Technical Comments

Comment on "Lethally Hot Temperatures During the Early Triassic Greenhouse"

Science  01 Mar 2013:
Vol. 339, Issue 6123, pp. 1033
DOI: 10.1126/science.1232924


Sun et al. (Reports, 19 October 2012, p. 366) reconstructed Permian to Middle Triassic equatorial seawater temperatures. After correct temporal positioning of their data points, their presumed trends of temperature changes, and hence their assumption of a one-to-one relationship between putative "lethally hot" seawater temperatures and a disputable equatorial "eclipse" of some organisms, are no longer supported by their data.

Sun et al. (1) reconstructed Late Permian to Middle Triassic equatorial seawater temperatures from δ18Oapatite of conodonts from sections in South China (their figure 2). We have reassessed their data and detected not only numerous minor inconsistencies but some major methodological flaws. Their conclusion that putative "lethally hot" ocean temperatures led to restriction of fishes, marine reptiles, and terrestrial tetrapods to high latitudes is not supported by their data.

Time calibration of lithologically positioned data points requires piecewise fitting into respective biozones. This is particularly important if various sections are compared or if a composite curve is used, as in (1). Yet, in Sun et al.'s figure 2, several points fall outside the conodont biozone to which they belong (their table S4).

After correct positioning of the data points, the Griesbachian Thermal Maximum and the subsequent Dienerian cooling are no longer recognizable (Fig. 1). Moreover, the samples plotted at the critical Smithian/Spathian transition upon which their "late Smithian Thermal Maximum" is based, fall in fact within an early Spathian zone (C13). The Smithian and earliest Spathian, interpreted to be ~2° hotter than the Dienerian, are not hotter than the last 2 million years of the Spathian.

Fig. 1

(A and B) Corrected and time-calibrated curves of the carbon isotopes of carbonates (A) and oxygen isotopes of conodont apatite (B) from the Nanpanjiang Basin, modified from (1). (C) Original curve of oxygen isotopes from (1). Gray bands represent the first-order trends [the dark band in (C) is from (1)]. Arrows indicate absolute radiometric ages (3). Chan., Changhsingian; Gries., Griesbachian; Diener., Dienerian. A1 to A12, ammonoid zones modified after (3): A1, Proptychites candidus; A2, Clypites/Prionolobus; A3, Kashmirites densistriatus; A4, Flemingites rursiradiatus; A5, Owenites koeneni; A6, Anasibirites multiformis; A7, Tirolitid n. gen. A; A8, Tirolites /Columbites; A9, Procolumbites; A10, Hellenites; A11, Neopopanoceras haugi; A12, Platycuccoceras. C1 to C18, conodont zones [according to figure 2 and table S4 in (1)]: C1, Neogondolella changxingensis; C2, Ng. yini; C3, Ng. meishanensis; C4, Hindeodus changxingensis; C5, H. parvus; C6, Isarcicella staeschei; C7, Is. isarcica; C8, Ng. planata; C9, Neospathodus dieneri; C10, Ns. waageni; C11, Parachirognathus sp.; C12, Ns. pingdingshanensis; C13, Icriospathodus collinsoni; C14, Ns. homeri /Ns. symmetricus; C15, Chiosella timorensis; C16, Ng. regalis; C17, Nicoraella germanica; C18, Ni. kockeli.

Consequently, the one-to-one relationship assumed by the authors between putative "lethally hot" seawater temperatures and the alleged diversity lows (coined "equatorial eclipse") of some organisms is unsustainable: The equally hot temperatures of the Smithian and Spathian should have been equally deleterious to fishes and other marine or terrestrial organisms; however, during the middle and late Spathian, fishes and tetrapods thrived and marine reptiles (ichthyosaurs) diversified.

Likewise, one of the greatest radiations of life in Earth's history occurred in the Middle Ordovician, a time period for which Trotter et al. (2) reported similarly hot temperatures [δ18Oapatite between 18.4 and 19.3 per mil (‰),Vienna Standard Mean Ocean Water; NBS120c renormalized to 22.6‰ for comparison with (1)].

Besides the fact that deep-time δ18Oapatite values should be interpreted in terms of absolute temperatures only with great caution, we propose that the dynamics of the temperature changes is at least as critical for organisms as absolute temperatures. Yet, Sun et al.'s biozonation is not resolved enough to capture such dynamics.

The data points are exclusively taken from table S4 of (1), where the y coordinates are given in centimeters from the base of the corresponding section. These y coordinates are converted to ages using the following procedure: The conodont zonation is slightly modified from figure 2 to take into account all the 18 zones mentioned in table S4. The continuous interval zones are converted into discrete zones, whose positions are constrained by biochronologic and radio-isotopic data (35). For a given section, sets of data points belonging to a particular zone are displayed such that the lowest and the highest points are placed at the bottom and the top of the respective zone, while respecting relative lithological positions for intermediary points. Single data points lacking further information are placed in the middle of the corresponding zone.

Because Parachirognathus is unknown from the Spathian, sample BYC19-1 is here reassigned to the conodont zone C11. The Smithian offset between the Zuodeng sections (Fig. 1, blue and black triangles) on one side and the Bianyang Quarry, Jiarong, and Jinya-Laren sections on the other cannot be explained by a putative water-depth offset (1), because in both areas the data points are based on elements belonging to the same genera. Differences in the intensity of thermal alteration as reflected by the higher color alteration index of the conodonts from Zuodeng (up to 4.5) provide an alternative explanation for this offset.

The late Griesbachian Thermal Maximum is based on a single ZDC24 sample, which falls in the ill-defined Neogondolella carinata zone. Except for this sample, inferred Griesbachian temperatures are all lower than Dienerian ones. Moreover, the stratigraphically closely spaced points of the early Smithian Neospathodus waageni zone (C10, samples ZDC29, ZDC30, ZDC44, ZDC45, and ZDC_B) show an average δ18Oapatite below 19.0‰, even when we exclude the two points corresponding to Platyvillosus, presumably a shallower water genus. Hence, the alleged late Dienerian or early Smithian cooling event is not substantiated by the present data.

In our corrected curve, the late Smithian Thermal Maximum is suggested by a single data point (sample ZDC50, which in fact belongs to the early Spathian zone C13). Furthermore, a sample (JRC51) from the presumed 0.7‰ offset series of the Jiarong section falls within C13 and would indicate lower temperatures. Hence, no thermal maximum can be claimed for this interval.

Fossil-rich beds with common fishes are found both in Chaohu (6) (their earliest fish occurrence in the Spathian, figure 3) and Jurong (7). Based on (7) and (8), these beds fall either in C12 or C13—i.e., they are coeval with the late Smithian Thermal Maximum. Moreover, the distribution of Smithian vertebrate localities is surprisingly at variance between figures 1 and S1 in (1). For the 30°N to 30°S interval, figures 1 and S1C (Smithian) and figure S1A (Late Permian) do not differ substantially (four tetrapod localities each, respectively four and five for fishes, and one and zero for marine reptiles). For these reasons, both the timing and the existence of an equatorial vertebrate "eclipse" are doubtful.

References and Notes

  1. Acknowledgments: This research is supported by the Swiss National Science Foundation project 200020-113554 (to H.B.). A.B. was also supported by the Région Bourgogne and the CNRS Institut National des Sciences de l'Univers Interrvie.
View Abstract


Navigate This Article