Peter Rockermann
This paper will describe the vegetation in five sites on each of five rivers in New York State, with attention to ash because of its susceptibility to the emerald ash borer.
General comments
I gather that the Results section is not complete. I hope you will have figures that illustrate your primary findings.
Use terms consistently to describe your sampling design. You use “river,” “plot,” “stand,” and “site.”
The experimental design needs to be clearly identified. It would be easy to use plots for the experimental unit, but incorrect, because they are not independent samples. Your experimental unit should be the river, right?
You decided not to use common names for trees, but for the emerald ash borer? I appreciate common names. You could look at other papers in your target journal to judge what your audience expects.
Table 1 currently doesn’t provide information of interest to your readers. The individual plot locations should be archived with your major professor or research sponsor, but your readers don’t care. The site designations are not useful to us. It could be useful to have a table that described the characteristics of the sites that are important to interpreting your results.
Specific comments
You don’t have page numbers and the line numbers aren’t unique! I’m going to cite line numbers and let you figure out what page these comments are on (they are probably mostly chronological).
3 This would be a good place to describe the geographic regions and what’s different about them. If it’s not important, maybe you don’t need to mention it. The list of the names of the rivers and the part of the state they are in doesn’t do much for me. If you have predictions of how things would be different in the different rivers based on their locations, this might even be worthy of moving to your Introduction.
6-10 This is a long sentence and any kind of numbering system is usually a sign that your sentence couldn’t hold up under its own weight. In this case, it seems like it might be reasonable to enumerate your criteria, except that I bet it’s not really important that there are four of them. The parts are not parallel in structure. You could make them parallel, and it would still be horrendously long. If you make short sentences, it won’t matter if they differ in structure.
23 Do you need to use an alpha or can you just report your P values?
“dominant” in forests can refer to canopy position. Is “common” better?
You give three significant digits in your results, but I bet the last one is not meaningful. This will be more clear when you have s.e. or CIs.
You give 5 decimal places in your lat and long, which looked like way too much to me, so I looked it up. “A value in decimal degrees to an accuracy of 4 decimal places is accurate to 11.1 meters.” (For longitude, it must depend on the latitude, but close enough.) Your sites are bigger than that. So even 4 decimal places is probably too much. If you want to archive the locations of all your plot corners, fine, but your readers don’t need that.
Comments not worthy of a response to the Editor
It would have been good to include your figure for review (the map).
7 I’m not sure we need to know
10 I bet you mean “plot”. If not, define what a site is.
Actually, I bet “river” is not right in line 3. You didn’t consider the entire length of the river, I bet. Maybe “site” would be the right term for each of the places you chose to put sites, one per river.
10 What’s a “stand” relative to plot, site, and river? Same question.
13 You can’t start a sentence with an abbreviation.
Jim Arrigoni
This paper describes the development of a biological index for monitoring water quality in streams in Hong Kong. Tolerance values are assigned to invetebratetaxa by family. The numbers of invertebrates are
General comments
The paper is long and there are a lot of steps to explain. I was going to suggest that section headings would help. But I think I will make an exception and suggest that you depart from the IMRD format, and use section headings that allow you to write a more naturally organized narrative.
Specific comments
4 I can’t believe you eliminated 55 of the 82 stations and used all that were satisfactory. If so, maybe you should list your criteria. I would believe you if you said you picked 27 and gave the criteria for your selection. I think “inappropriate” sometimes is a way to evade responsibility for defining what’s desirable.
48 An alternative is to assign the detection limit. I think I’ve also seen half the detection limit.
49 The metric you developed. It shouldn’t take me part way through the sentence to know you aren’t talking about someone else’s metric.
The verb tense switches back and forth between past and present in the Results. I suggest past tense throughout (even though some statements may still be true; you don’t know that without going back!)
97 I lost track. Are we still working with the PCA Axis 1 scores? So, you didn’t worry about breaks in the distribution, but your classes 1 through 5 are distinguished along this axis? The narrative is hard to follow because it’s so long.
105 This sounds like a different topic. The first sentence of the paragraph says you calculated the HKMI, but here you are talking about an assessment of the index. I
Table 1 is an alphabetical list that mixes a lot of categories together. Would a good way to organize them be by analytical methods? I object to the three columns because they are not meaningful, and the rows are not meaningful.
Table 2 is important. How are the entries ordered, taxonomically?
Table 3: I couldn’t understand the table heading, there is a “between” followed by two “and”s. I thought maybe the ASPT was being compared to each of the HKMIs, but that’s definitely not what the sentence says. Is it important to include the P values, when all of them are so small?
Table 4 has 8 values, did we say 6 was minimal? Do you think all those digits are significant? CVs are often reported in units of %, which would get you down to 1 digitand be easier to read. Are the differences important? Oh, you wanted to say that the dry season was better than the wet season? They don’t look that different to me.
Figure 1 caption doesn’t say 27 streams, your other captions do.
The Figure 1 variables could be identified in Table 1.
Comments not worthy of a response to the Editor
2 Good example of a problematic noun train. It sounds fine, but I can’t figure out how you should hyphenate it. If “streamwater” were one word, you’d be in business. I bet you can pick some words to drop out. Water quality appears later in the sentence.
This sentence is a mouthful. It won’t matter so much when it’s buried in the Methods; it’s not really your opening sentence.
22 verb tense change
25 comma in a list of two items
30 three visits occurred: passive is strange here, like they just happened to occur. “We scheduled” would show intent.
The following paragraph is awkward without an agent. “after carefully washing” is a dangling modifier because you have no actor in the sentence.
41 suggests that those without large numbers were not sorted completely
Sorting is different from counting? I don’t get it.
Figure 2: How about a dashed line for 1:1? Oh, you need it to be consistent with Figure 3.
Figure 3 is not consistent in format (font size) with Figure 2.
Angela Sirois
This paper describes the population dynamics of a threatened bog turtle in two sites in western Massachusetts. The two sites provide a contrast, as one is actively managed for turtle conservation and the other is not. Specifically, the managed sites have a greater variety of habitat types. I think the point of the study is to see whether managing for these habitat alterations pays off in turtle population growth. Maybe it would be good to pick a different question if this is not one you can answer.
General comments
The study is impressive for the long record of sampling. It seems a shame to exclude data from before 1995 and 1997-2005, if you have it. I’m looking at Figure 1 and wondering whether I would have a completely different impression if all the years were represented (I know this can happen with sampling forest floor mass, for example). Wouldn’t there just be larger or smaller error bars with lesser or greater sampling effort? We could see variation in the mean and also how much store to put by the changes over time.
The Results section has one subsection headed “Demography” and no other subsections. The results seem short (about 1 page) compared to the 5 pages of methods! So I’m guessing that there is more to come. If the focus of the paper is on habitat alternation, then I would expect there to be tables or figures showing me the success of turtles in different habitat types. The other thing your abstract highlighted was the comparison of pre-treatment to post-treatment observations. I didn’t notice these in your results.
The number of tables and figures seems excessive. Here are some suggestions for reducing them.
Tables 7, 10, and 13 present details of lots of models. I think the point was to find what variables best explained the observations. Is there more we need to know than what the best model was? I think I’d get more out of a paragraph of text and it would take less space than these tables. You could tell us what variables were tested and how much better the best model was than some of the other good models. I’m not sure it’s important to show us the details of the statistics for all the models.
I suggest combining Figures 1, 2, and 3 into panels of a single figure sharing an x axis that is quantitative for time. As above, I’d also like to see the data filled in for the dates where they are available, even if the design is not consistent.
Specific comments
38, 30 The first sites is a calcareous wetland, I don’t know why I expected the second site to be different. Could you give the features they have in common before you go on to the specifics?
p. 5 You should give references for these various tests. Not the statistical package, but the source for the method. I think you can reduce the number of acronymns, I bet these are not all recognized by your target audience. Some of them you introduce multiple times and some you don’t use consistently, which is probably a vote for not using them. See p. 6.
105-107 Is it important which way you coded it? I think you would report the same statistics either way. Similarly, we don’t need to know the units for your variables (e.g. 135); the significance of your tests does not depend on the units. We do need to know them when you display results.
100, 113 When you say data were pooled for both sites, in the first case I think you meant separately by site and in the second case you mean together. Make the language more distinct if these in fact mean different things.
123-126 This material looks like it belongs in the section where you catch and measure turtles, not where you analyze your data.
126-128 and 135-138 There is probably a way to organize your description so you don’t have to repeat this for both tests.
Tables and figures have to be numbered sequentially starting at 1.
Why are you reporting trapping hours instead of days, if it doesn’t add resolution? The number of days would be easier to read.
I don’t want to learn M and U, you can write them out.
What does the s stand for in Table 2?
Tables 1 and 2 would be more concise if the 2 years were set as different rows instead of in different columns.
Table 4 could be organized to first show the habitat types common to both sites and then the ones found only at the managed site. This would save the column for M & U and make it easier to get the description for a site. If the types are organized in an important order, you should ignore this suggestion. Capitalization should be consistent.
Table 5 seems like something you need for completeness rather than something the readers will care about, so are the two lines for totals worth the space they take? If you want readers to see a pattern, a figure would be better.
Table 6: Total captures includes radio-tagged turtles: I was confused. This suggests that the other counts do not? Without the footnote, I thought total captures meant that the same turtle can be counted multiple times, compared to the total number of turtles.
Comments not worthy of a response to the Editor
47 subjected, I think, but it’s not a great verb. “Burned” is a great verb, but I appreciate the need to distinguish controlled burning from wildfires.
48 Notable? Or someone noted them?
58-49 We don’t need to know what you were doing when you weren’t surveying. Here, I think we need the intensity of surveying activity. Tell us about the other activities elsewhere, if relevant.
65 If you have a comma in an item in your list, you need semicolons to separate the items. Maybe you can find a better solution (split into more sentences).
67, 68 Sex and age are described later (so you can omit them here). Health conditions sounds vague.
74 all turtles? How many?
108 Is this necessary?
109 I’m pretty sure this is not necessary.
142 “within 23 years” of what? “Over a 25-yr period” was better.
Is the number of turtles “captured and marked” different from the number captured or the number marked? If all the ones captured were marked, then “observed” or “studied” would work.
146 seems like the parenthetical comment should refer to the sites. Maybe you don’t want this sentence at all; it’s methods, right?
153 Rather than say that the variable was not important, you could say that males and females were captured at similar rates.
Are rho and phi conventional symbols for capture and survival? If not, don’t introduce new symbols. Actually, I’m not sure about showing this table.
Table 8. If you showed us the coefficients, it would matter what you dummy variables were, right? But since you don’t, we don’t care.
Anna Harrison
This paper describes the use of sites by beavers in an area of the Adirondacks where beaver lodges have been monitored for 30 years. The current study characterized 14 sites to see which factors best explained the frequency of use by beavers. I’m not an expert in this field, but the study seems novel and likely to be useful.
General comments
It seems like a major gap to have no information on the numbers of beavers. Maybe this topic will be addressed in your Introduction and Discussion. Still, in the Results, it is important to define what you mean by “occupancy.” It’s the fraction of years in which a site was occupied, and it doesn’t reflect whether there were a lot or only a few beavers. If “occupancy” is not a term that is well known to your audience to mean the frequency of occupation, maybe you could come up with a better term.
It’s important that you characterize the landscape only at the end of the 30-year period. You need to provide the dates for your field measurements, and make clear that they do not cover the same period as the occupancy data. Again, this will probably be more clear in the Introduction and the Discussion, but the reader who starts with your Results also needs to understand your study.
Is there any difficulty in determining what a site it? It has one lodge? If a big site could have been considered to be two little sites, this would clearly affect the probability of occupancy.
I’m not crazy about “drivers” of beaver occupancy: it suggests causality. What you have are correlations. You could use the word “predictors” which has a well defined meaning in statistical models.
The distinction between stream sites and other sites is important in the Results but is not described in the Methods. What is a stream site? Does it have no ponds? No, there are no sites in Table 1 with 0 ponds. There is a sentence about natural wetlands (l. 150), which I didn’t understand. The experimental design should be clearly explained in the Methods.
The Results as written seem to be about the statistics rather than about your subject matter. Can you write sentences about the beavers and the landscapes, citing the statistical models?
I don’t understand the purpose of presenting so many models. It reminds me of stepwise regression, but there a single analysis can be used to judge which variables are important to include. Do you expect all of these models to be used, or is it a matter of presenting the evidence that the best one is the best?
You have material in the Results about application of your research, which will be good for the Discussion session.
Specific comments
27 I don’t think we care about the extent of the HWF unless it coincides with your research area.