|“Sampling density and date along with species selection influence spatial representation of tree-ring reconstructions” is a clear, concise manuscript comparing a gridded drought construction made from network of updated, multi-species, tree-ring chronologies in the Ohio River Valley with the North American Drought Atlas. This comparison demonstrates the influence of increased predictor density and record length on gridded drought reconstructions. |
The clearly laid out discussion showed the power and limitations or increasing the predictor density in a spatial reconstruction. The authors’ conclusions regarding the “fading drought signal” of trees is important for future hydroclimate reconstructions – particularly in this region.
This manuscript should be published pending minor edits, please see the below line items.
Line 22: suggestion to replace “are represented” by “are modeled by few tree-ring chronologies”
Line 45: replace “represent” with “are”
Line 46: replace “/” with “and”
Line 55: delete “an” and replace “to extreme” with “for extreme”
Line 61: Suggestion “They are most useful for large, regional (what is regional here? maybe instead of regional be specific on what scale of climate/ecosystem ranges/ grid cells etc.) events”
Line 61-62: Instead of “each gridded reconstruction” “The reconstruction at each grid cell uses tree-ring data that are within a 450-km radius of that grid point”
Line 62: Rather than “therefore”, I suggest more specificity: by pulling from such a wide range of predictors, the NADA and LBDA models excel at representing large-scale phenomena as they tend to average out smaller scale features…
Line 77: perhaps elaborate on why the chronologies not being updated is important.
Line 79-83: Between this sentence and the preceding paragraph it feels a bit redundant
Line 96: add “despite a fading drought signal” at the end “of the instrumental data”
Line 98: you’re referencing the reconstructions with the new, dense, network right? Possibly specify which reconstructions you’re discussing.
Line 113: “significantly correlated” at what alpha level? Cook 2010 does 0.01 I believe?
Line 127: You mention “standard field methods” --- standard sampling design, or standard collection methods - from dendroclimatology (canopy dominant oldest) rather than ecology? Coring the tree at breast height seems more ecology to me
Line 142: regarding signal free standardization, possibly replace “standardization reduces” with “standardization can reduce”. Just to place my arbitrary horse into a stupid race – Signal Free isn’t a silver bullet to reduce the trend distortion- I'm glad it does here for this network.
Line 145: EPS value of 0.80 – why not 0.85 as in the reference cited (Wigley et al., 1984)
Line 152: Only the chronologies with significant at p <0.05 level – is this the same as Cook, 2010? Please state if you are as, more, or less stringent in your cut off
Line 164: replace “conducting” with “conducted”
Line 166: You have already defined JJA, don’t need to redefine
Line 168: as done by Cook 2010?
Line 176-Line177: You use simple Pearson’s correlation rather than beta weights of the species to the recon – why?
Line 196: Again, why 0.8 not 0.85?
Line 218-220: This is really neat. One of the reasons I think this paper is great!
Line 227: My suggestion to just say “whereas 18 of the sites included…” rather than “n=18”
Line 227: Do you mean multi-species when you say multiple chronology?
Line 238: really a maximum of 2006 years in this location as well? I'm surprised - I don't think it goes back much past 490CE in this region as far as I can tell from the raw data files. If you mean 2006 years over the entire network, better to specify.
Line 248: I would put the sentence starting with “Specifically” in the methods.
Line 386” “suggesting a wet period”… very very cool.
Figures: in general- the figures are much improved
Figure 2: suggestion for consistency put degree N and degree W on the grid
Figure 3: Make PMDI bigger
Figure 4: It would be better for the readers to contextualize this figure a bit more with a brief sentence on the climatology of 1954 in the maintext- we can see from the instrumental that we have wide spread drought in much of the area but slightly wetter conditions in the north – is there a reference about this particular summer?
Figure 7: increase the size of “number of flips” and make the font the same kind as the sup titles?
Figure 9: I suggest to put a color bar for the calibration R2 as well since every other figure has a color bar?
Figure 10: Again, ensure that the font is the same and increase the color bar label and color bar axes. Also I suggest to keep consistent with your figure formats: choose for your figures degree N and W (like figure 1) or “lat and lon” (like the rest of them).