Correction to “Individual and combined impacts of carbon dioxide enrichment, heatwaves, flow velocity variability, and fine sediment deposition on stream invertebrate communities”

Hunn, J. G., Orr, J. A., Kelly, A.-M., Piggott, J. J., & Matthaei, C. D. (2024). Individual and combined impacts of carbon dioxide enrichment, heatwaves, flow velocity variability, and fine sediment deposition on stream invertebrate communities. Global Change Biology, 30, e17336. https://doi.org/10.1111/gcb.17336

In the originally published version of this manuscript, the full names of the authors were omitted. They are

Julia G. Hunn, James A. Orr, Ann-Marie Kelly, Jeremy J. Piggott, Christoph D. Matthaei

This error has been corrected online.

In addition, due to a mistake that occurred during the typesetting process, negative symbols were omitted from some locations in the text. The corrected text is below:

2.2 Experimental setup

Each of the eight 135-L header tanks gravity-fed stream water to 16 mesocosms at a constant discharge of 2 L/min, measured at the start of the colonization period (Day −16) and recalibrated daily, via 4-m length of 13-mm polythene pipe controlled by a tap regulator. To create a bed substratum emulating small New Zealand streams (Matthaei et al., 2006), 500 mL of small- to medium-sized gravel was collected from the river floodplain, sieved to remove fine sediment (particles <2 mm; Zweig & Rabeni, 2001), and added to each mesocosm with 14 randomly selected 40- to 50-mm flat cobbles placed on top. On Day 0, a piece of PVC pipe (80 mm length, diameter 40 mm) was placed in the remaining space to act as a fish shelter, and a 1-cm stainless steel mesh covering was placed over the outflow to prevent fish escaping, scrubbed daily with filtered stream water to remove any trapped organic material.

2.3 Experimental design and procedures

CO₂, fine sediment, flow velocity variability, and temperature were manipulated in a full-factorial 2 × 2 × 2 × 2 design with eight replicates of each treatment combination. Flow to the mesocosms began on October 21, 2019 (Day −17), the start of a 17-day colonization period. During this, the CO₂ (from Day −17) and sediment (from Day −14) manipulations were already implemented. A 35-day “experimental” period (beginning on Day 0) followed, during which temperature and flow velocity were manipulated, as well (Figure 2).

CO₂ treatments were randomly assigned at the header tank level, with one CO₂-enriched header tank in each of four spatial blocks of two tanks per block. CO₂ was bubbled into CO₂-enriched header tanks continuously from the start of the colonization period (Day −17). On Days 14 and 28, 1-L water samples taken from 16 randomly selected channels (eight ambient and eight CO₂-enriched) were stored in sealed glass bottles and preserved with mercuric chloride for DIC analysis. Within 5 min of sampling, pH and temperature were also measured in these channels using a handheld pH meter (HI-98128; Hanna, Rhode Island).

On Day −2, natural invertebrate colonization was supplemented with taxa underrepresented in the drift by adding one standard sample of the Kauru River benthic invertebrate community to each mesocosm, following the methods described in Piggott, Townsend, and Matthaei (2015a).

Flow velocities were measured in all channels on Days −12, −1, 4, and 11 using a precision flow meter (MiniWater20; Schiltknecht, Gossau, Switzerland). Average near-bed velocity was 20 ± 1.1 cm/s on Day −12, with velocity gradually decreasing over time as prolific benthic algal communities (including filamentous taxa) formed in the channels. Mean velocity prior to beginning flow treatments (Day −1) was 17.9 ± 1.4 cm/s. On Day 4 (the first “fast” period), average velocities were 7.0 ± 2.8 cm/s in “constant” channels and 14.9 ± 3.2 cm/s in “variable” channels. On Day 11 (the first “slow” period), average velocities were 9.3 ± 2.6 and 2.0 ± 1.5 in “constant” and “variable” channels, respectively. Based on these two dates, “variable” channels experienced a mean velocity of 8.5 cm/s across the “fast” and “slow” periods. For “constant” channels, the corresponding mean velocity was 8.2 cm/s.

The publisher apologizes for this error.