The Southern Ocean, covering approximately one-third of the world's oceans, plays a vital role in global climate regulation. The biological carbon pump (BCP), encompassing ecosystem processes that absorb CO2 through photosynthesis and transport organic carbon to depth, significantly influences planetary-scale carbon sequestration. The Southern Ocean's contribution to the global biological pump is substantial (2-3 Pg C y⁻¹), representing a significant fraction of the global total (5-13 Pg C y⁻¹). However, the efficiency of the BCP varies across different Southern Ocean biomes. Predicting the BCP's response to climate change necessitates a thorough understanding of the ecological mechanisms driving its variability. Salps, gelatinous zooplankton grazers, are widely distributed in the Southern Ocean. Ocean warming associated with climate change may facilitate salp range expansion towards the Antarctic continent, potentially displacing krill. This southward expansion could significantly impact both the food web and the BCP. Therefore, a quantitative understanding of salps' effects on the marine carbon cycle is crucial for improving biogeochemical models. High particulate organic carbon (POC) export linked to salp blooms has been observed in various environments, including the Southern Ocean's Lazarev Sea and Western Antarctic Peninsula, as well as the Sargasso Sea, the northern Arabian Sea, and the California Current. The combination of high ingestion rates, extensive bloom formation, the ability to feed on a wide range of prey sizes, and the fast sinking of large fecal pellets (FPs) contribute to significant ocean carbon sequestration during salp blooms. While global-scale modeling efforts suggest substantial salp contributions to the BCP, further mechanistic studies are needed to understand the conditions leading to salp blooms and the temporal dynamics of associated export. Existing studies quantifying the impact of salp blooms on carbon export budgets are limited. It remains unclear whether salps significantly enhance export in situations where this production would not otherwise be exported, or if they primarily proliferate under conditions that independently lead to high export. Disentangling these factors requires simultaneous measurements of phytoplankton growth, micro- and mesozooplankton grazing, salp standing stocks, salp grazing or FP production, and carbon export, integrated over appropriate time scales (hours to days) in locations with diverse environmental conditions. This type of comprehensive study is currently lacking. Furthermore, salp grazing can alter both the composition of exported prey and the remaining assemblage, a process that is expected but challenging to demonstrate. Salps, being filter feeders, are largely unselective in their prey choice except for size, consuming submicron-sized cells, pico-, nano-, and microplankton. Consequently, their grazing can influence remineralization by exporting smaller cells that would typically remain in the euphotic zone. The episodic nature of salp blooms has historically hindered controlled investigations. This study leveraged successful predictions of salp bloom timing and location in the Southwest Pacific sector of the Southern Ocean to conduct the first whole plankton food web process study quantifying the impact of salp blooms on BCP efficiency.

Previous research has highlighted the significant role of salps in carbon export. Studies in the Lazarev Sea and Western Antarctic Peninsula have demonstrated high POC export associated with salp blooms. Similar observations have been made in other regions, such as the Sargasso Sea, the northern Arabian Sea, and the California Current. These studies have demonstrated the potential of salps to enhance carbon sequestration due to their high ingestion rates, extensive bloom formation, broad prey size spectrum, and the rapid sinking of their large fecal pellets. However, many of these studies lacked the comprehensive multi-faceted approach used in this study. While modeling efforts have suggested substantial salp contributions to the global BCP, there's a dearth of mechanistic studies examining the conditions that favor salp blooms and the temporal dynamics of their impact on carbon export. A key gap addressed by this research is the lack of studies simultaneously measuring phytoplankton dynamics, zooplankton grazing (both micro and meso), salp population dynamics, fecal pellet production, and carbon export using methods that provide accurate temporal integration.

The SalpPOOP (Salp Particle export and Ocean Production) study was conducted in Southern Ocean waters near the Chatham Rise, where Subtropical (ST) and Subantarctic (SA) surface water masses meet. The study area was chosen based on previous research indicating the ecological importance of salps in this region, supported by fisheries trawl data and predator diet analyses. Salp blooms are likely periodic events in this area, acting as a recurring food source for various deep-water fish, sea perch, warehou, sea lions, and Bryde's whales. The study aimed to (i) determine the effect of salp grazing on the microplankton community, (ii) quantify the difference in export flux under bloom and non-bloom conditions, (iii) determine if enhanced BCP efficiency during salp blooms is due to FPs or the phytoplankton community supporting the bloom, and (iv) investigate whether salp blooms alter the composition of prey exported from the euphotic zone. A Lagrangian experimental framework was employed, involving sampling of five water parcels over 3–7.5 days. Each parcel was characterized by water mass (SA or ST) and the presence or absence of a salp bloom. Measurements included hydrographic and nutrient conditions, salp abundance, biomass, and grazing rates, phytoplankton stocks and growth/grazing rates, and particle export flux. Particle export flux was assessed using free-drifting Particle-Interceptor Traps (PITs) at various depths (typically 70, 100, 300, and 500 m) and ²³⁸U/²³⁴U disequilibrium methods. The Lagrangian approach ensured the sampling of distinct water parcels over time, tracking natural changes in the system. Detailed methodologies were used to measure various parameters: * **Physical Oceanography:** CTD casts provided profiles of temperature, salinity, dissolved oxygen, and PAR. Nutrient samples were analyzed using a microsegmented flow analyzer. * **Phytoplankton:** Chlorophyll a (total and size-fractionated), chemotaxonomic pigments (HPLC), and 18S rDNA metabarcoding were used to assess phytoplankton biomass, community composition, and physiological status. Phytoplankton physiology was evaluated using a fast repetition rate fluorometer (FRRF). * **Net Primary Production (NPP):** ¹⁴C assays were used to measure NPP over 24-h incubations. * **Phytoplankton Growth and Grazing:** Dilution techniques were used to measure phytoplankton growth and microzooplankton grazing rates. * **Pigment Analysis:** Chlorophyll a and phaeopigments were determined using a fluorometer. * **16S and 18S rRNA gene amplicon sequencing:** DNA metabarcoding was used to determine the composition of prokaryotic and eukaryotic communities in the water column, PITs and sediments. * **²³⁸U-²³⁴Th Disequilibrium:** This method was used to estimate export fluxes. * **Zooplankton Abundance and Biomass:** Zooplankton net tows quantified salp and other zooplankton abundance and biomass. Salp size and stage were recorded. * **Salp Grazing:** Chlorophyll a and phaeopigments in salp guts were measured to estimate grazing rates. * **Fecal Pellet Production:** Fecal pellet production rates were calculated using two methods: one based on grazing and egestion efficiency, and another using a published length-FP production relationship from Antarctic waters, adjusting for temperature differences. * **Particle Interceptor Traps:** Free-drifting PITs collected sinking particles at different depths, which were analyzed for POC, PON, chlorophyll a, and DNA metabarcoding. * **Statistical Analyses:** Various statistical methods including Wilcoxon signed-rank tests, PERMANOVA and DESeq2 were used.

The study revealed a significant effect of salp blooms on microbial dynamics and POC export. * **Phytoplankton Dynamics:** Salp grazing resulted in negative phytoplankton growth rates in all salp bloom locations, except for the non-salp subtropical site. The rate of decrease was influenced by both salp and zooplankton grazing, and the magnitude of the effect varied across sites. In the early salp bloom site (Salp SA-Sc A), the addition of salp grazing resulted in a roughly one-third reduction in net primary productivity (NPP) between two measurement periods. * **Carbon Export:** Sediment trap data (PITs) indicated significantly higher POC export fluxes in salp bloom areas compared to non-salp areas. The ratio ranged from 2- to 10-fold, averaging approximately 5-fold higher in salp areas. This higher export flux was observed across various depths, below the euphotic zone. Analysis of trap contents showed a substantial contribution (20-40%) of intact salp FPs to export at the base of the euphotic zone, with up to 50% contribution to mesopelagic flux. * **BCP Efficiency:** The BCP efficiency was substantially higher in salp bloom areas. The efficiency of exporting NPP below the euphotic zone increased from 5% to 46% in subtropical waters and from 11% to 42% in subantarctic waters in the presence of salps. Overall BCP efficiency increased from 5% to 28% in the presence of salp blooms, among the highest recorded globally. * **Microbial Communities:** Analysis of microbial communities in the water column and PITs revealed that water mass was the primary driver of community composition, but the presence or absence of salp blooms also explained a significant portion of the variance, particularly in the PIT samples. The composition of sinking phytoplankton in the PITs did not indicate any single algal group as the primary driver of export across high export (salp) locations. * **Salp DNA:** Salp DNA was detected in four of five cycles, primarily concentrated in the euphotic zone and also found in PITs and sediments, suggesting both rapid transport to depth and deposition in the sediment record. * **Export Composition:** There was greater similarity in community composition between water column and PIT samples in salp bloom locations compared to non-salp locations, particularly in subtropical waters. In Salp ST, the dominant prymnesiophyte, *Gephyrocapsa oceanica*, which is generally not prevalent in exported material, was significantly enriched in the PIT samples, indicating a role for salps in altering the export composition. * **²³⁸U-²³⁴Th Disequilibrium:** While this method provided additional evidence for enhanced export in salp bloom areas, comparisons with PIT fluxes showed some inconsistencies, highlighting the complexities of integrating export signals over different time scales. In particular, the ²³⁸U-²³⁴Th data suggested some potential for advection of exported material from western to eastern bloom sites.

The results clearly demonstrate the significant impact of salp blooms on phytoplankton dynamics and carbon export in the study region. Salp grazing substantially reduced phytoplankton biomass, exceeding the impact of microzooplankton grazing in several locations. The large increases in POC export fluxes in salp bloom areas were primarily attributed to salp fecal pellets, either directly sinking or contributing to downstream export after disaggregation. The observed enhanced BCP efficiency in salp bloom areas highlights the importance of considering zooplankton grazing as a key driver of carbon export, not simply changes in primary production. The study suggests that salp blooms may not be dependent on a specific phytoplankton group, as blooms were observed in waters dominated by either diatoms or prymnesiophytes, yet in the case of prymnesiophytes, the export was disproportionately greater in the presence of salps. The spatial patterns across the bloom suggest a temporal progression in bloom evolution, likely driven by advection from the west (near-shore) to the east (off-shore). This interpretation is supported by the ²³⁴Th disequilibrium data, which integrate over longer time periods than the sediment traps. While the ²³⁸U-²³⁴Th data suggested a high-flux event in the non-salp subantarctic area, the elevated salp DNA in sediments from this area may indicate repeated particle delivery associated with the earlier salp bloom events. The results of this study have broad implications for understanding carbon cycling in the Southern Ocean and the potential impacts of climate change. The findings emphasize the importance of considering factors beyond NPP, such as changes in zooplankton communities, when modeling and predicting future changes in Southern Ocean biogeochemistry. The disproportionately high export of *Gephyrocapsa oceanica* in the presence of salps also indicates a potential impact on inorganic carbon fluxes.

This study provides compelling evidence for the substantial role of salp blooms in enhancing passive POC export and BCP efficiency in the Southwest Pacific sector of the Southern Ocean. Salp grazing significantly impacted phytoplankton dynamics, and the increased export was largely attributed to salp fecal pellets. The observed high BCP efficiencies in salp bloom areas highlight the importance of zooplankton grazing in Southern Ocean biogeochemistry, emphasizing the need to incorporate these factors into climate change predictions. Future research should focus on exploring the interplay between water mass characteristics, salp bloom initiation, and the complex interactions within the microbial food web that contribute to the remarkable carbon export observed in this study.

The study's Lagrangian design, while providing valuable insights into temporal dynamics, limited the number of sampling locations. The ²³⁸U-²³⁴Th disequilibrium method, while providing a longer-term integration, showed some inconsistencies with the PIT data, likely due to the inherent challenges of integrating over different time scales in a dynamic system. The study focused primarily on fecal pellet export, potentially underestimating the total impact of salps on the carbon cycle by neglecting other export pathways, such as sinking carcasses and active diel vertical migration. Further research is needed to quantify these additional export pathways.