The authors present a manuscript using 27 years of data on sex-specific spring emergence and breeding success in Columbian ground squirrels. They find that the plastic response between sexes is similar, maintaining synchronous emergence throughout the study period and that earlier emergence of males relative to females increases the likelihood of males securing a territory and increases reproductive success. I really enjoyed the manuscript as a whole and find the topic and study system really interesting; it will be a valuable contribution to phenology literature. However I have two main concerns; first regarding a component of the analyses that I believe to be important and second relating to the writing around the topic of climate change which I found to be confusing (though I appreciate there seems to be a more complex range of climate effects occurring than in other regions) and I think better clarity on this will improve the broad appeal of the work.

I am concerned about the lack of year random effect in the main models presented. I see models including a year random term were run and are referred to in supplementary information, however I strongly feel these should be the main analyses. Personally, I would not have run models that do not include a year random term as I believe the results from the model including year will be more accurate (for example see Knape, J. (2016) and Daskalova et al., 2021). With that in mind, the models including year are not currently discussed and interpreted at all. The within individual slope is not significant for the Kite field site but this differs to the effect identified from Meadow B. The difference in between-individual slope between sexes is also interesting yet currently undiscussed. Testing for differences between the between- and within-individual slopes for each sex would be a valuable addition as the manuscript doesn’t currently address or discuss this. Particularly if focusing on the models including year as a random term this would be a valuable addition in interpreting the within- and between-individuals coefficients. I believe the method for testing the difference in slopes is presented in van de Pol & Wright (2009).

As mentioned above, my other concern regards the clarity of description around climate change and how it is presented throughout the manuscript.

Using the term favourable in the context of spring conditions is a bit confusing/unclear- I’ve seen that it is a response to previous comments about the use of “more benign” but it still makes it a bit confusing to read for me. Do the conditions need to be referred to as better or worse? Having described that the environmental conditions become suitable for biological events to occur earlier in warmer years, it may then be clearer to the reader to just refer to them as warmer or earlier springs (although see next para).

To me, the discussion of climate change as a whole (particularly in the introduction in paragraphs 3-5 and lines 112-5) is confusing. Warmer conditions are used when discussing the effects of climate change (as would be expected by the reader), but climate change so far doesn’t seem to have brought warmer springs/earlier spring emergence for the Columbian ground squirrels as their timing has delayed. I think the introduction of how climate change ties into this work needs a bit more attention and to be laid out quite clearly as it is quite different and more complicated than the most commonly reported situation of spring temperatures increasing and timing advancing. How has the climate changed- ref 32 seems to suggest no temporal trend in temperature but increase in late winter precipitation so delayed snowmelt? From previous work what is known about how temperature, precipitation and snowmelt timing affect emergence date independently? Given the temporal trends in climate and snow melt date, how has their timing changed? The focus on temperature makes me wonder whether it is predicted that in coming decades temperature will increase and drive a temporal trend in the timing of snowmelt? If so then that should be made clear and if not then discussing the hypotheses in the context of warming seems somewhat removed from the study system, unless clearly described that the hypotheses have been introduced in the more general context and then describing how you interpret them within this system.

Refs:
Knape, J. (2016). Decomposing trends in Swedish bird populations using generalized additive mixed models. Journal of Applied Ecology, 53(6), 1852-1861.

Daskalova, G.N., Myers-Smith, I.H., Phillimore, A.B. (2021). Accounting for year effects and sampling error in temporal analyses of population and biodiversity change – Response to Seibold et al. 2019 “Arthropod decline in grasslands and forests is associated with landscape-level drivers”. Insect Conservation and Diversity.

van de Pol M, Wright J. 2009 A simple method for distinguishing within- versus between-subject effects using mixed models. Animal Behaviour 77, 753–758

Comments by line

Title: I’m not sure whether it is quite accurate to say sexual selection is maintaining synchrony if earlier relative timing in males has higher fitness? Also see comments below about the age effects on fitness

Line 32-33: I think “increases in protandry … [to the end of the sentence]” could be removed as it is a bit unclear what is meant.

Line 114: “in response to warming temperatures” seems inaccurate? And is ref 22 meant to be there? Didn’t seem relevant but I could have missed something

Line 123: “Here we evaluate” might read better

Line 186: “the laboratory birth of young” has come out of no-where, is there a citation for this? I’m unclear on how it is relevant to this work, apologies if I’ve missed something

Line 252: Why was age not included in the territorial status model?

Line 281: Have snowmelt dates delayed? The SE is bigger than the mean slope! Is that right?

Line 285: Are those estimates for behavioural emergence or torpor termination?

Line 309: You say it weakened the effect, but the within-individual effect is no longer significant when year is included as a random term. The models with year random terms should also be mentioned in the methods.

Fig 2: showing some uncertainty on the slopes would be good, especially in plot a.

Fig 5b/d. Shouldn’t the age plots here show a quadratic trend? That coefficient doesn’t seem to have been accounted for in the plotted slopes

Line 369: Is there no table for this model (either main or supplementary?) and should this say “significant variance in individual random slopes”?

Line 377/8: The age and age^2 effects should jointly describe the trend in timing as a function of age so it’ll be easier to interpret the difference between males and females by plotting the trend including both in Fig 5d. As the age^2 coefficient is positive, I think the negative age:male effect suggests the males breed at their earliest at a later age than the females, but that will also depend on the effect of the age^2:male coefficient has on the turning point in the curve.

Line 380: The significant age^2:sex effect with year included is interesting as that suggests a stronger quadratic trend in timing with age in males than females.

Fig S5: There isn’t a Table 3b is there? It would be good to state what the uncertainty around the lines is- as in confidence intervals or SEs (this goes for other figures too).

Line 425: ref 21 isn’t Møller

Line 427/8: with advancing warmer spring conditions – a suggestion for removing increasingly favourable

Line 458: The equivalent for Columbian ground squirrels would be nice for comparison

Line 484: The non-significance of emergence date with the inclusion of age could do with more discussion/interpretation and presents a topic that would be interesting to investigate further in future.

Line 494-6: Again on the topic of the age effect overshadowing the effect of relative timing- perhaps the timing of breeding itself is not having a causal effect on breeding success and the effect identified when age was excluded may more reflect higher quality/prime age individuals breeding earlier. Given the variation in timing with age it is hard to know the direct pathway from your results so presenting the various options is important. If the effect of timing on fitness is purely a correlation driven by the effect of age on fitness and variation in timing with age then there isn’t necessarily directional selection acting on relative timing which would mean similar plasticity between sexes isn’t surprising.