Ecological Archives C006-086-A1

Barbara L. Peckarsky, Angus R. McIntosh, Maruxa Àlvarez, and Jennifer M. Moslemi. 2015. Disturbance legacies and nutrient limitation influence interactions between grazers and algae in high elevation streams. Ecosphere 6:241. http://dx.doi.org/10.1890/es15-00236.1

Appendix A. Detailed description of streams where surveys and experiments were conducted.

Stream site selection

Twenty 50 m sampling sites (stream reaches) were selected in the upper East River drainage basin north of RMBL to provide a wide range of conditions for comparisons of stream size, flow-related disturbance, predation regimes, invertebrates and aquatic plant assemblages. Sampling sites were separated spatially to ensure they could be considered independent replicates (see map in Peckarsky et al. 2014).

Biological variables surveyed in 20 streams

During July and August 2006 we compared the composition of benthic invertebrates, aquatic plant assemblages, and predatory fish densities among streams with different disturbance regimes, and tested for associations among trophic levels. We estimated the relative proportions of diatoms, moss, and filamentous forms of algae at all 20 sites. We took five randomly located quadrat samples per site to estimate the percent of the stream bottom covered with diatoms, filamentous forms of algae, and moss when benthic algal biomass was at its peak (Table A1). We also disturbed the substrate inside a bottomless plastic bucket (diameter 30 cm) and collected organic matter in 100 mL cups to estimate the quantity of benthic organic matter (BOM) ash free dry mass (AFDM) and chlorophyll (chl a) to compare the standing crop of basal resources used by primary consumers in all streams (Table A3). Blanks in the table indicate streams with no data. The 10 streams that were used for the mesocosm experiment in 2007 are indicated in bold font, and the five streams used for in-stream channel experiment in 2010 are indicated in bold caps (Table A3).

We estimated macroinvertebrate density and species composition of grazers, shredders, and predators (the predominant functional feeding groups) using a collapsible modified box sampler with a stainless steel frame that covers an area of substrate = 0.104 m², and can be fitted with a standard Wilco ® drift net with mesh size 0.364 μm. This sampler was invented by David Hoffman at the University of Wisconsin. Plans are available from B. L. Peckarsky ([email protected]). We took five benthic samples totaling 0.52 m² of stream bed one time during July 2006 to estimate density and species composition of each functional group, the density of Chironomidae and Oligochaeta, which could not be identified beyond family and were thereby not classified into functional feeding groups, and total invertebrate density (Table A2).

We also estimated the density of predatory fish (mostly brook trout) using a backpacking electroshocker in all 20 streams in August 2006. Thirteen streams were completely fishless (Fig. 1A), and the other seven had a range of densities of brook trout. We have presented the mean density of brook trout in those seven streams based on annual fish surveys that have been conducted since 1996 (Table A3).

Multivariate axis of disturbance

We directly characterized variation among flow regimes by continuously monitoring all 20 sites using Trutrack stage height data loggers from 2004–2009 (Peckarsky et al. 2014). We used stage data from loggers, calibrated by manual estimates of discharge over as wide a range of stages as possible, to compare the hydrologic regimes among sites and combined metrics related to the hydrologic regime with empirical estimates of substrate instability and a qualitative assessment of stream bed stability to classify all 20 sites according to variation in hydrology and geomorphology (low numbers are stable and high numbers are most disturbed) using a principal components analysis (PC axis 1; Table A3) (Peckarsky et al. 2014).

Other abiotic variables

In July 2006 we also measured other abiotic variables in all 20 streams (Table A3) that could explain not only variation in disturbance (see Peckarsky et al. 2014 for additional variables measured), but also variation in the abundance of algae and invertebrate grazers. Drainage basin areas were estimated from digital elevation maps using Arc-GIS software. Maximum discharge (Qmax) in each stream over the 5-year period of the study was obtained from calibrated estimates from stage-height data loggers. We also collected water samples during two dates during July 2006 to estimate soluble reactive phosphorus-P (SRP), ammonium-N (NH₄), and nitrate-N (NO₃) following standard protocols described in the methods. Mean values of those two dates are displayed in Table A3.

Biological variables in experimental streams (2007 and 2010)

Before and at the conclusion of the mesocosm experiment (12–13 July and 24 July 2007) we collected four rocks from each stream used for the experiment and estimated the biomass of chl-a using protocols described in the Methods section. Values represent the means of the two sampling dates (Table A4). We estimated the densities of macroinvertebrate grazers using the collapsible modified box sampler. We took five benthic samples totaling 0.52 m² of stream bed in July 2007 at the end of the mesocosm experiment, and five benthic samples at the end of the stream channel experiment in July 2010. Densities of grazing invertebrates collected in 2007 and 2010 are displayed in Figs. 1B, 2B, and 3B.

Abiotic variables in experimental streams

In each year when experiments were conducted (2007 and 2010) we measured abiotic variables that could be related to the accrual of algae and the strength of grazer-algal interactions, which were the response variables for the experiments. Disturbance regimes, drainage basin areas and maximum discharge in each stream over the 5-year period of the study are displayed in Table A3 (see stream names in bold font). Water temperatures were measured year-round using the Trutrack stage-height data loggers and Hobo data loggers during the summer months. We took one-time measurements of light levels (photosynthetically active radiation - PAR) during the mesocosm experiment in July 2007. We also collected water samples during the mesocosm experiment in 2007 to estimate soluble reactive phosphorus-P (SRP), ammonium-N (NH₄), and nitrate-N (NO₃) following standard protocols described in the methods. Results from 2006 experiments deploying nutrient diffusing substrates (NDS) were used to calculate an index of phosphorus limitation in the 10 streams used for the mesocosm experiment (Moslemi 2010). The P limitation response index represents the effect size of increases of algal biomass when P was added to the NDS (higher numbers indicate greater P limitation).

Conditions in the mesocosms

We used the densities and species composition of grazing mayflies and caddisflies sampled during 2006 (Fig. 1A) to estimate the relative proportions of each species of invertebrate grazer to stock in mesocosms to establish assemblages that reflected the legacy of disturbance in each stream. Analysis of the 2007 invertebrate samples (Fig. 1B) confirmed that those assemblages remained consistent from year to year. At the end of the experiment we counted and weighed the grazers recovered from each mesocosm to estimate final densities and biomasses of invertebrates grazing algae under ambient and enriched nutrient conditions (Tables A5 and A6), and used those values to calculate the grazer-algal interaction strengths.

Table A1. Proportions of aquatic plants in quadrats and densities of invertebrates (number /m²) in benthic samples from 20 streams surveyed in July 2006, sorted from least to most disturbed.

Table A2. Dominant taxa in each functional feeding group of invertebrates in 20 streams surveyed in July 2006, sorted from least to most disturbed.

Table A3. Conditions in the 20 streams surveyed in 2006, sorted from least to most disturbed (PC Axis 1; Peckarsky et al. 2014). Abbreviations are described in the text of Appendix A.

Table A4. Conditions of 10 streams used for the mesocosm experiment in 2007, sorted from least to most disturbed. Abbreviations are described in the text of Appendix A.

Table A5. Conditions in mesocosms with ambient (limited) nutrients adjacent to 10 streams sorted from least to most disturbed.

Table A6. Conditions in mesocosms with enriched nutrients adjacent to 10 streams sorted from least to most disturbed.

Literature Cited

Peckarsky, B. L, A. R. McIntosh, S. C. Horn, A. Wilcox, K. McHugh, D. Booker, W. S. Brown, and M. Alvarez. 2014. Characterizing disturbance regimes of mountain streams. Freshwater Science 33:716–730.

[Back to C006-086]