The geological record represents the only source of data available for documenting long-term historical patterns of extinction intensity and extinction susceptibility. Such data are critical for testing hypotheses of extinction causality in the modern world as well as in deep time. The study of extinction is relatively new. Prior to 1800, extinctions were not accepted as a feature of the natural environment. Even after extinctions were recognized to have occurred in Earth's geological past, they were deemed to have played a minor role in mediating evolutionary processes until the 1950s. Global extinction events are now recognized as having been a recurring feature of the history of life and to have played an important role in promoting biotic diversification. Interpretation of the geological extinction record is rendered complex as a result of several biasing factors that have to do with the spatial and temporal resolutions at which the data used to study extinctions have been recorded: fluctuations in sediment accumulation rates, the presence of hiatuses in the stratigraphic sections/cores from which fossils are collected, and variation in the volumes of sediments that can be searched for fossils of different ages. The action of these factors conspires to render the temporal and geographic records of fossil occurrences incomplete in many local stratigraphic sections and cores. In some cases, these stratigraphic and sampling uncertainties can be quantified and taken into account in interpretations of that record. However, their effects can never be eliminated entirely. Testing hypotheses of global extinction causality requires acknowledgment of the uncertainties inherent in extinction data, the search for unique predictions of historical patterns of variation or associations that can, in principle, be preserved in the fossil record and tied logically to the operation of specific causal processes, and to adoption of an explicitly comparative approach that establishes the presence of multiple instances of the predicted cause-effect couplets within a well-documented chronostratigraphic context.

The temporal link between mass extinctions and large igneous provinces is well known. Here, we examine this link by focusing on the potential climatic effects of large igneous province eruptions during several extinction crises that show the best correlation with mass volcanism: the Frasnian-Famennian (Late Devonian), Capitanian (Middle Permian), end-Permian, end-Triassic, and Toarcian (Early Jurassic) extinctions. It is clear that there is no direct correlation between total volume of lava and extinction magnitude because there is always sufficient recovery time between individual eruptions to negate any cumulative effect of successive flood basalt eruptions. Instead, the environmental and climatic damage must be attributed to single-pulse gas effusions. It is notable that the best-constrained examples of death-by-volcanism record the main extinction pulse at the onset of (often explosive) volcanism (e.g., the Capitanian, end-Permian, and end-Triassic examples), suggesting that the rapid injection of vast quantities of volcanic gas (CO 2 and SO 2 ) is the trigger for a truly major biotic catastrophe. Warming and marine anoxia feature in many extinction scenarios, indicating that the ability of a large igneous province to induce these proximal killers (from CO 2 emissions and thermogenic greenhouse gases) is the single most important factor governing its lethality. Intriguingly, many voluminous large igneous province eruptions, especially those of the Cretaceous oceanic plateaus, are not associated with significant extinction losses. This suggests that the link between the two phenomena may be controlled by a range of factors, including continental configuration, the latitude, volume, rate, and duration of eruption, its style and setting (continental vs. oceanic), the preexisting climate state, and the resilience of the extant biota to change.

The recent discovery of the direct link between Deccan volcanism and the end-Cretaceous mass extinction also links volcanism to the late Maastrichtian rapid global warming, high environmental stress, and the delayed recovery in the early Danian. In comparison, three decades of research on the Chicxulub impact have failed to account for long-term climatic and environmental changes or prove a coincidence with the mass extinction. A review of Deccan volcanism and the best age estimate for the Chicxulub impact provides a new perspective on the causes for the end-Cretaceous mass extinction and supports an integrated Deccan-Chicxulub scenario. This scenario takes into consideration climate warming and cooling, sea-level changes, erosion, weathering, ocean acidification, high-stress environments with opportunistic species blooms, the mass extinction, and delayed postextinction recovery. The crisis began in C29r (upper CF2 to lower CF1) with rapid global warming of 4 °C in the oceans and 8 °C on land, commonly attributed to Deccan phase 2 eruptions. The Chicxulub impact occurred during this warm event (about 100–150 k.y. before the mass extinction) based on the stratigraphically oldest impact spherule layer in NE Mexico, Texas, and Yucatan crater core Yaxcopoil-1. It likely exacerbated climate warming and may have intensified Deccan eruptions. The reworked spherule layers at the base of the sandstone complex in NE Mexico and Texas were deposited in the upper half of CF1, ~50–80 k.y. before the Cretaceous-Tertiary (K-T) boundary. This sandstone complex, commonly interpreted as impact tsunami deposits of K-T boundary age, was deposited during climate cooling, low sea level, and intensified currents, leading to erosion of nearshore areas (including Chicxulub impact spherules), transport, and redeposition via submarine channels into deeper waters. Renewed climate warming during the last ~50 k.y. of the Maastrichtian correlates with at least four rapid, massive volcanic eruptions known as the longest lava flows on Earth that ended with the mass extinction, probably due to runaway effects. The kill mechanism was likely ocean acidification resulting in the carbonate crisis commonly considered to be the primary cause for four of the five Phanerozoic mass extinctions.

Deccan volcanism, one of Earth's largest flood basalt provinces, erupted ~80% of its total volume (phase 2) during a relatively short time in the uppermost Maastrichtian paleomagnetic chron C29r and ended with the Cretaceous-Tertiary boundary mass extinction. Full biotic recovery in the marine realm was delayed at least 500 k.y. or until after the last Deccan eruptions in C29n (phase 3, 14% of the total Deccan volume). For over 30 yr, the mass extinction has been commonly attributed to the Chicxulub impact, and the delayed recovery remained an enigma. Here, we demonstrate that the two phases of Deccan volcanism can account for both the mass extinction and delayed marine recovery. In India, a direct correlation between Deccan eruptions (phase 2) and the mass extinction reveals that ~50% of the planktic foraminifer species gradually disappeared during volcanic eruptions prior to the first of four lava megaflows, reaching ~1500 km across India, and out to the Bay of Bengal. Another 50% disappeared after the first megaflow, and the mass extinction was complete with the last megaflow. Throughout this interval, blooms of the disaster opportunist Guembelitria cretacea dominate shallow-marine assemblages in coeval intervals from India to the Tethys and the Atlantic Oceans to Texas. Similar high-stress environments dominated by blooms of Guembelitria and/or Globoconusa are observed correlative with Deccan volcanism phase 3 in the early Danian C29n, followed by full biotic recovery after volcanism ended. The mass extinction and high-stress conditions may be explained by the intense Deccan volcanism leading to rapid global warming and cooling in C29r and C29n, enhanced weathering, continental runoff, and ocean acidification, resulting in a carbonate crisis in the marine environment.

We review the extensive record of plant fossils before, at, and after the Cretaceous-Paleogene event horizons, recognizing that key differences between plants and other organisms have important implications for understanding the patterns of environmental change associated with the Cretaceous-Paleogene event. Examples are given of the breadth of prior environmental conditions and ecosystem states to place Cretaceous-Paleogene events in context. Floral change data across the Cretaceous-Paleogene are reviewed with new data from North America and New Zealand. Latest Cretaceous global terrestrial ecology was fire prone and likely to have been adapted to fire. Environmental stress was exacerbated by frequent climate variations, and near-polar vegetation tolerated cold dark winters. Numerous floristic studies across Cretaceous-Paleogene event horizons in North America attest to continent-wide ecological trauma, but elsewhere greater floral turnover is sometimes seen well before the Cretaceous-Paleogene boundary rather than at it. Data from the Teapot Dome site (Wyoming) indicate continued photosynthesis, but during or immediately after the Cretaceous-Paleogene event, growth was restricted sufficiently to curtail normal plant reproductive cycles. After the Cretaceous-Paleogene transition in New Zealand, leaf form appears to have been filtered for leaves adapted to extreme cold, but at other high-southern-latitude sites, as in the Arctic, little change in floral composition is observed. Although lacking high-resolution (millimeter level) stratigraphy and Cretaceous-Paleogene event horizons, gradual floral turnover in India, and survival there of normally environmentally sensitive taxa, suggests that Deccan volcanism was unlikely to have caused the short-term trauma so characteristic elsewhere but may have played a role in driving global environmental change and ecosystem sensitivity prior to and after the Cretaceous-Paleogene boundary.

The Boltysh meteorite impact crater formed in the Ukrainian Shield on the margin of the Tethys Ocean a few thousand years before the Cretaceous-Paleogene boundary and was rapidly filled by a freshwater lake. Sediments filling the lake vary from early lacustrine turbidites and silts to ~300 m of fine silts, organic carbon–rich muds, oil shales, and lamenites that record early Danian terrestrial climate signals at high temporal resolution. Combined carbon isotope and palynological data show that the fine-grained organic carbon–rich lacustrine sediments preserve a uniquely complete and detailed negative carbon isotope excursion in an expanded section of several hundred meters. The position of the carbon isotope excursion in the early Danian stage of the Paleogene period, around 200 k.y. above the Cretaceous-Paleogene boundary, leads us to correlate it to the Dan-C2 carbon isotope excursion recorded in marine sediments of the same age. The more complete Boltysh carbon isotope excursion record indicates a δ 13 C shift of around -3‰, but also a more extended recovery period, strikingly similar in pattern to the highest fidelity carbon isotope excursion records available for the Toarcian and Paleocene-Eocene hyperthermal events. Changes in floral communities through the carbon isotope excursion recorded at Boltysh reflect changing biomes caused by rapidly warming climate, followed by recovery, indicating that this early Danian hyperthermal event had a similar duration to the Toarcian and Paleocene-Eocene events.

We analyzed the plant macro- and mesofossil records deposited in the Paleocene oil shales of the Boltysh crater (Ukraine) in terms of leaf morphology and its implication for reconstruction of the vegetation and paleoecology of the region. During the early Cenozoic, the Boltysh astrobleme formed a geothermal crater lake that accumulated sediments, preserving a record from the Paleocene to the early middle Eocene. These sediments contain fossil leaf fragments of ferns and angiosperms that grew close to the lake. The occurrence of the Mesozoic fern Weichselia reticulata is of importance. This discovery suggests the survival of this Jurassic to Cretaceous fern into the early Paleogene in the refugial geothermal ecosystem of the Boltysh crater area. Our finding is the youngest record of this fern, although it was a widespread and common element of secondary vegetation during the Cretaceous. The local survival of this fern may have been fostered by the unique combination of edaphic environmental factors of the Boltysh hydrothermal area. Other plant fossils include fragments of leaves that represent ferns likely belonging to lineages that diversified in the shadow of angiosperms, as well as remains of the flowering plants Pseudosalix , Sorbus , Comptonia , and ? Myrica leaf morphotypes.

Palynological studies of the late Maastrichtian infratrappean and intertrappean sedimentary beds from various stratigraphic levels in the Deccan continental flood basalt of the Nand-Dongargaon Basin in central India show that infratrappean sediments (Maastrichtian C30n-C29r) are characterized by gymnosperm ( Araucariacites , Classopollis , Cycadopites , Podocarpidites ) and angiosperm ( Cretacaeiporites , Compositoipollenites , Graminidites , Longapertites , Palmaepollenites ) palynomorphs. A distinct floral turnover is observed in intertrappean sediments with the initiation of volcanic activity in the basin. At the lowest stratigraphic level, the earliest floral change is recorded by the appearance of angiosperm-pteridophyte–dominated association ( Aquilapollenites , Azolla , Gabonisporis , Tricolpites , Triporopollenites , and Normapolles group pollen) and the appearance of peridinoid dinoflagellates. At higher stratigraphic levels in C29r, this palynofloral association continues with the appearance of new forms, such as Scabrastephanocolpites spp., Scollardia conferta , and Triporopollenites cracentis . A sharp decline in diversity of titanosauriform-abelisaurid–dominated dinosaurian fauna is also observed with the onset of Deccan volcanic activity. At this time, out of at least eight dinosaur species, only a single taxon of titanosauriform dinosaurs survived the onset of volcanism. This suggests that the floral change and decreased dinosaur diversity are strongly linked with the initiation of Deccan volcanism in C29r in India. The peak eruption of this volcanic phase resulted in the Deccan's largest volume of lava flows, which played a significant role in the global mass extinction at the Cretaceous-Paleogene boundary.

During the last two decades, extensive paleontological research in the main Deccan volcanic province has led to a better understanding of biodiversity close to the Cretaceous-Paleogene boundary. Several infratrappean localities exposed in Jabalpur, Kheda, Balasinor, Rahioli, Dohad, and Bagh in the Narmada Valley (India) preserve one of the most geographically widespread dinosaur nesting sites known in the world. The well-studied intertrappean beds, such as those of Naskal on the southern margin, Asifabad and Nagpur on the eastern margin, Kisalpuri and Mohgaon Kalan on the northeastern margin, and Anjar on the northwestern margin of the main Deccan volcanic province, have yielded Maastrichtian fish ( Igdabatis ) and dinosaur remains and palynofossils ( Aquilapollenites - Gabonisporites - Ariadnaesporites ), either separately or in association, that suggest a Maastrichtian age for these beds. Only two intertrappean sections, Papro on the northern margin and Jhilmili on the northeastern margin of the main Deccan volcanic province, have produced Paleocene fossils. The fossil record from the infratrappean and intertrappean beds demonstrates that the dinosaurs survived the early phase of volcanism, though there was an apparent decline in their diversity, and that freshwater vertebrate fauna was least affected by the initial volcanic activity. The episodic nature of Deccan volcanism may possibly explain the survival of many freshwater and terrestrial communities during the periods of quiescence. In addition, as in the case of the late Maastrichtian sections in eastern Montana, North America, detritus-feeding freshwater vertebrate communities possibly had greater potential for survival than the terrestrial communities dependent on primary productivity. A close examination of the vertebrate faunal distribution across the two stratigraphic intervals (infratrappean and intertrappean) suggests that sampling bias in the infratrappean beds may have also masked the actual diversity of these beds.

The record of dinosaurs over the last 10 m.y. of the Cretaceous, as well as surrounding the Cretaceous-Paleogene boundary, helps to define extinction scenarios. Although Late Cretaceous dinosaur fossils occur on all present-day continents, only in North America do we find a terrestrial vertebrate fossil record spanning the Cretaceous-Paleogene boundary, although promising work may yield comparable records in South America, India, China, and Europe. For the present then, the North American record represents the proxy for our knowledge of dinosaur extinction. Over the last 10 m.y. of the Cretaceous (late Campanian to late Maastrichtian) in the northern part of the western interior of North America, the number of nonavian dinosaur species dropped from 49 to 25, almost a 50% reduction, even though a 16% greater extent of fossil-bearing exposures record the last dinosaurs in the latest Cretaceous in the western interior. Important, but less-well-exposed, nonavian-dinosaur–bearing units suggest this drop occurred around, or at least commenced by, the Campanian-Maastrichtian boundary. These losses began during climatic fluctuations, occurring during and possibly in part caused by the last major regressive cycle of the Cretaceous, which also reduced the expanse of the low coastal plains inhabited by nonavian dinosaurs. The pulse of Deccan Trap emplacement that began some time later in the latest Cretaceous was also likely a major driver of climatic change. As for the dinosaur record near the Cretaceous-Paleogene boundary, even the best-known records from North America remain enigmatic and open to interpretation. Newer studies suggest some decline in at least relative abundance approaching the Cretaceous-Paleogene boundary, but the cause (or causes) for the final extinction (if it was the case) of non-avian dinosaurs remains unresolved, although the Chicxulub impact undoubtedly played a major role.

Currently, it is believed that volcanogenic outgassing of CO 2 during Karoo-Ferrar igneous province eruptions caused prolonged global warming and a multiphased extinction event during the Pliensbachian–Toarcian interval of the Early Jurassic. Warmer water temperatures are thought to have caused a release from the methane hydrate reservoir and global marine anoxia in the Early Toarcian (dubbed the Toarcian oceanic anoxic event). Recently, the timing and geographic extent of these events have been questioned, with emphasis placed on regional conditions in the Tethys Ocean area rather than global controls. Our study compares paleontological and geochemical data from western North America with previously established correlative data in Europe to provide a perspective on the duration, extent, and controlling mechanisms of this Early Jurassic extinction event. Our data indicate that during Pliensbachian–Toarcian time, concurrent with Karoo-Ferrar magmatism: (1) there were six globally correlative intervals of taxonomic diversity decline that constitute evidence of a multiphased extinction event; (2) there was a major disruption in the long-term δ 13 C profile (~3‰–7‰), suggesting volcanogenic CO 2 outgassing as a preeminent factor driving global warming and mass extinction; (3) there was a large negative excursion in the Early Toarcian δ 13 C profile in two successions on Haida Gwaii (formerly known as Queen Charlotte Islands), Canada, suggesting global methane release; and (4) the northeast Panthalassa Ocean did not contain anoxic water during the Early Toarcian. Although the Toarcian oceanic anoxic event does not appear to have globally affected every marine environment in the same manner, it is possible that local anoxic water masses occurred in restricted basins outside the Tethys Ocean area.

In the last decade, major advances have been made in our understanding of the end-Triassic mass extinction, related environmental changes, and volcanism of the Central Atlantic magmatic province. Studies of various fossil groups and synoptic analyses of global diversity document the extinction and subsequent recovery. The concomitant environmental changes are manifested in a series of carbon isotope excursions (CIE), suggesting perturbations in the global carbon cycle. Besides the earlier-recognized initial and main negative anomalies, a more complex picture is emerging with other CIEs, both negative and positive, prior to and following the Triassic-Jurassic boundary. The source of isotopically light carbon remains debated (methane from hydrate dissociation vs. thermogenic methane), but either process is capable of amplifying an initial warming, resulting in runaway greenhouse conditions. Excess CO 2 entering the ocean causes acidification, an effective killing mechanism for heavily calcified marine biota that appears implicated in the reef crisis. The spatial and temporal extent of Central Atlantic magmatic province volcanism is established through a growing data set of radiometric ages. Since the Central Atlantic magmatic province is one of the largest Phanerozoic large igneous provinces, volcanic CO 2 -driven warming is plausible as a key factor in the chain of Triassic-Jurassic boundary events. Greenhouse warming may have been punctuated by short-term cooling episodes due to H 2 S emission and production of sulfate aerosols, a process more difficult to trace in the stratigraphic record. Taken together, recently generated data significantly increase the support for Central Atlantic magmatic province volcanism as a viable trigger for the environmental and biotic changes around the Triassic-Jurassic boundary.

Eight climatic events can be distinguished in the Triassic–Jurassic (ca. 201 Ma) continental strata of Poland. These events are distinguished by kaolinite/illite ratio, chemical index of alteration (CIA), color of sediments, and palynomorphs. The first transition to wetter climate, evidenced by a shift from smectite- to kaolinite-dominated mudrocks, coincides with the earlier (“precursor”) Rhaetian negative δ 13 C org excursion, which means that the beginning of climate perturbations predates the oldest known Central Atlantic magmatic province flood basalts by some 100–200 k.y. The later global, late Rhaetian “initial” negative δ 13 C org excursion is divided into two subpeaks, each corresponding to hot and humid events, separated by a cooler and drier event. The upper subpeak is also associated with perturbation of the osmium isotope system (attributed to volcanic fallout), and darkened miospores, pointing to acid rains. Between the “initial” excursion and the Triassic-Jurassic boundary interval, five climatic fluctuations are inferred from the changing kaolinite/illite ratio, the last two of which are also associated with an Os isotope perturbation, polycyclic aromatic hydrocarbon (PAH) occurrences, a “spore peak,” and darkened miospores. A series of periodic atmospheric loading events by CO 2 , CH 4 , or alternatively by SO 2 , sulfate aerosols, and toxic compounds, is inferred to have caused this series of rapid climatic reversals and resulting extinction of many less-adapted forms. Just above the palynofloral extinction level, appearance of new forms commenced Jurassic palynofloral recovery. Tetrapod evolution events in the end-Triassic–earliest Jurassic were related to the extinction of the Pseudosuchia, Dicynodontia, Capitosauroidea, Plagiosaroidea, and Rhynchosauria, while appearance of highly diversified tetrapod ichnofauna in the earliest Jurassic strata indicates a rapid recovery and refill of ecological niches by dinosaurs.

The Hirnantian mass extinction is recognized as the first of the “big three” extinctions and, along with the end-Permian and end-Cretaceous events, is the result of an acceleration in biotic extinctions concomitant with a rise in originations. The Hirnantian mass extinction is characterized by high taxonomic impact and within-community extinctions. The Hirnantian mass extinction is also unusual in that (1) it is associated with glaciation, but there is little evidence elsewhere in the younger Phanerozoic that glaciations have been a cause of mass extinction, and (2) there is limited understanding of how glaciation could directly cause mass extinction, particularly in the marine realm. In this review, we argue that coordinated extinctions occurred at the onset and termination of glaciation and were due to climatically induced changes in relative sea level, ocean redox stratification, and sea-surface temperature gradients. These earth system changes resulted in a reduction in prospective niche space, both in the water column and on the seafloor, which in turn led to increased competition and selection pressures, resulting in extinctions where the carrying capacities of particular ecological niches were exceeded. The long-term ventilation of the oceans broke the link between glaciation and mass extinction.

There are at least a dozen Phanerozoic continental flood basalts and oceanic plateaus (large igneous provinces) that roughly obey a “rule of two times one million” (volume of extruded lava of one million cubic kilometers and duration of volcanic activity ~1 m.y.). The correlation between large igneous province ages and mass extinctions (and oceanic anoxia events) is excellent, but quantitative scenarios are still wanting. We hypothesize that the temporal sequences of extrusions determine the severity of extinction: Volcanic pulses separated by thousands of years allow the ocean-atmosphere system time to recover, whereas large volcanic pulses occurring in a shorter sequence may result in a runaway effect and cause a mass extinction. Detailed flow-by-flow magnetic stratigraphies of thick sections have identified directional groups (sequences of superimposed lava flows with the same paleomagnetic direction that cooled in a time too short to record secular variation). With help of this simple tool, many single eruptive events with a volume larger than 1000 km 3 , some in excess of 10,000 km 3 , emplaced in possibly less than a decade, have been identified. We review this evidence, mainly for the following flood basalt provinces: Columbia, Brito-Arctic, Deccan, Karoo-Ferrar, Central Atlantic magmatic province, and Siberian Traps. Large igneous province volcanism occurs in a highly discontinuous way, on embedded time scales, on the order of 10 m.y., 1 m.y., 100 k.y., and 10 yr. This provides constraints for models of plume-lithosphere interaction and magma production. A next step is to model the consequences of massive injection of gases that can be derived from these time and duration estimates. Early attempts are reviewed in a companion paper in this volume.

Continental flood basalt provinces are the subaerial expression of large igneous province volcanism. The emplacement of a continental flood basalt is an exceptional volcanic event in the geological history of our planet with the potential to directly impact Earth's atmosphere and environment. Large igneous province volcanism appears to have occurred episodically every 10–30 m.y. through most of Earth history. Most continental flood basalt provinces appear to have formed within 1–3 m.y., and within this period, one or more pulses of great magma production and lava eruption took place. These pulses may have lasted from 1 m.y. to as little as a few hundred thousand years. Within these pulses, tens to hundreds of volumetrically large eruptions took place, each producing 10 3 –10 4 km 3 of predominantly p3hoehoe lava and releasing unprecedented amounts of volcanic gases and ash into the atmosphere. The majority of magmatic gas species released had the potential to alter climate and/or atmospheric composition, in particular during violent explosive phases at the eruptive vents when volcanic gases were lofted into the stratosphere. Aside from the direct release of magmatic gases, magma-sediment interactions featured in some continental flood basalt provinces could have released additional carbon, sulfur, and halogen-bearing species into the atmosphere. Despite their potential importance, given the different nature of the country rock associated with each continental flood basalt province, it is difficult to make generalizations about these emissions from one province to another. The coincidence of continental flood basalt volcanism with periods of major biotic change is well substantiated, but the actual mechanisms by which the volcanic gases might have perturbed the environment to this extent are currently not well understood, and have been little studied by means of atmospheric modeling. We summarize current, albeit rudimentary, knowledge of continental flood basalt eruption source and emplacement characteristics to define a set of eruption source parameters in terms of magmatic gases that could be used as inputs for Earth system modeling studies. We identify our limited knowledge of the number and length of non-eruptive phases (hiatuses) during continental flood basalt volcanism as a key unknown parameter critical for better constraining the severity and duration of any potential environmental effects caused by continental flood basalt eruptions.

Climatic and environmental changes are now widely recognized as the main cause of mass extinctions. Global warming that immediately preceded the Cretaceous-Tertiary boundary is regarded as a consequence of CO 2 released during the main phase of Deccan Trap emplacement. Modeling has shown that such global warming cannot be explained by the continuous release of volcanic carbon dioxide. In the present paper, we use a biogeochemical model, coupled to a climate model, to further our understanding of climate changes caused by continental flood basalts. The response of the global climate–carbon-cycle system to sulfur dioxide (SO 2 ) and carbon dioxide (CO 2 ) emissions is investigated, assuming a degassing history consisting of a series of evenly spaced pulses. We find that CO 2 -related warming is enhanced when large-scale SO 2 injections are added. According to our model, we observe that the succession of drastic cooling events induced by sulfate aerosols decreases the efficiency of silicate weathering and destabilizes the carbon cycle during the full time span of trap emplacement. In the case of the Deccan Traps, these transient dis-equilibria lead to a 25% increase in p CO 2 and ensuing warming. The environmental consequences of emplacement of large igneous provinces appear to be even more complex: A SO 2 -related climate feedback may have enhanced the long-term warming due to CO 2 emissions.

Environmental changes linked to Deccan volcanism are still poorly known. A major limitation resides in the paucity of direct Deccan volcanism markers and in the geologically short interval where both impact and volcanism occurred, making it hard to evaluate their contributions to the mass extinction. We investigated the low-magnetic-susceptibility interval just below the iridium-rich layer of the Bidart (France) section, which was recently hypothesized to be the result of paleoenvironmental perturbations linked to paroxysmal Deccan phase 2. Results show a drastic decrease of detrital magnetite and presence of scarce akaganeite, a hypothesized reaction product formed in the aerosols derived from reaction of a volcanic plume with water and oxygen in the high atmosphere. A weathering model of the consequences of acidic rains on a continental regolith reveals nearly complete magnetite dissolution after ~31,000 yr, which is consistent with our magnetic data and falls within the duration of the Deccan phase 2. These results highlight the nature and importance of the Deccan-related environmental changes leading up to the end- Cretaceous mass extinction.