General
This manuscript provides interesting information describing the migration patterns and wintering distribution for Common Loons using both satellite telemetry data and geolocator data. The sample sizes achieved are impressive for a species that occurs at low densities and can be difficult to work with. The dive data presented are particularly interesting and one of the first studies to identify dive patterns of loons.
Currently the manuscript is very long and in its current form it can be difficult for the reader to follow all of the individual pieces (e.g., satellite transmitters vs. geolocators, adults vs. juveniles, departure dates from breeding and wintering areas, migration routes, stopover duration and sites, wintering duration and location, foraging patterns, differences between locations, mortality of juvenile loons, etc). I suggest splitting the manuscript into multiple manuscripts. One manuscript could describe the wintering distribution and migration patterns and the other could describe the foraging patterns using the dive data. This would shorten and greatly strengthen the current manuscript by narrowing the focus.
I also have a few concerns regarding the methods. The entire Methods section is difficult to follow and lacking important details. The reader is left to guess about how the data are being used making it difficult to evaluate the methodology. Specifically, it is not clear how the PTT data and geolocator data were used (e.g., were geolocator data used to determine wintering locations?). The authors should also address the large differences in accuracy between satellite telemetry and geolocator data.
Much of the Results section lists the patterns for each individual loon. I suggest summarizing this information and not discussing individual loons. Similarly, most of the Tables (2-5) are just lists of information on individual loons. These Tables would be better off summarizing the information (e.g., mean distance to shore, water depth, winter range sizes, etc) and the current tables listing individual loons could be moved to the supplemental material.
Finally, many of the paragraphs in the Discussion are simply restating the Results without any interpretation. The Discussion would benefit from additional information about what the results mean and how they can be applied.

Specific comments by line
Lines 4-6. The Abstract would benefit from more context added to the beginning introducing the study and include the rationale for the research and main objectives.
Lines 7-9. The percentages of loons wintering in the different areas are hard to interpret because sample sizes of marked birds are not given.
Lines 28-35. The manuscript begins by describing the distribution of the study species. The Introduction would greatly benefit from a broader opening paragraph (or two) setting up the study. Perhaps a paragraph about why it is important to identify migration routes and timing of species.
Lines 28-53. The sentences in these paragraphs are somewhat disjointed without much connection to one another or how they might relate to the study. For example, the sentence in Lines 48-51 is related to juveniles remaining on wintering areas, while the following sentence is about loon survival with no connection to the previous sentence about wintering areas. Further, the first two paragraphs simply list the distribution of common loons but it isn’t clear to the reader why this information is being presented or why they should care. The Introduction could be greatly improved by adding some context and connections between the sentences in these paragraphs. For example, adding a sentence following line 53 about why it is important to quantify juvenile survival rates would help add context to the paragraph.
Lines 30-33. The percentages don’t add up to 100%. Where do the other 1% breed?
Line 37. Some what? Individuals?
Line 54-56. The citation here is from 2007. Has there been a more recent Assessment and Conservation Plan for Common Loons?
Lines 56-59. It seems likes much of this has already been done in Kenow et al. (2002), although the data presented in this manuscript are more recent data and have much larger sample sizes. It would be interesting to compare results from the two studies (i.e. identify any changes over time).
Lines 61-68. Most of this paragraph discusses the potential impacts from oiling; however, oiling is not addressed again until the last sentence of the Discussion.
Lines 69-73. This section needs to be expanded as two sentences outlining the objectives of the study is not enough background or justification. Further, significant parts of the study, such as the foraging data, are not mentioned anywhere in the Introduction.
Line 70. The locations where loons winter are on the ocean (i.e., water) and therefore it would be more accurate to call them “wintering areas” instead of “wintering grounds”. Change throughout the manuscript.
Line 77. Change “obtained from” to “captured on”
Lines 79-81. What about Michigan DNR?
Line 89. Since a molecular genetic assay was used it can be assumed that a tissue sample (e.g., blood or feathers) was collected but this isn’t explicitly stated.
Line 90. Change “was” to “were”
Lines 92-93. Geolocators do not collect “daily location estimates.” They record light levels (at an interval much more frequent that daily) which are then typically used to calculate twilights to obtain location estimates. Replace “daily location estimates” with the frequency that geolocators were programmed to record light levels (most manufacturers default is every five minutes).
Lines 109-111. This information should be moved to the Acknowledgements.
Line 116. It seems like locations received from satellite transmitters were not filtered? I suggest using a filter, such as the Douglas Argos Filter Algorithm, instead of the best location class for each 8-hour transmission period.
Lines 116-120. It would be helpful to have the information regarding PTT duty cycles, and geolocator processing methods in this manuscript as well so the reader doesn’t have to look at another paper to find this information and guess what methods were used.
You should also clearly state how the geolocator data and PTT data were used. These two methods have very different accuracies. Were geolocator data used to determine stopover/wintering locations or just for the foraging information? Lines 120-122 suggest geolocator data were used to determine gross movement patterns and locations, but this is not clear.
Lines 125-129. It isn’t clear from the Methods why the carcasses were recovered and necropsies were performed.
Lines 129-130. This information should be moved to the previous paragraph because information about geolocator processing doesn’t make sense when the rest of the paragraph is about loon mortalities.
Lines 139-141. How did you determine the latitudes to estimate gravity? Were the geolocator data used to estimate locations or just the PTT data? Given the low accuracy of geolocator data, it seems like there could be error in dive depths if location estimates from geolocators were used to estimate gravity.
Lines 151-154. Again, the differences in how you used PTT and geolocator data are not clear. The Methods state that all PTTs were deployed on adult male loons (Lines 101-102) but here there are models that account for differences in sex? It also isn’t clear the sample sizes for each sex.
You should explicitly state why you included a sex x transmitters presence/absence interaction. I assume this had to do with differences in body sizes between the sexes.
Also, Lines (88-89) state that a genetic assay was used to determine sex of juvenile loons but the manuscript doesn’t address how sex of adult loons was determined. Adult plumage is the same for males and females so how was sex determined (e.g., mass measurements)?
Did you compare migration routes and wintering areas between the sexes?
The farthest right column of Table 1 suggests that only geolocator data from adults were used in dive analyses but this is not clear from the text in the Methods section. It seems like you obtained geolocator data from 12 juveniles. If data from juveniles were used in the dive analyses, why didn’t you evaluate an adult x juveniles interaction? It seems like there might be foraging differences between age classes.
Lines 163-164. Do you mean that different bouts were separated by 360 seconds but individuals dives within 360 seconds were considered part of the same bout?
Lines 168-169. It seems like only LC3s should be used for determining the distance to shore and water depth variables. Accuracy estimates for LC1s are between 500-1500m and could results in a large error when calculating water depths and distance to shore.
Lines 170-171. It might be nice to provide the reader a short summary of the differences between these winter range estimates and why you used all three instead of picking one method.
Line 188. Here you could add the mean and range of exposure days.
Lines 189-204. This is just a long list of each individual loon that died or the signal was lost. This could be shortened by simply stating the number of loons that died and the number of signals that were lost and then provide the specific information in the supplemental material.
Lines 207-208. This sentence isn’t clear. Are you saying that recapture rates were 65% and geolocator retention was 83%?
Lines 211-212. You could reference Figure 3 here.
Lines 222-223. This information regarding loons using Lake Michigan is already stated in Lines 211-212.
Lines 227-232. You could cite Figures A1-A16 here to show the migration routes.
Line 313. Why is it important for the dive to have a “square shape”?
Lines 328-336. The individual Loon IDs throughout the manuscript don’t add anything to the paper and mean nothing to the reader. Delete
Line 332. Delete the extra “(“
Lines 333-334. This information would be better off in the mortality section.
Lines 335-336. This information would be better off in the Wintering Areas section.
Lines 338-339. Delete this sentence and reference Figure 12 following one of the next sentences. For example, Departure of juvenile loons from natal lakes varied from 17 September to 19 November (Figure 12).
Lines 343-344. Clarify that some juveniles returned to their natal lakes following initial departure. How long were they gone before returning to their natal lake?
Lines 346-347. Did you actually look at dates of ice formation and departure dates? No ice formation data are presented.
Lines 366-368. One of the loons in Table 5 has an MCP of 112,000 km^2 but this sentence states that the range varied from 651.5 to 48,818.5 km^2.
Further, many of the juvenile loons have a “winter range size” of > 10,000 km^2 using the MCPs. These certainly do not seem like winter ranges and winter range should be better defined (i.e., 112,000 km^2 is a huge area and likely doesn’t represent a winter range). I assume this is because the juvenile loons moved a lot throughout the winter. This needs to be acknowledged and addressed.
You also should address the huge differences in in winter range sizes between methods (e.g., MCP vs 95% kernel UD). Given the large size of the MCP estimates perhaps they are worth removing from the manuscript or at least stating why they are so large compared with the kernel and core use areas.
Lines 376-380. Instead of citing individual loon ID numbers, you could cite individual Figures from A13-16.
Line 399. Change “expire” to “was depleted”.
Line 401. Clarify that you are discussing juvenile loon mortality in this section.
Lines 447-450. These numbers are already presented in the Results.
Line 462. “Elsewhere” used here suggests that common loons from other breeding populations use the Gulf of Mexico. If so, please state the other breeding populations that use GOM during winter.
Line 465. What is considered the midday period?
Line 470. Where were these other wintering areas?
Lines 472-478. Why do you think the dive times were longer in the GOM than Lake Michigan? This likely suggests greater energy expenditure during winter to catch food.
Lines 493-502. This section listing individual diet studies should be shortened. For example, Wintering common loons have been observed eating benthic fish species, crabs, winter flounder, Gulf menhadem, gulf silversides, Atlantic croaker, and spot in the Gulf of Mexico and Atlantic (Haney 1990, Ford and Greg 1995, Spitzer 1993, Vliestra 2000).
Lines 505-513. This section is simply restating the Results with no interpretation.
Lines 522-523. Did you collect data from marked pairs of adults and their juveniles to compare? This suggests that juveniles do not follow adults during first migration and are on their own. It would be very interesting to compare winter locations of juveniles with the wintering locations of their parents.
Lines 539-542. Why do you think juveniles are migrating faster? Perhaps this is a result of the lack of awareness of stopover sites as stated in previous paragraph (Lines 529-531).
Lines 546-547. Why such broad distribution by juveniles? Maybe juveniles were searching for good foraging habitat?
Lines 554-565. This paragraph is another example where the Results are being restated without adding any new information or discussing the implications. Delete or expand on this section to include interpretation of results.
Lines 566-570. This long sentence should be broken up into two separate sentences.
Lines 581-586. Explicitly state why juveniles might escape botulism exposure. Is it related to differences in arrival dates or length of stay at the Great Lakes between juveniles and adults?
Lines 612-627. Delete the Loon IDs throughout this section.
Lines 621-625. Instead of stating that juvenile loons “returned to the same general wintering areas used during their first winter” you could calculate the amount of overlap of the winter ranges between the two years of data (e.g., report proportion of winter range overlap).
Line 627. This information about a single area seems superfluous. Delete “including a protected bay (Barataria Bay, Louisiana) in the GOM.”
Line 628. This idea is already brought up in Lines 458-462.
Lines 632-633. Based on Figures 2 and 5 in Evers et al. 2020 Birds of the World COLO account, the duration of the non-breeding period of the annual cycle is more than half a year.
Lines 640. How do you protect with such large home ranges and wide distribution.
Lines 644-647. Either expand on this section or delete. Explicitly state how this information could be useful for conservation planning and inform mitigation efforts related to oil spills.
Table 1. I suggest adding columns regarding the number of males and females.
Table 2. Instead of a long table listing individual loons, it would be better to summarize the mean distances and water depths for the GOM and Atlantic. Then this information with each individual could be moved to supplemental material.
Table 3. Same comment as above where it would be much better to summarize mean winter range sizes than listing the range size for each individual loon.
Tables 4-5. Same comment as above for Tables 2 & 3.
Table 9. This table doesn’t provide much information and should be moved to supplemental material. Where the birds died are already included in Figure 15.
Figure 1. Change “Distribution” to “Locations”. What do the different colors represent?
Figure 3. What do the different colors arrows represent?
Figure 5. Are the dark circles the 50% core use areas? Why are these only male loons? Why not include female wintering locations?
Figure 6. “Various areas” is vague. Could you call these “stopover sites” or “staging areas” instead?
Figure 15. The base-layer of this figure is not consistent with Figures 5 and 13.

General Comments
The overall structure of the manuscript has been improved from the original. The added sub-headings make it easier to follow all of the different pieces of information. The Results have also been improved by removing most of the information about individual loons. However, many of my previous concerns were not addressed. If a reviewer states they cannot find information, a response saying it is in the paper or clear to the authors is insufficient. Papers must be clear to the reader not the author. Instead of simply answering reviewer questions in a response letter, add information to the manuscript addressing their points because other readers might also have these same questions. Further, I still find some of the Methods difficult to follow and much of the discussion still strictly repeats the results without any interpretation (e.g., Lines 467-468, 474, 475-476, 479-480, etc). After reading the updated draft, I also still feel like the manuscript is too long and would be better broken into multiple manuscripts to help readers digest all of the information (it is a lot of information to wade through) but understand why the authors would want to keep it as a single manuscript.
Specific Comments by Line
Line 37. Common loon breeding range is mostly restricted to the boreal forest region and is very limited in Arctic North America.
Line 53. See suggestion for rewording sentence below.
“Pairs of adult common loons generally migrate independently of each other and their chicks.”
Line 65. Impacts of the Deepwater Horizon Oil Spill on common loons has been evaluated (see Paruk et al. 2014, 2019).
Lines 86-87. This is the first mention of dive/foraging patterns in the Introduction.
Lines 97-100. This suggests that lakes were not selected randomly
Line 105. Is this so that you could compare location estimates derived from geolocators vs satellite transmitters? If yes, then explicitly state this.
Line 113. How frequently did the geolocators collect light level data?
Line 119. Replace “were not” with “cannot be”
Line 121. Replace “devices” with “data”
Lines 152-156. Detail is greatly lacking. It is unclear how sea surface temperature and dive depth information were used to improve location estimates. Citations are also needed.
Line 157. Throughout the manuscript multiple terms are used to describe the data collected from geolocators (e.g., “geolocator tag data” in line 147, and “geotag data” in line 157). Be consistent with usage.
Many geolocators are equipped with a wet/dry sensor. Were these also used to interpret flight times?
Lines 158-159. What does it mean to “set bounds on the distances of migration events?”
Line 162. Was there a temperature threshold used to distinguish a mortality?
Lines 162-171. Why is only juvenile loon mortality reported? If there were no adult mortalities, this should be stated in the Results.
Lines 177-178. The link provided here as a citation is broken. How did you determine to use this equation?
Lines 178-181. According to Figure 8, loons ranged from ~26-36 degrees latitude. It seems like using estimates of gravity for the mean (29 deg N) incorporates a lot of potential for error. Because wintering locations of loons were estimated, shouldn’t you just use the latitude for each wintering location to calculate the gravity?
Lines 185-190. Not sure what this means. Daily ascent and descent rates for individual dives?
Line 197. The URL link is broken.
Line 198. Does this mean you used the mean depth on each dive? Clarify.
Line 209. What does coupled with bathymetric data mean? I assume this means that you calculated the water depth at each location for each loon. Be specific.
Line 211-212. It seems like only LC3s should be used for determining the distance to shore and water depth variables. Accuracy estimates for LC1s are between 500-1500m and could results in a large error when calculating water depths and distance to shore. Think about the potential difference in water depth if there is over 1km of error for a loon location. Just because the authors are comfortable using location classes 1 and 2 doesn’t mean the resulting water depth and distance to shore variables are accurate!
Lines 219-220. This sentence is unclear. What does “the intersections of each home range” mean?
Lines 233-236. State that movements and fates are for each individual.
Line 259-260. Citing a different publication that found that Common Loons use Lake Michigan in the fall is not a result of this study and should be deleted or moved to the Discussion.
Lines 271-274. Why did you differentiate male geotagged loons versus radiomarked loons? Did you think there might be differences in migration speed or arrival/departure dates due to transmitter attachment methods, etc? If so, this should be addressed somewhere in the manuscript (the Methods). Also, this is the only place where this difference is compared (e.g., Lines 280-281 don’t differentiate between tag types and lump all males together). Why only make this comparison for arrival dates at wintering areas?
Lines 283-288. Clarify that these numbers include multiple stops by some individuals.
Lines 293-344. Did you deploy transmitters on juveniles that were from the nest of adults that were also tagged? If so, it would be very interesting to compare arrival/departure dates and staging/wintering areas of the juveniles/adults from that nest.
Instead of responding to the reviewer that you did not mark an adequate sample of progeny of previously radiomarked adult loons, add a sentence to the manuscript that addresses this point for other readers who might have the same questions.
Lines 307-308. Did you actually look at dates of ice formation and departure dates? No ice formation data are presented. Being “cognizant of when conditions were favorable for ice formation” does not mean that you know specifically when ice formation happened.
Line 321. The loons “appeared” to cross the Florida Peninsula. Why use the term appeared? Did they cross the peninsula or not?
Line 328. Insert “juvenile” after “five”
Line 333. Replace “grounds” with “areas”
Lines 335-336. “Near the same areas used during their first winter” is vague. How close was the area used in the second winter to the area they used during their first winter? You could look at home range overlap between the two winters to explicitly evaluate this.
Line 345. Winter distribution is already addressed in the previous section. Replace “distribution” with “home range size”
Line 355. It is unclear if the term “radiomarked” used here means that the majority of the PTT marked birds used the Florida continental shelf or if this includes individuals with geolocators as well. The prior paragraph uses all 103 individuals.
Lines 355-366. Is this paragraph only discussing adult male loons as is suggested in the caption of Figure 8?
Lines 362. How did you determine 19 as the minimum number of wintering locations to use to quantify winter home range sizes?
Lines 396-399. Here is another comparison between radiomarked and non-radiomarked individuals (and sexes). This needs to be set up in the Methods so readers have some idea why these are being compared.
Line 400. Does Southeastern Reservoirs need to be capitalized? Elsewhere in the manuscript it is lowercase (e.g., Line 393).
This sentence could be simplified to say, “Maximum common loon dive depths were…”
Lines 401-402. The manuscript states that dive characteristics of SE reservoirs are not analyzed but data from SE reservoirs are still included in Figures 10 and 11.
Line 404. Instead of saying mean dive depths “appeared” to be similar, this should be explicitly evaluated.
Line 407. Why is the shallower dive depth in the Atlantic Ocean so “unusual”?
Lines 408-412. These sentences are very difficult to follow. Clarify.
Line 416. End of sentence missing period.
Line 427. Other subheadings in the manuscript also use the term “Spatiotemporal patterns”. Simplify subheading to “Juvenile loon mortality”
Line 428. Add total number of juvenile loons marked with satellite transmitters.
Lines 436-443. This information is all presented in Figure 15 and could be removed (or shortened).
Line 438. Extra “(“ included in this sentence
Line 448. Replace “range” with “locations”
Lines 447-465. This paragraph is rather disjointed with little connection between sentences.
Lines 466-486. Outside of a single sentence about bird movement and weather (Lines 470-473) these paragraphs simply restate the results. Either expand these paragraphs to add interpretation explaining the results or delete.
Line 466-468. Direct repetition of results
Line 468. When the term “movements” is used in this sentence, do you mean “departures” from MN and WI?
Lines 474-477. More repetition of results
Line 485-486. No ice-out dates are presented anywhere in the manuscript.
Lines 493-494. The subject of this subheading is unclear. Perhaps simplify to “Juvenile loon natal lake departure and fall migration”. Also, the subheading is indented and is not in italics like the rest of the subheadings in the manuscript (e.g., Line 446).
Lines 523-533. This paragraph is unnecessarily long. See proposed edit below that shortens and simplifies the paragraph.
“Radiotelemetry studies have documented juvenile migration independent of adults in a variety of bird species (Hake et al. 2003, Gschweng et al. 2008, Peron and Gremillet 2013, Thorup et al. 2003, Yoda et al. 2017). Dispersal movements of juveniles preceding direct migration are common in avian species, similar to the migratory behavior exhibited by many juvenile loons in this study. Juvenile birds may follow a more direct migratory route than adults, as was also the case in the juvenile loons in this study. It is suspected that naïve birds learn new migration routes and dependable foraging opportunities en route from adults during subsequent migrations.”
Line 557. Shorten this subheading. Perhaps to “Common loon wintering habitat in the Gulf of Mexico”
Lines 562-565. This sentence would be better compared to the loons wintering in the Atlantic Ocean than loons in the GOM.
Lines 566-573. These sentences could be shortened. See proposed edit below.
“Florida offshore waters in the GOM are likely a critical wintering area for loons breeding in the Upper Midwest (Kenow et al. 2002, this study) and Canadian interior (Paruk et al. 2014). Preservation of this wintering habitat should be considered as an important conservation strategy to support breeding loon populations as well as populations of other marine birds that also utilize this habitat (e.g., wintering northern gannets [Morus bassanus]; Montevecchi et al. 2012).”
Lines 593-596. This is unclear. How are these dive characteristics consistent with observations of loons staging on Lake Michigan?
Lines 596-600. This sentence could be shortened. Deeper dives typically mean longer dive durations.
Lines 621-622. What about differences in fish communities between the west coast of FL, east coast of TX, and the Atlantic?
Line 603. Satellite telemetry data usually are not considered to be “fine-resolution” given the large errors in accuracy for location estimates (e.g., LC2 has >200m error).
Line 604. Loon dive depth information is presented in the Results/Tables but water depth information is lacking. As currently written, there is no strong case presented that loons are foraging benthically along the bottom (i.e., no water depth data presented).
Line 623. Juvenile mortality was very high in this study and should be addressed somewhere in the Discussion. Do you think this high mortality rate reflects reality (i.e., low juvenile COLO survival) or do you think this is a function of implanting satellite transmitters in young birds?
Lines 630-633. This sentence is very long. See proposed edit below.
“Only a small proportion of juvenile loons stopped in the southern half of Lake Michigan during fall migration. Juvenile arrival at Lake Michigan occurred later in the fall compared with adults, and juvenile stopover length at Lake Michigan was relatively short.”
Lines 634-645. This paragraph is simply just a long list of factors that could contribute to loon mortality. I suggest deleting lines 638-645 and merging lines 634-638 with one of the other paragraphs in this section (perhaps at the beginning).
Line 652. Delete “listed as”
Line 667. Delete “(and migratory routes)”
Line 681. Lines 568-573 already state that wintering habitat is important for conservation of COLOs breeding in the Upper Midwest.
Line 695. Limited information is known about loon diet (as is acknowledged in Lines 611-612. Before implementing strategies that might improve forage fish abundance, researchers would need to identify what those fish species are that make up wintering loon diet.
Line 699. Replace “grounds” with “areas”
Table 2. This still feels like a table that should be in the supplemental material as most readers will not care about individual loons. Instead of a long table listing individual loons, it would be better to summarize the mean distances and water depths for the GOM and Atlantic. Then this information with each individual could be moved to supplemental material.
Table 3. Same comment as above for Table 2. Further, the response letter from the authors states that these two tables were merged, which does not seem to be the case.
Table 4. Replace “gender” with “sex” in table heading and table.
Table 7. Most readers will not be interested in this LONG table that simply lists dates and locations of individual juvenile loon mortalities. Further, most of this information is already presented in Figure 15. This Table should be in the supplemental material.
Figure 1. The figure caption still does not explain what the different colors represent.