Benthic Nitrogen Loss And Transformations In The East China Sea

Nitrogen plays a vital role in regulating global marine primary productivity andconsequently climate change. Its availability depends on the balance between nitrogenfixation through diazotrophs and fixed nitrogen loss via denitrification and anammox.However, whether marine nitrogen budget is in balance or not is still on debate. Thisis due to the scarcity of direct measurements for each process of marine nitrogen cycleand its complicated composition. Thus accurate rate measurement and nitrogentransformation process confirmation determine marine nitrogen budget to a largeextent. Marine sediment accounts for50⁷0%of global marine N-loss, and thus playsa dominant role. Among N-loss from sediment, continental shelf accounts for morethan50%of total benthic N-loss although it only comprises7.5%of the total seafloor.The East China Sea （ECS） shelf is one of the widest and most productive shelves. Itwas speculated that the ECS would be a hotspot for benthic N-loss; however, poorlystudies of benthic N-loss were conducted in this area. The aims of this thesis are toidentify the specific nitrogen transformation processes, to distinguish respectivecontribution to benthic nitrogen transformation, and to determine the extent of benthicN-loss on the ECS shelf. In this thesis, we chose the prevalent¹⁵N isotope pairingmethod (¹⁵N IPT) to investigate the benthic nitrogen cycle in the ECS from both fieldmeasurements and mathematical model. Furthermore, both sediment slurry and intactcore incubations were conducted in the field measurement to give a full understandingof benthic nitrogen cycle. Preliminary results were as follows:（1） The application of isotope pairing technique （IPT） in sediments whereanammox, denitrification and DNRA coexist was discussed in detail basedon a mathematical model. The mathematical expression demonstrated thatco-occurrence of DNRA with anammox and denitrification would change¹⁵N combination in anammox. Coexistence of DNRA would underestimateanammox and overestimate denitrification if their rates were calculatedaccording to the equations given by Thamdrup and Dalsgaard （2002） forslurry incubation. Therefore, ra was underestimated. The underestimation ofanammox was proportional to¹⁵NH₄⁺fraction in NH₄⁺pools （FA）; however,the overestimation of denitrification was related to both FAand ra. For intact core incubation, genuine N₂production was decreased since DNRA wouldcompete with anammox and denitrification under nitrate limiting. Threealternative procedures were proposed to correctly quantify anammox,denitrification and DNRA rates in slurry incubation. The first two methodsbased on estimating30N₂production via anammox. The third one based onconstraining FAto lower than5%in the way of elevating14NH₄⁺concentration. The validity and feasibility of the proposed calculationprocedures were verified by field studies in the ECS and Yellow Sea. Ourresults implied that DNRA did not influence anammox and denitrificationsignificantly in estuarine sediment with relative low ra （<10%）, however,those sediments with high ra and DNRA rate should be treated verycautiously for future studies and DNRA was strongly recommended to bemeasured.（2） Benthic nitrogen transformation pathways were investigated in the sedimentof the East China Sea （ECS） in June of2010using the¹⁵N isotope pairingtechnique. Slurry incubations indicated that denitrification, anammox anddissimilatory nitrate reduction to ammonium （DNRA） as well as intracellularnitrate release occurred in the ECS sediments. These four processes did notexist independently, nitrate release therefore diluted the¹⁵N labeling fractionof NO₃^-–, and a part of the¹⁵NH₄⁺derived from DNRA also formed30N₂viaanammox. Hence current methods of rate calculations led to over andunderestimations of anammox and denitrification respectively. Following theprocedure outlined in Thamdrup and Dalsgaard （2002）, denitrification rateswere slightly underestimated by an average6%without regard to the effectof nitrate release, while this underestimation could be counteracted by thepresence of DNRA. On the contrary, anammox rates calculated from¹⁵NO₃^-–experiment were significantly overestimated by42%without consideringnitrate release. In our study, this overestimation could only be compensated14%by taking DNRA into consideration. In a parallel experiment amendedwith¹⁵NH₄⁺+¹⁴NO₃^-–, anammox rates were not significantly influenced byDNRA due to the high background of¹⁵NH₄⁺addition. The significantcorrelation between potential denitrification rate and sediment organic mattercontent （r=0.68, p<0.001, Pearson） indicated that denitrification wasregulated by organic matter, while, no such correlations were found for anammox and DNRA. The relative contribution of anammox to the totalN-loss increased from13%at the shallowest site near the Changjiang estuaryto50%at the deepest site on the outer shelf, implying the significant role ofanammox in benthic nitrogen cycling in the ECS sediments, especially on theouter shelf. N-loss as N₂was the main pathway, while DNRA was also animportant pathway accounting for20³1%of benthic nitrate reduction in theECS. Our study demonstrates the complicated interactions among differentbenthic nitrogen transformations and the importance of consideringdenitrification, DNRA, anammox and nitrate release together when designingand interpreting future studies.（3） Benthic N-loss and DNRA rates on the ECS continental shelf were directlymeasured in two cruises in June and October-November2010usingprevalent¹⁵N isotope pairing technique （IPT） through sediment intact coreand slurry incubations. Measured benthic N-loss rate ranged from0.13to0.85mmol N m^-2d^-1with an average of0.37mmol N m^-2d^-1using revisedIPT suggested by Risgaard-Petersen （2003）, where anammox accounts for11⁴9%with an average of24%. DNRA ranged from0to0.05mmol N m^-2d^-1with an average of0.02mmol N m^-2d^-1. According to the Fick’s first law,we speculated that just a pseudo-steady state was established for¹⁵NO₃^-when IPT was employed.¹⁵NO₃^-could not diffuse more than1cm insediment, which caused a substantial part of benthic N-loss missingcontributed by NO₃^-below¹⁵NO₃^-penetration zone. Furthermore, severalevidence from re-examination based on published literature also supportedthat benthic N-loss was underestimated when IPT was applied. After addingthe missing benthic N-loss based on sediment slurry incubation, total benthicN-loss increased by a factor of0.8².2with an average of1.6and DNRAincreased by1order of magnitude to an average of0.29mmol N m^-2d^-1when the missing benthic N-loss based on sediment slurry incubation wasadded. The additional benthic N-loss was calculated with an average of1.6Tg N a-1in the East China Sea. If this additional N-loss could be extrapolatedto the global continental shelves, then an excess of80Tg N a-1can then beadded to the global marine nitrogen loss budget, accounting for52±23%ofthe global marine benthic nitrogen loss, which further aggravates theunbalance between nitrogen fixation and loss in marine environment. （4） The response of benthic nitrogen cycle to bottom water hypoxia wassynthetically investigated in the Changjiang estuary and adjacent area in twocruises in May and August2011. In this study benthic oxygen uptake,sediment oxygen profile, net flux of nitrate and ammonium, anammox,denitrification, DNRA, nitrification and mineralization were studied using¹⁵N isotope pairing and O₂microelectrode techniques. Sediment oxygenuptake and oxygen penetration depth averagely decreased23%and29%witha natural50%decline of bottom water oxygen from²00μM in May to¹00μM in August. NH₄⁺inclined to release from sediment to overlyingwater, while NO₃^-was prone to shifting from efflux to influx to sediment. rasignificantly decreased from20%to7.4%, leading to anammox ratedecreased by a factor of2.5from0.15to0.06mmol N m^-2d^-1. However,denitrification showed a slight increase causing total benthic N-loss stablizedat0.85mmol N m^-2d^-1. DNRA showed a significant increase by a factor of5from0.02to0.10mmol N m^-2d^-1. These changes were consistent with theresults derived from artificially bottom water oxygen condition controlledexperiments （oxic, ambient and severe hypoxia）: sediment oxygen uptakedecreased by as much as91%when bottom water oxygen dropped92%fromnormal oxic condition （²00μM） to severe hypoxia （¹6μM）. NH₄⁺showedan enhanced release to overlying water from⁰to0.60mmol N m^-2d^-1andNO₃^-shifted from an efflux of0.14mmol N m^-2d^-1to an influx of0.79mmol N m^-2d^-1. Denitrification and anammox showed an average decreaseof38%and43%under severe hypoxia, leading to total benthic N-lossdecline of38%from0.92to0.57mmol N m^-2d^-1. DNRA showed anelevation by a factor of3although it only accounted for less than10%oftotal nitrate reduction.Our work in the ECS provided preliminary understanding of benthic nitrogentransformations, but there is still a long way to go to exhibit a comprehensive andintensive picture of nitrogen cycle due to its inherent complexity. With the boomingdiscoveries in the field of marine nitrogen cycle, studies will surely face numerousopportunities and challenges.