Macroinvertebrate assemblages from a stream-wetland complex: a case study with implications for assessing restored hydrologic functions

Abstract

Legacies of past land use persist today in the form of incised, single-threaded stream channels with dramatically different hydrologic functions of pre-colonial stream valleys. Restoration practices that aim to return lost hydrologic functions by re-establishing floodplain and groundwater connections should result in stream habitat and biological assemblages that differ from modern, single-threaded channels. The aim of this case study was to identify attributes of macroinvertebrate assemblages that might serve as biological indicators of improved hydrologic functions following the restoration of a stream-wetland complex, similar to a Stage 0 restoration, of a headwater valley in the Western Allegheny region of the USA. We monitored hydrologic functions and macroinvertebrate assemblages from stream reaches of a restored and unrestored site over multiple years during the early years following restoration. Reduced bed mobility and increased flow duration indicated improved hydrologic functions from the restored site. Aggregate metrics that capture functional attributes of macroinvertebrate assemblage (i.e., density and biomass) were consistently greater from the restored site. EPT biomass from restored pools was 3-4 × greater than amounts from the unrestored site as a result of consistently greater mayfly abundance. Restored pools also supported a subassemblage of taxa with life history attributes that are aligned with habitat conditions created from improved hydrologic functions. Results from this case study demonstrate the importance of habitat-specific sampling designs that report the absolute abundance of potential biological indicators. Findings from this case study should help guide the development of rapid biological indicators of improved hydrologic functions.

Keywords: Biological indicators; Hydrologic functions; Post-restoration monitoring; Stage 0 stream restoration.

Fig. 1

Location of study sites within the Licking River Drainage, KY, USA

Fig. 2

Photographic evidence of habitat conditions at the restored site relative to single-threaded headwater reaches in the Daniel Boone National Forest, USA: a restored site before restoration, winter; b restored site post-restoration, late summer; c unrestored site, winter; d unrestored site, late summer

Fig. 3

Hydrologic monitoring results from the a unrestored site and b restored site during the 1st through 6th years post-restoration

Fig. 4

Mean (± SE) invertebrate a, d richness; b, e density; and c, f biomass between riffles (n = 5) and pools (n = 5) of the restored and unrestored sites over the multi-year monitoring period

Fig. 5

Mean (± SE) density of the insect orders a, d Ephemeroptera; b, e Plecoptera; and c, f Trichoptera between riffles (n = 5) and pools (n = 5) of the restored and unrestored sites over the multi-year monitoring period

Fig. 6

Mean (± SE) biomass of the insect orders a, d Ephemeroptera; b, e Plecoptera; and c, f Trichoptera between riffles (n = 5) and pools (n = 5) of the restored and unrestored sites over the multi-year monitoring period

Fig. 7

Nonparametric multidimensional scaling ordination analysis of macroinvertebrate abundance from restored riffles (black triangle), unrestored riffles (white triangle), restored pools (black circle), and unrestored pools (white circle). Labels indicate season (SU, summer; F, fall; W, winter; SP, spring) and year (2013, 13; 2014, 14; 2015, 15; 2016, 16). Stress = 10.5. MRPP results for habitat comparisons within and between sites: restored riffles vs. restored pools A = 0.15, p < 0.0001, unrestored riffles vs. unrestored pools A = 0.04, p = 0.0207; restored riffles vs. unrestored riffles A = 0.07, p = 0.0005, restored pools vs. unrestored pools A = 0.15, p = 0.0001