Contrasting behaviors of the atmospheric CO 2 interannual variability during two types of El Niños

Abstract. El Nino has two different flavors: eastern Pacific (EP) and central Pacific (CP) El Ninos, with different global teleconnections. However, their different impacts on carbon cycle interannual variability remain unclear. We here compared the behaviors of the atmospheric CO 2 interannual variability and analyzed their terrestrial mechanisms during these two types of El Ninos, based on Mauna Loa (MLO) CO 2 growth rate (CGR) and Dynamic Global Vegetation Models (DGVMs) historical simulations. Composite analysis shows that evolutions of MLO CGR anomaly have three clear differences in terms of (1) negative and neutral precursors in boreal spring of El Nino developing years (denoted as “yr0”), (2) strong and weak amplitudes, and (3) durations of peak from December (yr0) to April of El Nino decaying year (denoted as “yr1”) and from October (yr0) to January (yr1) during EP and CP El Ninos, respectively. Models simulated global land–atmosphere carbon flux (F TA ) is able to capture the essentials of these characteristics. We further find that the gross primary productivity (GPP) over the tropics and extratropical southern hemisphere (Trop+SH) generally dominates the global F TA variations during both El Nino types. Regionally, significant anomalous carbon uptake caused by more precipitation and colder temperature, corresponding to the negative precursor, occurs between 30° S and 20° N from January (yr0) to June (yr0), while the strongest anomalous carbon releases, due largely to the reduced GPP induced by low precipitation and warm temperature, happen between equator and 20° N from February (yr1) to August (yr1) during EP El Nino events. In contrast, during CP El Nino events, clear carbon releases exist between 10° N and 20° S from September (yr0) to September (yr1), resulted from the widespread dry and warm climate conditions. Different spatial patterns of land temperature and precipitation in different seasons associated with EP and CP El Ninos account for the characteristics in evolutions of GPP, terrestrial ecosystem respiration (TER), and resultant F TA . Understanding these different behaviors of the atmospheric CO 2 interannual variability along with their terrestrial mechanisms during EP and CP El Ninos is important because CP El Nino occurrence rate might increase under global warming.


Detailed comments
(1) Given the strong similarity of broad focus of this work with the recent Chylek et al paper, it might be worth adding a paragraph to the discussion that summarises the differences and similarities in approach and results e.g. exclusion of events that coincide with volcanic eruptions, identification of different events, inclusion of TRENDY and inversion results, focus on lag by Chylek, conclusions etc. Do you also see a difference in the lag? Is there anything from the TRENDY results that could shed light on the hypothesis from Chylek that the shorter time lag between the temperature rise and an increase in CO2 emissions with CP El Ninos is influenced by fire response, while the longer time lag in EP El Ninos is dominated by vegetation response, noting although that the TRENDY models exclude or underestimate the effect of fire (maybe therefore there isn't anything you can add here, but at least worth thinking about)? Although there is a strong overlap of focus of this work with Chylek there are also significant differences, so I do believe that there is value in both studies.
Reply: Thanks very much. We have added a paragraph in the discussion section to simply illustrate the differences and similarities between our work and Chylek et al.  Reply: Thanks for your constructive suggestion. We have modified it accordingly.
(3) page 2, line 44 -the sentence that begins "Regionally, significant anomalous ..." is long and you don't know which type of El Nino event this sentence refers to until the end. I suggest beginning the sentence something like "Regional analysis shows that during EP El Nino events significant anomalous ..." or some other way to mention EP at the start.
Reply: Thanks for your suggestions. We have modified it accordingly.
(4) Page 5, line 111 -word "carefully" should be unnecessary Reply: Thanks very much. We have deleted it.  Reply: Thanks very much. We have modified it accordingly.
(7) Page 9, line 198 -".. with noticeable increases *in CO2 growth rate* during ..." Reply: Thanks very much. We have modified it as "…with noticeable increases in CGR during El Nino and decreases during La Nina, respectively". Reply: Thanks very much for your suggestion. We have modified it accordingly.
(9) page 12, line 267 -how are you defining the MLO CGR peak here?
Reply: Thanks very much. We have added the definition in the text. We define the peak duration as the period above the 75% of the maximum CGR or F TA anomaly, in which the variabilities of less than 3 months below the threshold are also included.
(10) page 14, line 305 -"GPP anomalously increases ...etc" Can you check this sentence reflects the variations in Fig 4b? Would it be more accurate to say that there is a peak in GPP during austral fall (yr0), and is low from austral spring and winter (yr1)?
Because austral summer spans from one year into the next, be more precise when you mention austral summer. Also be careful with the word increase (could be interpreted as talking about the trend) versus high values through this section.
Reply: Thanks very much for your suggestions. We have checked it and modified into "GPP showed an anomalous positive value during austral fall (yr0), and an anomalous negative value from austral fall (yr1) to winter (yr1), with the minimum around April (yr1) during the EP El Niño (Fig. 4b), …" (11) page 16, line 349 -perhaps swap the order of figs S3 and S4 in the supplement, as S4 is always discussed before S3.
Reply: Thanks for your suggestion. We have swapped their order. Reply: Thanks for your suggestions. We have pointed out this and modified as "Both GPP and TER showed the anomalous decreases ( Supplementary Figs. S3a and b), and stronger decrease in GPP than in TER makes the anomalous carbon releases here (Fig.   6c)."