Get access

Flow variability and longitudinal patterns in parafluvial water chemistry, aquatic invertebrates and microbial activity


Correspondence: Scott T. Larned, National Institute of Water & Atmospheric Research Ltd, P. O. Box 8602, Christchurch, New Zealand.E-mail:


  1. Parafluvial zones occur in emergent gravel bars and other river sediment structures where exchange with surface water is limited to horizontal inflow and outflow. Within parafluvial zones, interstitial water flows down longitudinal flowpaths in response to hydraulic-head gradients. Changes in physical and biogeochemical conditions along these flowpaths lead to the formation of longitudinal gradients in solute concentrations, biota and microbial activity. The influence of river-flow variability on interstitial residence times, flowpath sizes and solute influx suggests that flow variability also affects longitudinal gradients in parafluvial zones.
  2. We studied associations between longitudinal gradients and flow variation in the forms of flow fluctuations and variable water sources. We used data from parafluvial zones in a New Zealand river to test four predictions: (i) longitudinal gradients in solutes, invertebrates and microbial activity are steeper in runoff-dominated losing and neutral reaches compared with groundwater-dominated gaining reaches; (ii) moderate-to-large floods eliminate longitudinal gradients, which reform during recessions; (iii) between floods, gradients in parafluvial chemistry change in steepness and direction in response to changes in river-flow levels; and (iv) between floods, invertebrate abundance and diversity are positively related to river flow.
  3. In Study 1, we measured solute concentrations, invertebrate composition and microbial activity over a range of flow levels in vegetated gravel bars with contrasting water sources (alluvial groundwater in a gaining reach, and hillslope runoff in losing and neutral reaches). In Study 2, we made repeated measurements in unvegetated gravel bars before and after a one-in-2-year flood to assess its effects on longitudinal gradients in nutrients and invertebrate communities.
  4. In Study 1, we detected gradients along parafluvial flowpaths in inline image, dissolved reactive phosphorus and major ion concentrations and in pH, electrical conductivity and invertebrate density. There was little congruence between bars in different reaches, or among variables; some gradients became steeper as flow increased, and others shifted from negative to positive or vice versa or approached zero. Parafluvial chemistry was strongly affected by water source; median concentrations of most solutes were 2–20 times higher in the groundwater-dominated parafluvial zone than in the runoff-dominated parafluvial zones.
  5. Prior to the flood during Study 2, the parafluvial zones were characterised by negative longitudinal gradients in dissolved organic carbon (DOC) and dissolved organic N (DON) and positive gradients in inline image. These gradients were eliminated by the flood, but inline image and DOC gradients redeveloped during the post-flood period. Differences between the steepness and direction of gradients in Study 1 and the pre-flood period in Study 2 may be related to differences in terrestrial vegetation, as well as differences in flow regimes.
  6. Associations between river-flow levels and parafluvial invertebrate communities differed in the two studies, which corresponded to different flow ranges. Invertebrate density and taxon richness were positively correlated with flow over a range of low flows in Study 1, but there was no detectable correlation over a range of high flows in Study 2. These results suggest that beneficial effects of flow fluctuations on invertebrate communities peak at moderate flow levels, beyond which invertebrate–flow relationships become unstable.
Get access to the full text of this article