Digitalized Earth's most severe sea-level regression and extinction

~ 212 words 11 Text: ~ 6066 words 12

but also in Canadian Arctica [1] , Oman [2] , Canadian Rockies [3] , Norway [3] , North America [3] all 25 over the world. New sections and digitalized sea-level regression demonstrate that the period 26 of extinction falls within the hiatus, a break in deposition between the uppermost Permian 27 carbonate strata and the clasts reworked from Permian platforms, representing a duration of 28 sea-level drop 354 m. Carbonate clasts, Siberian Traps volcanism, global warming, anoxia, 29 and ocean acidification are all post-extinction geological events. Why did the extinction occur 30 during the falling stage? We will never know because we can't study a hiatus unrepresented 31 by strata unless we associate the extinction with the sea-level drop. 32   33   34   35   36  37  38  39  40  41  42  43  44  45  46  47  48  49  5 Less than 1 centimeter in thickness of the hiatus-equivalent bed is inferred from the general 110 sediment accumulation rates of 0.36-0.17cm/ka [14] . It would not be available to find out a 111 complete PTB sequence [22] with higher resolution. All the PTB sections are represented either 112 by a centimeter-thick extinction bed in distal basin center areas or by a hiatus in a 113 shallow-water shelf environment.   A systematic decreasing trend in δ 13 C value is suggested [23] from basin margin to basin 122 center or from stratigraphic position to EPME both below and above. Sediments in the distal 123 basin center areas preserved the minimum δ 13 C values, where few Permian records were 124 missing. The minimum δ 13 C values were maintained at a depth of ＞354 m, during marine 125 regression interval between 251.939 and 251.925 Ma, with the lowest sea-level occurring at 126 251.932Ma. The negative excursion of δ 13 C isn't considered related to EPME due to reduced 127 carbon-isotope shifts with decreasing stratigraphic distances to EPME [23] . 128 Because the average δ 13 C values of CO 2 released suggest a source of large quantities of 129 carbonate-derived carbon [10] , removing at least 354 m-thick carbonate rocks during the end 130 Permian regression period best explained the negative excursion of δ 13 C. A delayed source 131 with small quantities of carbon from sill intrusions [24] or thermal erosion [25] showing a 132 homogenized δ 13 C trend [26] failed to explain the globally recognized excursion of δ 13 C [27] . A 133 negative shift in δ 13 C occurred before the onset of volcanism, global warming, or negative 134 excursion of δ 18 O has been verified [8] . 135

Initiation of volcanism at 251.928 Ma 136
Volcano eruptions occurred at the end of the low-stand interval at 251.928 Ma, as indicated 137 by the Xiadawu section[ Fig. 2D] [28] , containing a 752-m-thick volcano with two complete 138 breccia-tuff-basalt cycles, which witnessed an uplift-erosion-basalt process, following the rule 139 of Campbell's plume theory [29] . Volcano  Above geological events based on the incomplete PTB sections occurred very close to the 150 extinction event but never preceded it. A fluctuation of temperature ~3-7°C [8,9,16] is not 151 severe enough to cause the largest bio-crises. Moreover, the regression-induced 152 conglomerates are consolidated by calcareous (85%) and ferric (15%) oxides, suggesting an 153 un-happening of ocean acidification or anoxia during the extinction interval. All of the above 154 direct towards that a regression-extinction mechanism might be an un-neglected reasonable 155 cause. 156 3. Interpretation for EPME occurred during the falling period 7 Hallam [3] didn't think Newell's theory tenable for interpreting EPME. But he provides a series 165 of crucial evidence supporting the end-Permian regression. "all sections, the latest Permian is 166 missing, and lower Triassic strata rest unconformably on middle Permian or older 167 strata…… the oldest Triassic rocks has been lacking." Correlation between these sections in 168 North America, Norway, and Canadian Rockies [3] and those in Canadian Arctica [ [3,32] . Both involved reworking and redeposition during the falling 175 period, the carbonate platforms bearing the latest Permian ammonoid markers, and the 176 deep-slope shales containing the Changhsingian Age [ Figure 2].  Changma River (left) and Xiadawu (right). Numbered lines ( a to d ) represent a period for sea-level drop 181 (extinction hiatus), sea-level low-stand, and sea-level rise, as well as volcanic eruptions, respectively, as shown in

253
Xiadawu and (7) Yama in the middle study area.

254
The study area [ Figure 3C] is located at the northeastern margin, extending along the south Palaeofusulina-Reichelinia-collaniella assemblage, which can correlate with Beds 1-2 271 immediately underlying the PTB in Chongyang section [50] .  Bed 11 contains post-extinction organisms widely distributed in Changma River [31,42] and 311 24km east of Maduo county [45,51]  In ascending order from east to west, the section contains lower-slope faces of silty 318 slate-dominated turbidites representing Early-Triassic deposits over 3000 m. Taupe

343
Numbers lines (a to c) mean a period of sea-level drop (extinction hiatus), sea-level low-stand, and sea-level rise, 344 respectively, as shown in Fig.1

The PTB sequences show a coarsening-upward character 377
The Xiadawu and North Changma River sections display sequence changes from marine 378 shales to coastal conglomerates. Xiadawu section witnessed an uplift-erosion-basalt process, 379 following the rule of Campbell's plume theory [29] .

The sediment surge over tens of times, as indicated by a sharp increase in 392
87 Sr/ 86 Sr[ Figure 6] 393

27.36
The considerable thickness of accumulation and high deposition rate [ Table 1]  A sharp increase in ratios of 87 Sr/ 86 Sr [34,35,56,57] , intensified weathering [5,35] , climate change 419 from humid to arid [18,35] , continentalization, desertification, wildfires [13] , or drought enhanced 420 terrestrial input [37] and anomalous marine sediment fluxes [36,37] . Restriction events within the 421 Paleotethys ocean [59] , Tibetan-sized plateaux [60] in supercontinent Pangaea, and an estimated 422 change of sea-level 400-650 m [61] , as well as a minimum shallow-water area of 13% [20] during 423 the latest Permian, are all marine regression signals. syn-depositional bedding [ Figure 8.6], showing rootless features and no tectonic characteristics. Consequently, they couldn't be the "nappes" formed during the orogenesis 437 stage [62] . So-called middle Permian palaeoseamounts [38] are, in fact, of km-sized megabreccias 438 because the shallow-water carbonate seamounts couldn't grow in basin center areas with a 439 depth greater than 1500 m [38] , and the benthic organisms on the palaeo-seamounts couldn't 440 live on an aphotic sea-floor. It is speculated that the palaeo-seamounts may have been 441 reworked from carbonate platforms during the end Permian regression (should be of earliest 442 Triassic Age) due to their similar bio-and lithostratigraphy of the strata. 443

The synchrony between the regression and the onset of carbonate collapse 444
Global sea level controls the growth of carbonate platforms [67] . The sea-level rise created 445 more accommodation spaces, and carbonate platforms grow upward and keep up with the 446 level. When sea level drops, carbonate platforms are exposed to erosion until the platforms'

Debates focusing on the end Permian regression 457
Many authors [3,4,6,8,9,58] deny the end-Permian regression due to its continuous PTB 458 sequences or ascribe the end-Permian hiatus to submarine erosion [ Fig. 7below], even think 459 "latest Permian……coincides with a major transgression" [3] . 460 The so-called continuous sequences are, in fact, of freshwater deposits rather than marine 461 sediments. The Wujiapingian-Changhsingian boundary of late Permian recognized a global 462 regression of sea level [3,23] . The shallow-water sedimentary structure (equivalent to Bed 8 of 463 the Zhihela section) of the storm deposits suggests a water depth of 30~50 m. This paper also 464 admits this late Permian regression. But it didn't result in the widely distributed bioclasts 465 because of some bearing the latest Permian fossils [Fig. 5A, Fig. 8.3,8.4]. Moreover, the thickness of the earliest Triassic (Griesbachian) turbidite sequence ranges up to more than 1 467 km in the basins of Chinese Bayan Har (North Changma River section, 1087m) and Canadian 468 Sverdrup [23] . How could the late Permian basins with 50-meter deep in distal center areas 469 accommodate up to 1 km thick sediment? 470 Intensified weathering [5,35] , climate change from humid to arid [18,35] , enhanced terrestrial 471 input [37] , anomalous marine sediment fluxes [36,37] wouldn't be the results of the hot climate. 472 Can the hot weather produce giant carbonate blocks ranging from hundred meters to 473 kilometers [ Fig. 5D, Fig. 8.1, 8.2, 8.5, 8.6], which were reworked from Permian platforms and 474 embedded in a matrix of earliest Triassic slates, siltstones, and sandstones? Abraded bioclasts 475 that occurred in strata of later Griesbachian [3] may result from re-delivery and redeposition 476 from the earliest Triassic regression unit (equivalent to Bed 10 of the Zhihela section).  The last 15-Myr Permian shallow-water platforms were eroded not only in Chinese Bayan 533 Har but also in Canadian Arctica [ Beds 5-9 of the Zhihela section, whereas Beds 2-4 or previous strata eroded through 536 submarine erosion[ Table 2]. Although this magnitude calibration is speculative due to the 537 large ranges in thickness of the fusulinid-basalt zone, it is consistent with the regional doming 538 of 364 m indicated by the Xiadawu section [ Fig.2D] from deep ocean shale Beds 5-7 to 539 coastal conglomerate Bed 8. The latest papers have verified this magnitude calibration, "the 540 [70] and 541 "restriction events within the Paleotethys ocean" [59] .  Why did the extinction occur during the falling stage? We will never know because we 559

amplitude of end-Permian sea level drop is at least 190 m in Sichuan Basin"
can't study a hiatus unrepresented by strata unless we associate the extinction with the 560 sea-level drop, which reduced the shallow-water habitat regions to a minimum. The severe 561 marine regression and the resultant drought also killed land plants and animals. The study of 562 the past helps to protect the human future. The global threat to human security could be 563 habitat loss rather than global warming, although global warming could also cause habitat 564 loss. 565 Figure 1 Digitalized geological events (numbered black circles) and the corresponding relative magnitude (green curve) across the PTB interval: (1) extinction event, (2) minimum values of δ13C, (3) initiation of Siberian Traps volcanism, (4-6) onset of warming, anoxia, and ocean acidi cation, etc.; Numbered lines ( a to d ) represent a ~3 kyr-long period for sea-level drop (extinction hiatus), ~8 kyr sea-level low-stand, sea-level rise ~0.6 Myr, and volcano eruptions ~10 Myr, respectively.   Lake in the center basin areas, volcanos of (6) Xiadawu and (7) Yama in the middle study area.Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors. Correlation of series PTB sections 1, 2, 3, 5 along with the pro le A-B ( Figure 4C) from shelf to basin center areas and a reconstructed Permian sequence representing the missing strata of section 3 are displayed. Numbers lines (a to c) mean a period of sea-level drop (extinction hiatus), sea-level low-stand, and sea-level rise, respectively, as shown in Fig.1. A. a photo of giant carbonate collapse (size:13m×4m) bearing the uppermost Permian corals, ammonites, and brachiopods embedded in a matrix of early Triassic siltstones or slates; C. wave-polished conglomerates reworked from a carbonate platform, are enclosed in black slates and siltstones, containing one basalt and two fusulinid-bearing carbonate boulders. D. two kilometer-sized megabreccias (B, E) bearing mid-Permian fusulinid assemblages on google maps 35°03′18.97″N, 97°57′54.22″E with an angle of view altitude 5.7km. Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.

Figure 6
The apex of the long-term 87Sr/86Sr curve[34, 54] (left) shows a 10-Myr lag after the peak of the average deposition rate curve (right) across the PTB. Left: the 1-Ma step 87Sr/86Sr curve with peak values occurred at the Olenekion-Anisian boundary, a 10-Myr-long lag after the PTB. Right: green curve shows the average deposition rate, the blond bar showing the extinction interval when the sea-level dropped sharply, and the resultant deposition rate increased dramatically.