PREreview of Invasion dynamics of quagga mussels (Dreissena rostriformis bugensis) within a reservoir and its spatially intermittent watershed: Lake Piru and the Santa Clara River, California, USA

This review was performed by a pre-review journal club of ecologists including grad students, post docs, and faculty. We were excited to review this paper because we are interested in ecosystem health and population dynamics. 

Review authors: Grace Leuchtenberger, Kindall Murie, Fiona Boardman, Zachary Bengtsson, Olivia Cattau, Annie Colgan, Chuck Flaherty, Sanford Leake, Rachel Potter, Monica Sheffer, Julia Smith

Summary

This article was a comprehensive report on the state of quagga mussel invasion in Lake Piru, a southern Californian, man-made lake. The authors quantified quagga abundance via dive and photo surveys from 2014-2021, removing quaggas in the process as part of a mitigation plan. The authors also monitored water conditions and water quality, lake fill, veliger (larval mussel) abundance, and settlement and growth rate of quaggas. This data was parlayed into models that examined the impacts of reproductive mussel density and environmental parameters on veliger abundance, and the impacts of veliger abundance on settlement rate. 

The authors presented comprehensive data on mussel biomass, density, settlement, and growth throughout the lake at different sites, as well as downstream from the lake. The first model predicting veliger abundance did not include the amount of reproductive mussels at all, with the simplest model including only lake fill status, temperature, and dissolved oxygen. The second model found that veliger abundance predicted settlement. We appreciate the many years of work that went into collecting very comprehensive data. Long term data sets are rare and we believe that this paper is important for both quagga mussel invasion and mitigation. We find that the manuscript could be improved substantially by making slight alterations to the framing, so that it is question-driven and ecologically informed. To this end, below, we raise some points/concerns and minor corrections that we believe would improve this paper.

Major Concerns

• We feel like this paper reads much like a report and less like a scientific paper and has a lot of results listed without proper framing in the introduction. The introduction includes good background information but is missing a more defined purpose and questions. Lines 52-54: would be a great place to state some of the specific questions addressed later  in the results by your models. Lines 52-54 as of now just give a general statement about the overarching aim of this study. 
• While the comprehensive nature of the paper is a selling point, we believe many of these graphs could be supplementary, while highlighting the graphs that convey your main results. We agreed that Figures 1, 2, 4 ,6 and 9  seem the most valuable. 
• Results from statistical analysis are not present in the paper. There is a table 2 caption stating the results of the mixed effect models, but the table (pdf) is missing. In the caption you list the models are for testing the “1) the relationship between reproductive mussels (>5mm) and veliger abundance and 2) the relationship between veliger abundance and mussel recruitment in Lake Piru. The top 10 or 11 models are shown, shown in order of the best fit model from AIC criteria.'' However, there seems to be many other statistical analyses performed (ANOVA’s) that are throughout the paper and may also benefit from a summary table in the supplemental. 
• Line 560: Because there is only one site (Bridge 40km) and only one year where mussels were detected, we feel like table one is unnecessary. We also think that the formatting of table one needs improving if kept. It's very hard to read and could be made in a more informative way. We suggest maybe envisioning it as a map similar to figure one; with dots for where mussels were detected and not detected each year. 
• After reading and discussing this paper all reviews saw the value in the results of this paper but felt that the authors need to include the following information within the discussion section of the paper: 1) how does this paper come back to improving ecological knowledge? 2) what goals were accomplished (this will be stronger if authors lay out more detailed questions addressed beforehand in the introduction),  and 3) are there any suggestions for future efforts in managing quagga mussel invasion? We believe answering these questions above and summarizing them in a concluding paragraph after lines 499-504 will help to end the paper strong instead of ending on a “troubling finding of this study.” 
• Question: We were wondering if it was possible to do any predictive analysis looking at the future population of quagga mussels in Lake Piru and the Santa Clara River?

Minor Concerns 

Lines 33-34: Mussels are filter feeders and could potentially help to filter out harmful algal blooms and not facilitate algal blooms. More detailed explanation on why they contribute to harmful blooms. 

Lines 35-42: Could be interesting to include how much economic cost is associated with quagga mussel removals and their effect on Southern California Steelhead populations. 

Line 42: Are there any positive effects of quagga mussels on the environment? Do they create habitat for smaller invertebrates or filter out harmful pollutants? 

Line 59: Why are figure references highlighted yellow throughout the paper? 

Line 68-69: Needs a transitioning sentence, lines 69-71 seem out of context or randomly placed. 

Lines 74-76: Physical property measurements like temperature, dissolved oxygen, turbidity, specific conductance and pH were measured, but with what instrumentation and were the sensors calibrated, were the data corrected in any way for drift over time? Specific conductance is not mentioned or shown again in the paper, why include it here? Also why not convert to salinity? Was dissolved oxygen corrected using the specific conductance or was the instrument assuming a salinity of 0? Were any water samples taken to verify sensor readings?  

Line 80 + others: When citing R authors do not need to cite Program R. Instead try: All analyses were done in R (version 3.43) . Citation is as follows:

RCoreTeam.R:ALanguageandEnvironmentforStatisticalComputing.RFoundationforStatisticalComputing,Vienna,Austria.http://www.R-project.org/ (2013). 

You can then list citations for packages below as well. 

Line 81: For Bathymetric data provided by UWCD, need to list the company’s full name before abbreviating when first mentioning them and cite them. 

Line 84: How many quadrats were done at each site? Need justification for why quadrats were chosen over say transect sampling. Were the quadrats randomly placed or targeted?

Line 85: GPS coordinates needed for all sites. 

Line 89: The measurement of mussel length has a lot of error in it considering that the margin of error (5 mm) is the same as the qualifying measurement for a reproductive mussel (>5 mm). The authors need to justify why that margin of error is acceptable if they’re measuring size classes, and also why it’s acceptable for measuring amounts of reproductive mussels. Additionally, the authors need to justify why 5 mm is the cutoff for reproductive adults. Also do not need to state you used calipers- unnecessary detail. 

Line 93: fines habitat? 

Line 95: Weird wording- Try images were corrected for curvature of camera lens in Adobe Photoshop CS6. Needs citation for Adobe. 

Line 106: land-side? Or Landslide? 

Line 155: Not sure you need to include the model in (), hard to read the way it's written. 

Line 169: This is the start to the models and what questions they address, and it feels like it comes out of the blue again (see first major concern). But it would be nice to see these questions addressed in the introduction. 

Line 171: lmer package in in R. 

Line 173: Specific conductivity does not need to be capitalized. 

Line 193: “used the “acf” and “ccf” functions in R to generate autocorrelation and cross-correlation plots, respectively. 

Line 239: Needs a comma? ( Figure 5A; 27,069 ± 6,964 mussels m-2), This is throughout the paper. 

Line 441: no data that shows what happened if they didn’t remove the quaggas

Line 497: Instead of saying veliger towing isn’t helpful, could have justified it a bit more

Graphs: 

All graphs’ text axes need to have a bigger font- hard to read. 

Figure 5: maybe say cumulative biomass removed, because times that biomass was removed were not equally spaced.  

Figure 6: Need to be referenced somewhere in the text, if not it's not needed in the paper, and don’t need panel B if kept. 

Line 554: Figure 10 panel A-resize y axis- some data is cut off.