| 06 Nov 2025
Carbon export and burial pathways driven by a low-latitude arc-continent collision
Abstract. Chemical weathering of silicate rocks of low-latitude arc–continent collisions has been hypothesized as a driver of global cooling since the Neogene. In low-latitude regions, monsoon and tropical cyclone precipitation also drive intense physical erosion that contribute to terrestrial carbon export and nutrient-stimulated marine productivity. Despite this, the role of physical weathering on carbon sequestration has largely been overlooked. To address this gap, we analyse late Miocene–early Pleistocene sedimentary and geochemical records from the Taiwan Western Foreland Basin and time-equivalent records from the northern South China Sea.
Along the continental slope, organic carbon is largely marine in origin, and its accumulation controlled by long-term sea-level fall and glaciation. In contrast, on the continental rise, organic carbon burial is controlled by high sedimentation rates related to Taiwan’s uplift and erosion (since ~5.4 Ma). Despite increased terrestrial erosion of Taiwan, the organic material remains mainly marine in origin, suggesting that primary production was enhanced by nutrient exported from Taiwan. Marine organic matter along Taiwan’s shore was subsequently remobilized by turbidity currents through submarine canyon systems and accumulating on the continental rise of Eurasia. The onset of Northern Hemisphere Glaciation (~3 Ma) and subsequent intensification of the East Asian Summer Monsoon and persistent tropical cyclone activity all further amplified nutrient export across the basin, further stimulating marine primary production.
Our findings demonstrate that arc–continent collision influences carbon sequestration through two pathways: (1) direct burial of terrestrial organic matter and (2) nutrient-fuelled marine productivity and burial. This work establishes a direct link between the erosion of an arc-continent collision and long-term carbon burial in adjacent ocean basins.
This is an intriguing study that looks at the impact of an arc-continent collision on regional carbon budgets and particularly the export of terrestrial organic material from continental southern China and from Taiwan. Taiwan is the world type example of arc-continent collision. The study compares the flux of organic carbon from the land with the amount of organic material of marine origin which is buried in the deep South China Sea. The paper reaches an interesting conclusion that much of the buried organic material is of marine origin and that greater rates of carbon burial are achieved as a result of the delivery of nutrients from the land into the coastal ocean increasing productivity which is then buried by the significant flux of clastic sediment, especially from Taiwan. In general the work is useful in noting the potential importance of arc-continent collision zones as being places with significant amounts of organic carbon may be sequestered although any material buried like this in the South China Sea is liable to be recycled in the relatively near geological future because of the Southeast migration of Taiwan through time, meaning that the storage is not long lived. I was only partly convinced that two neighbouring drill sites really had quite such distinct provenance with one being almost exclusively related to sediment delivered by the Pearl River and the other from Taiwan, but I think even if we take this with a pinch of salt that the overall conclusions are interesting in showing the coupling between the tectonics and the burial of organic carbon and the potential influence of high amplitude climatic variations modulated through the Asian monsoon and linked to Northern hemisphere glaciation. There was no estimate of the total flux it so it’s not possible to see whether this region might have a global impact although it seems unlikely based on recent work. I had a number of smaller questions which I left below and which the authors are encouraged to respond to.
Line 13 - In low-latitude regions, monsoon - Monsoon is more of a mid latitude phenomenon
Line 15 - physical weathering - You mean erosion?
Line 18 - accumulation controlled by long-term sea-level fall and glaciation. – You mean because sediment is stored on the shelf during sea level highstands? Is Taiwan glaciated?
Line 22 - Marine organic matter along Taiwan’s shore was subsequently remobilized - How do you know the organic matter is not produced further from the coast and just falls out through the water column to be preserved in deeper water?
LIne 24 - intensification of the East Asian Summer Monsoon - Yes but only during interglacial times
LIne 29 - erosion of an arc-continent collision - Yes but anywhere with fast sedimentation has the potential to busy more organic material, See Pleistocene of the Bengal Fan.
Clift, P.D., Jonell, T.N., Du, Y., Bornholdt, T., 2024. The impact of Himalayan-Tibetan erosion on silicate weathering and organic carbon burial. Chem. Geol., 656: 122106. doi:10.1016/j.chemgeo.2024.122106.
Line 34 - during the Neogene (Clift and Jonell, 2021),- Yes but also see more recent work by same authors. Clift et al. (2024)
Line 47 - that were considerably more intense and frequent than at present - But we are in an interglacial, Why would typhoons be stronger during previous interglacials?
Line 70 - Since its emergence in the early Pliocene, Taiwan - Emergence in this location. The collision zone has been migrating SW through time so there would have been an emergent island before then but located further to the NE.
Line 86 - late Pliocene =upper Pliocene
Line 92 - The targeted time interval (~6.27–1.95 Ma – Why was this time chosen?
Line 92 - spans the initiation of Eurasian-Philippine plate collision – No. Only in this place Its not the first time of collision
Figure 2 - Dark writing against dark background is hard to read. Why not have a blue shaded bathymetry?
Line 113 - Da’an River – If you are going to name geographic features you need to show them on a map. Likewise, Tachia River, Houlong River
Line 116 – ODP - Explain abbreviation.
Line 120 - ODP Sites 1146 and 1148 cores – Why do you think these contain Taiwan material and not just from southern China?
Line 131 - achieving a precision of better than 0.3% (REFS) - REFS??
Line 155 - Thirty-three oriented palaeomagnetic core specimens (25-mm diameter) were collected at ~3.5 m intervals – But from where?
Line 165 - serve as proxies for physical erosion - this is not justified. The MAR at a single site cannot be considered proxy for the amount of physically erosion in the source areas. There are lots of factors not least sediment transport in the deep water and accommodation space issues on the continental shelf that mean that these are undoubtedly unrepresentative.
Line 170 - sourced from the South China – you need to be more specific. Where is the South China data from? If you use material from the Pearl River then that is not acceptable because the river is highly anthropogenically disrupted and not representative of the flux in the past.
Line 178- Sedimentary TOC content provides a measure of organic carbon accumulation through time – yes, but as for the clastic mass accumulation it is not acceptable to use the rate of organic carbon MAR at a single site as approximately for the total flux
Line 189 - Hematite typically forms through iron oxidation under arid climates – that’s not completely correct. It may be indicative of a seasonal climate with a dry season.
LIne 218 - Stratal ages from ODP Site 1148 (Clift, 2006) are constrained using biostratigraphic ages of benthic foraminifera (Wang et al., 2000a) – the age model is from Wang et al.
LIne 221 the those - Delete “the”
Line 227 - Chinshui shale - Chinshui Shale
Figure 3 – why is the label “covered interval” so enormous compared to the rest of the text?
Line 239 - after which both increase, with a maximum MAR of 3.5 cm—that only tells you that the accumulation increased in this particular location but it doesn’t tell you anything about the discharge from the river. Sedimentation rates at a single location being governed by a variety of factors. For example the increase may simply refer to moderate change in the peed and location of a contour current.
Line 275 - which channel sediment - which channels sediment
Lines 285 - sediment records diverge despite their spatial proximity - I;m not sure I agree. The records parallel each other. They are not identical but there are plenty of local reasons that MAR would not be identical and even TOC could be affected by local dilution by clastic or biogenic sedimentation.
Line 290 - Pearl River sediment discharge is controlled by long-term sea-level changes and East Asian Monsoon variability (e.g., Liu et al., 2016) - Sea level does not control the discharge. Only where the sediment is deposited. The monsoon impact on Pearl River discharge was noted long before Liu et al. (2016), e.g., Clift (2006)
LIne 291 - shallow Taiwan Strait – What does the Taiwan Strait have to do with the Pearl River? The modern/Holocene flux is to the west. As noted by
Liu, J. P., Z. Xue, K. Ross, H. Wang, J., Z. S. Yang, A. C. Li, et al. (2009), Fate of sediments delivered to the sea by Asian large rivers: Long-distance transport and formation of remote alongshore clinothems, Sed. Record, 7(4), 4-9.
At the LGM flux was to the deep SCS via the Canyon.
Figure 5 - Why do you show Taiwan getting smaller in the past? The collision is migrating to the SW but there is no reason to believe the collisional orogen was smaller in the past
Line 302 - major element data suggest a mixture - Major element data is not a good provenance tool. Its mostly affected by weathering. Also Isotopic (87Sr/86Sr, εNd) methods are not suitable either because the sources are all from South China and Sr is affected by weathering which changes with climate.
Line 304 - clay mineral records - Specifically what do you mean?
LIne 306 - supported by rare-earth element studies – these are not reliable because of the common sources to the these areas and using modern Pearl River sediment is a bad fingerprint for the past because of anthropogenic disruption of the catchment.
LIne 319 - Increase in magnetic minerals since ~6.27 Ma reflects increased sediment input from Eurasia - How do you know this is a sign of Eurasian input?
LIne 328 - a shift towards long-term drying – Started much earlier ~10 Ma in southern China region. See Clift et al. (2014), Clift (2025)
LIne 330 - reflecting enhanced sediment export – You don’t know that. Sedimentation rate at a single site can be controlled by a wide range of processes.
LIne 335 - does not appear to track long-term the monsoon drying – Why do you say that?
LIne 337 - as it emerged from the Pacific Ocean - Taiwan (or the collision orogeny) has been progressively migrating to the SE. Any increase is because the collision is approaching the drill site.
LIne 344 - tracks the orogenic evolution of Taiwan at both ODP sites - How does it do that? You just argued that Site 1146 is mostly derived from the Pearl River
LIne 355 - fine-grained material enriched TOC - fine-grained material enriched in TOC?
LIne 369 - reflect increased Eurasian clastic influx under conditions of long-term sea-level fall-How do you know this isn’t related to stronger erosion driven by a stronger EASM?
Figure 6 - Text is too small to read
LIne 382 - The onset of orogenesis in Taiwan at ~5.5 Ma - The onset of deposition by sediment eroded from the Taiwan Mountains at ~5.5 Ma
LIne 384 indicates significant export of terrestrial sediment - No. As noted about MAR at a single site could be controlled by many processes, such as currents in the SCS.
LIne 387 enhanced by intensified sediment export - Or better preserved by rapid burial
LIne 405 - confined to proximal coastal environments - Or oxidized before deposition which seems unlikely compared to the Pearl River
LIne 417 rapidly declining offshore due to swift uptake - And dilution?
LIne 424 - makes the northern SCS a depocenter for organic carbon burial - Not too much based on this regional synthesis
Clift, P. D., T. N. Jonell, Y. Du, and T. Bornholdt (2024), The impact of Himalayan-Tibetan erosion on silicate weathering and organic carbon burial, Chem. Geol., 656, 122106, doi:10.1016/j.chemgeo.2024.122106.
LIne 430 - nutrient inputs from Taiwan and the Yangtze River, - and presumably the rivers of Southeast China as well
LIne 445 - Cholan Fm - Don’t abbreviate Fm
LIne 450 - The enhanced in export of coarser-grained sediment - The enhanced export of coarser-grained sediment
LIne 453 - would in turn increase sediment supply to the South China Sea - 453 would in turn have increased sediment supply to the South China Sea - I’m not sure I understand the logic. Also, you were abbreviating South China Sea earlier in the paper.
LIne 459 - coupled with intensified monsoon – are they really intensified? I thought it was mostly during the interglacial periods when the monsoon is strong.
LIne 467- South China Sea (SCS) – you’ve already defined this abbreviation
LIne 480 the East Asian Summer Monsoon - you were abbreviating that before
LIne 488- rather than to local tectonics - do you mean in Taiwan or are you talking about near the drill sites in the South China Sea?