chloroperla spp.,Rhithrogena semicolorata, Simulium spp. Values ofR, \(\bar X\) ,X _p not significantly different from those obtained for dead invertebrates at modal water velocities ≧19 cm/sec. At low velocities (10–12 cm/sec),R was significantly greater and \(\bar X\) ,X _p significantly smaller than values obtained for dead invertebrates. Erpobdella octoculata, Gammarus pulex, Hydropsyche spp.,Ecdyonurus venosus, Ephemerella ignita, Baëtis rhodani. Values ofR, \(\bar X\) ,X _p significantly different (R greater, \(\bar X\) andX _p smaller) from those obtained for dead invertebrates. The exponential law was not a good model for experiments with cased caddis larvae (Agapetus fuscipes and a mixed group ofSericostoma personatum, Drusus annulatus, Potamophylax cingulatus). Larvae sank rapidly after release and drifted over very short distances. Values ofR for each taxon were significantly higher at site 3 than at site 4, and the more rapid return at site 3 was presumably due to dense stands of aquatic macrophytes. The increase inR was greatest forSimulium spp. andE. ignita. The relationship between \(\bar X\) and modal water velocity (V cm/sec) was well described by the regression equation: $$\bar X = aV^b $$ wherea andb are constants. The relationship betweenR andV, orX _p andV, was described by similar equations. Values ofa andb were calculated for each taxon at sites 3 and 4. The drift distance was also investigated by blocking the total drift and taking drift samples at several stations downstream from the blockage. Drift rate was markedly reduced immediately below the blockage and then gradually increased downstream until it was similar to that recorded before blocking. The mathematical model developed from the detailed experiments was a good fit to the results of the blocking experiments. The behaviour of the drifting invertebrates and their ability to return to the bottom are discussed. From the results of the present study and those of other workers, it is concluded that the mathematical model is a good model for invertebrate drift. The implications of the model are discussed, especially the relationship between drift rate and water velocity, and the proportion of the benthos in the drift.