Dataset: Lake Quagga Mussels

Study Location

We sampled 25 locations in Lakes Michigan (17 sites) and Huron (8 sites) during spring (May 24 – 29) and summer (July 31 – August 7) of 2019 (Figure 1, Table S1). Sites ranged from 20m to 130m deep. To account for differences in water chemistry between lake depth regions, sites were designated as ‘shallow’ (30 – 45m), ‘mid-depth’ (46m – 90m), or ‘deep’ (> 90m). Green Bay, with sites ranging 25 – 35m deep, was also considered a separate lake region due to consistent differences in trophic status compared to other locations.

Site Characterization 

At each site we collected water samples from five meters below the surface (“surface”) and two meters above the sediment-water interface (“near-bottom”). Water samples were collected using 12-L Niskin bottles on a Seabird 32 Carousel Rosette sampler. We filtered water samples through GF/F filters to determine particulate organic matter (POC, PON, and PP) concentrations and 0.45μm cellulose nitrate filters for chlorophyll a measurements. Filters for particulate phosphorus (PP) measurements were rinsed with 15 – 20mL 1% HCl and 50mL MilliQ water. All filters were frozen until analysis. We also collected chlorophyll a fluorescence and temperature profiles at each site using a SeaBird Model 911 plus CTD. We measured average dreissenid biomass (g/m2 ) at our sites using underwater video surveys and Ponar grab sampling (see Zalusky 2021 for more detail on mussel survey methods). 

Mussel Collection 

We used Ponar grabs to collect mussels for excretion experiments and tissue composition analysis. To minimize stress and damage, we carefully separated mussels from sediment by hand. Mussels were also gently cleansed using a soft-bristled toothbrush to ensure removal of adhering biofilm that might impact respiration and dissolved nutrient measurements. To mimic environmental conditions close to those insitu, we stored mussels in unfiltered near-bottom lake water during processing. Mussels were rinsed with ultrapure MilliQ water before we placed them in incubation chambers for excretion and respiration rate measurements. 

Excretion and Respiration Rate Measurements 

Mussel respiration and nutrient (N and P) excretion rates were measured using shipboard incubations. To test for the influence of body size on metabolic rate, at each site we incubated up to 12, 5, and 3 small (< 15.5mm), medium (15.5mm – 25mm), and large (> 25mm) mussels, respectively. Incubations were conducted in acid-cleaned vials with 0.7μm GF/F filtered near-bottom lake water. We used five replicates of each mussel size class (small, medium, large) and incubated three containers without mussels to act as controls (18 vials total). We mostly used 40mL amber glass vials as experimental containers, although a small number of incubations were performed in 60mL falcon tubes when working with very large mussels. Before each incubation began, we determined initial DO concentration and took initial dissolved nutrient samples in duplicate, which we syringe filtered (0.22μm polyethersulfone filter) and froze for later analysis. We then added mussels to the experimental containers, sealed the containers with no headspace for gas exchange, and incubated them for three hours in the dark at 4°C (Ozersky and others 2015; Mosley and Bootsma 2015; Tyner and others 2015; Vanderploeg and others 2017; Zalusky 2021). After three hours, we determined final DO concentrations for each vial and syringe filtered incubation water from each vial (0.22μm polyethersulfone filter) then froze samples for later analysis. We measured DO concentration using a WTW Optical IDS dissolved oxygen sensor (FDO® 925, precision = 0.01mg DO ±1.5%, Wilheim, Germany), paired with a multi-parameter portable meter (MultiLine® Multi 3510 IDS). The DO sensor was rinsed with cold MilliQ water between each measurement. DO concentration did not go below 327μmol O2/L in the incubation vials, indicating no instances of hypoxia. Finally, we froze mussels from all incubations for later analysis. 

Laboratory Methods 

Water Chemistry 

Total seston POC and PON concentrations of water samples were determined using a Vario EL cube elemental analyzer (Elementar, Langenselbold, Germany). We dried sample filters for 24 hours at 60°C and then encased them in tin capsules for analysis. Water PP was then determined using a SEAL Analytical AQ400 discrete analyzer (Mequon, Wisconsin). We combusted PP sample filters for two hours (550°C) and then digested them using sulfuric acid (10N H2SO4). PP samples were then reacted with acidic molybdate and antimony and reduced by ascorbic acid to form phosphomolybdenum blue, which was measured photometrically (880nm) (modification of US EPA 119-A method, Murphy and Riley 1962). Water column chlorophyll a filters were extracted into 10mL of 90% acetone for 24 hours (Welschmeyer, 1994) and analyzed using a UV-Visible spectrophotometer (Turner Designs, 10-AU, Sunnyvale, CA) at an excitation wavelength of 436nm and 680nm emission. We ran these and other nutrient samples (see below) in duplicate. We measured dissolved N and P (as NH3 + -N and SRP) concentrations in incubation water from mussel excretion experiments using a SEAL Analytical AQ400 discrete analyzer (Mequon, Wisconsin) (alkaline phenate method for NH4 + - EPA-103, acidic molybdate method for SRP - EPA-155). Respiration and excretion rates were determined by normalizing final incubation concentrations to control values, mussel biomass (as SFDM, see next section), and volume of incubation chambers. Values were then divided by the length of incubation time, yielding rates as μmol/g SFDM/h. 

Mussel Biomass and Tissue Composition Analysis 

Mussels were removed from their shells, dried for 24 – 48 hours in a drying oven at 60°C, and weighed to determine total biomass as shell-free dry mass (SFDM) for each incubation replicate. We then combined replicates for each size class at each site and analyzed them in duplicate (C and N) or triplicate (P). Dried tissue was homogenized using a mortar and pestle and stored in a desiccator until we weighed and analyzed them. We then weighed samples of homogenized tissue for POC and PON (target weight 0.6mg) and PP (target weight 1mg) analyses and processed them the same way as our seston samples (see above).