Carbon cycling in the deep eastern North Pacific benthic food web: Investigating the effect of organic carbon input

The deep ocean benthic environment plays a role in long-term carbon sequestration. Understanding carbon cycling in the deep ocean floor is critical to evaluate the impact of changing climate on the oceanic systems. Linear inverse modeling was used to quantify carbon transfer between compartments in the benthic food web at a long time-series study site in the abyssal northeastern Pacific (Station M). Linear inverse food web models were constructed for three separate years in the time-series when particulate organic carbon (POC) flux was relatively high (1990: 0.63 mean mmol C m d), intermediate (1995: 0.24) and low (1996: 0.12). Carbon cycling in all years was dominated by the flows involved in the microbial loop; dissolved organic carbon uptake by microbes (0.80–0.95 mean mmol C m d), microbial respiration (0.52–0.61), microbial biomass dissolution (0.09–0.18) and the dissolution of refractory detritus (0.46–0.65). Moreover, the magnitude of carbon flows involved in the microbial loop changed in relation to POC input, with a decline in contribution during the high POC influxes, such as those recently experienced at Station M. Results indicate that during high POC episodic pulses the role of faunal mediated carbon cycling would increase. Semi-labile detritus dominates benthic faunal diets and the role of labile detritus declined with increased total POC input. Linear inverse modeling represents an effective framework to analyze highresolution time-series data and demonstrate the impact of climate change on the deep ocean carbon cycle in a coastal upwelling system. The oceanic biological carbon pump (BCP) involves the transfer of particulate and dissolved organic carbon from the surface to the deep ocean (Sigman and Boyle 2000; Boyd and Trull 2007). The BCP also plays a vital role in removing carbon dioxide from the atmosphere (Sanders et al. 2014) and sequestering carbon in deep ocean sediments for thousands of years (Honjo et al. 2014). However, understanding of the response of the BCP and the deep ocean carbon sink to changes in primary production and organic carbon export fluxes as a result of climatic change is limited (Honjo et al. 2014; Sanders et al. 2014). Ocean stratification is increasing with rising ocean surface temperatures, which has resulted in reduced nutrient exchange to surface waters via upwelling (Falkowski and Oliver 2007). Globally, primary producer communities are thought to have shifted from diatoms to smaller species to adapt to these nutrient limitations, resulting in changes in the quantity and quality of particulate organic carbon (POC) export to the deep ocean (Buesseler et al. 2007; Smith et al. 2013). However, some coastal areas have been experiencing increased wind stress and nutrient upwelling, which have led to peaks in primary production and high POC flux to the deep sea floor. The deep ocean benthic environment occupies almost two thirds of the earth and plays a role in centennial and longer-term carbon sequestration and regeneration of nutrients (Smith et al. 2013). Therefore, understanding deep ocean food web and carbon cycle processes and how they will alter with climate change informs on the basic function of Earth systems. *Correspondence: [email protected] Present address: Heriot-Watt University, Edinburgh, EH14 4AS This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. 1956 LIMNOLOGY and OCEANOGRAPHY Limnol. Oceanogr. 61, 2016, 1956–1968 VC 2016 The Authors Limnology and Oceanography published by Wiley Periodicals, Inc. on behalf of Association for the Sciences of Limnology and Oceanography doi: 10.1002/lno.10345 Determining the climatic impacts on biogeochemical processes of the deep ocean carbon cycle requires long-term observations of POC flux and benthic community dynamics. Developments in deep ocean instrumentation have enabled significant advances in high resolution, inter annual to decadal monitoring of deep ocean communities and POC input to assess temporal and spatial changes in relation to climate (Sherman and Smith 2009). Deep ocean tethered sediment traps (> 1000 m) are the only means by which sustained measurements of seasonal and inter-annual variability in POC flux have been measured (Antia et al. 2001; Sanders et al. 2014). The 26-year time-series study at Station M, North Pacific Ocean, estimates changes in surface food supply (POC), sedimentation events, megafauna activity and seasonal sediment community oxygen consumption (SCOC) as a measure of the organic carbon consumed by benthic communities (Smith et al. 2014). Time-series data are also available on the density, biomass and respiration of abyssal megafauna (Ruhl et al. 2014) and macrofauna (LaguionieMarchais et al. 2016). This represents one of the most comprehensive time-series studies related to the deep ocean benthic carbon cycle, and has substantially improved understanding of the connections between surface food supply and deep ocean benthic communities and the role of the deep ocean benthic environment in the oceanic BCP (Smith et al. 2014). Analysis of abyssal time-series data with satellite estimates of near-surface chlorophyll a, net primary production and export flux has found that changing surface ocean conditions translate directly to biological and biogeochemical activity in the deep ocean. Also that episodic (days to weeks) inputs of POC can have a significant effect on the variation of POC input to the deep benthic environment. Decadal peaks in supply, remineralization, and sequestration of organic carbon have broad implications for projections of global carbon sequestration (Smith et al. 2013). Quantifying present and future changes in the BCP will require modeling studies to be combined with long-term time-series datasets (Smith et al. 2013; Honjo et al. 2014), such as those collected at Station M. The complex structure of the deep ocean benthic food web and the limited accessibility to collect data makes directly quantifying processes in the deep ocean benthic carbon cycle difficult (Van Oevelen et al. 2009). Inverse modeling techniques were developed to resolve food web models with data deficiency by combining various sources of quantitative data with a topological flow network (V ezina and Platt 1988). The technique enables unmeasured flows of mass between food web compartments to be estimated. The model solves flow networks based on the constraint that elemental mass is conserved between compartments and must comply with rate measurements, biological constraints and the observed structure of the food web (V ezina and Platt 1988; Kones et al. 2009). Inverse food web modeling has been used to quantify carbon flows at two other deep benthic time-series study sites: the Porcupine Abyssal Plain (PAP) in the Northeast Atlantic (Van Oevelen et al. 2012) and the deep-sea observatory at HAUSGARTEN, eastern Fram Strait (Arctic Ocean) (Van Oevelen et al. 2011b). Linear inverse models (LIMs) have also been used to study the effect of characteristics of the upper, middle and lower sections of the Nazar e Canyon, eastern Atlantic Ocean, on carbon flows (Van Oevelen et al. 2011a). We are aware of no applications of this technique that resolve carbon flows in a time-series to identify how the benthic community responds to temporally variable conditions such as POC input. In this study, benthic food web models of carbon stocks and flows were developed through linear inverse modeling using data from the time-series from Station M. Models were developed for three years in the time-series when POC input were at relatively high, medium and low levels to investigate the effect of changing POC on organic carbon transfer in the deep ocean benthic food web. POC input to the deep ocean is changing in relation to climate change and the present retrospective model analysis will improve our understanding of the response of the deep ocean benthic carbon cycle in relation to climate-induced changes in POC deposition. Material and methods Study site The Station M abyssal 26-year time-series, in the eastern North Pacific Ocean, is located 220 km west of Point Conception, off the coast of central California (348 500N, 123800W; 4000 m water depth), on the Monterey Deep-Sea Fan. The site is characterized by low topographic relief and oxygenated silty-clay sediments (Smith and Druffel 1998). POC flux and benthic community dynamics have been monitored with measurements and model estimates of atmospheric and ocean surface conditions beginning in 1989 to examine the effect of changing climate on the deep ocean benthic ecosystem (Smith et al. 1994; Smith and Druffel 1998; Smith et al. 2013). Data on abiotic and biotic carbon stocks have been collected at Station M, which has enabled the development of linear inverse food web models. The years 1990, 1995, and 1996 were selected as they represent years of contrasting POC input and for which the most comprehensive dataset for the required elements of the food web is available. Food web structure Food web models consist of abiotic and biotic compartments with specific links between compartments that represent carbon flows (Soetaert and van Oevelen 2009). The deep ocean abyssal benthic food web is driven by the input of POC produced by primary production in the euphotic zone (Smith 1992). In the food web model carbon is split into labile (carbon linked to chlorophyll a (Chl a) content), semilabile (linked to lipid, proteins, and carbohydrates) and refractory detritus to reflect differences in quality and quantity. The model has compartments for labile (lDet_w), semiDunlop et al. Carbon Cycling in the North Pacific Benthic Food Web