October 21, 2024

Dear Editor

Thank you very much for the opportunity to review the study, ¨Tree diversity increases forest temperature buffering¨ submitted by Schnabel et al. and currently under consideration for publication at Ecology Letters. The paper argues that since forests ‘create’ their own microclimate below the canopy, consequently tree species diversity may play a role in temperature buffering. They define temperature buffering effect to occur when ‘below canopy’ temperature fluctuations are smaller than macroclimate fluctuations. Further, they analyse this temperature buffering across different temporal scales (daily, monthly, and annual) and across seasons (winter cf. summer). Using microclimate measurements over 6 years (2015-2020) over a network of field plots along a richness gradient of 1 to 24 species, they find that temperatures inside forest are higher at night (likely due to heat retention and longwave emission) and lower during the day (due to evapotranspiration). Further, they find that buffering is a function of season (warmer night time microclimate minimum temperature in winter, cooler day time maximum microclimate temperature in summer) and soil moisture (less buffering magnitude in dry years but similar slope). The paper is well written, timely, and investigates an interesting and pertinent topic since macroclimate measurements might not fully represent the impact of climate change on forests. I would like to add that I am not an expert in Structural Equation Modelling. Therefore, while methods look sound to me, I am not able to evaluate them fully. I really do not have any major comments or concern and think this this manuscript is ready to be published with little changes based on other reviewer comments.

Some minor comments are included below:
1. The referencing to the supplementary section and associated figures is not very clear throughout the manuscript. It might be better to divide or clearly label supplementary items as Suppementary Text xx or Supplmentary Figure xx so that the reader know what section they should be looking at.

Line Numbers based on continuous numbering (not the page-wise ones)
Line 85-86: This difference is larger than the average warming of land and ocean temperatures in 2001–2020 compared with 1850–1900 (0.8 to 1.1) °C (IPCC 2021).
Not a big deal but, perhaps warming over land may be more appropriate than global (land+ocean temperature). Since oceans warm less due to their lower head capacity

Line 91: Does photosynthesis scale exponentially with leaf temperature? Respiration yes, but this is probably not correct for photosynthesis?

Line 179 & Supplementary Figure 1 – what does VIP stand for? “VIP Plots”. Not defined until later.

Lines 198-202: How the CRU dataset is used is not fully explained. I would be concerned if CRU is used as the representation of macro-climate and in-situ below canopy sensors for micro-climate. Given the methods using which CRU is created, it should be expected that there are biases at the local/ground scale at which CRU is not really appropriate. Could the authors explain how CRU data are used? And why they are used, since presumably the have local in-situ data that samples ‘macro-climate’ conditions.

Line 312: On the daily scale, we found below-canopy air temperatures to decrease with tree species richness during daytime, while they increased with species richness during the night (Fig. 1a).
Could you explain how this figure (panel A) accounts for changing daily temperature over the course of the year?

Line 421-422: Confirming H1, species-rich forests cooled high and insulated against cold macroclimate temperatures better than species-poor forests.
I think experimental plots is probably more accurate than the word ‘forests’. These forests plots are not fully independent forests.

Figure S3: What is the solid black line? It doesn’t fit the points and is not explained

Figure S7: Macro climate effects on the relationship between tree species richness and maximum “microclimate” temperature. The change in these diversity effects with macroclimate conditions is examined for maximum (A), average (B) and minimum (C) monthly macroclimate temperatures and monthly values of SPEI. The figure is similar for S8 and S9 but for Median and Minimum temperature.
This figure is a bit confusing (to me) in how it is described. I think it might be easier to understand if they title clearly explains that the diversity effect is on microclimate conditions and the subheadings should be Maximum microclimate temperature, median microclimate temperature, and minimum microclimate temperature (if I understood correctly).

I am happy with the revised manuscript. The authors addressed my main concern about the appropriateness of using the CRU dataset for macroclimate data instead of their instrumental dataset (Figure S12).