The Vertical Distribution Structure Of Phytoplankton Biomass And Its Response To Marine Environment

After nearly 30 years of development,ocean color remote sensing satellite data has played an important role in the research of marine ecosystems and carbon cycle.The ocean color satellites have not only achieved the high-precision estimation of the global surface chlorophyll-a concentration and the monitoring of their long-term variations,but also puts forward many Remote sensing Productivity Model.These data have provided a preliminary map of the global phytoplankton biomass and their pigment concentration and has greatly promoted the progress in the research of the ocean carbon cycle.At the same time,satellite data is also a key parameter for studying ecosystems and biogeochemical environments,and has made great contributions to deepening our understanding of the coupling of marine biology and the physical/chemical environment.However,we must realize that the ocean color remote sensing can only get the weighted average information from the upper surface layer of the ocean,but it is hard to obtain the parameter profiles in the water column.For a long time,the research of marine biogeochemical processes has also been seriously restricted by the lack of vetical observation data.In recent years,the progress of theoretical research in the underwater light field and the widespread deployment of the BGC-Argo buoy provide a basis for further exploration of the vertical distribution of marine phytoplankton and its pigments,which makes the research on the response of the marine ecosystem to the changing physical&chemical background fields as possible.This paper combines satellite data,multiple in-situ datasets from ship-borne cruises,and the BGC-Argo buoy to invert the vertical distribution structure and the total concentration of marine phytoplankton and its pigments(chlorophyll-a),then analyze two typical biogeochemical coupling cases:the formation mechanism of the spring bloom in the North Atlantic,and the response of the new production in the North Pacific to ENSO events,such as El Nino/La Nina.①The conventional use of optically determined 1%of surface photosynthetically available radiation(PAR)depth(Z1%PAR,λ=400 to 700 nm)as a metric for the euphotic zone depth(Zeu)has been a matter of debate for several decades because of frequent inconsistencies with the base of euphotic zone determined biologically,i.e.,the compensation depth(Zc).In this study,we attempt to reconcile between optical and biological determinants of the euphotic zone through use of a large dataset of coincidental profiles of downwelling irradiance and primary production.These measurements cover open ocean waters in the tropics,subtropics,temperate regions,and from two time-series stations,the Hawaii Ocean Time series(HOT)and the Bermuda Atlantic Time-series Study(BATS).We report that Z1%PAR is too shallow(by～15.5%)compared to Zc,while Z0.1%PAR is too deep(by～27.3%).Further,the irradiance at Zc(i.e.,the compensation irradiance,Ic)varies by a factor of more than five,but its ratio to surface irradiance is relatively stable.In general,Ic corresponds to 0.48±0.23%of surface PAR,or 0.87±0.40%of surface usable solar radiation(USR,for λ=400 to 560 nm),or 1.50±0.6 7%of surface downwelling irradiance at 490 nm.These results suggest that Z0.5%PAR,or Z0.9%USR,or Z1.5%490 could be promising alternatives to bridge the gap between optical and biological determinants of the euphotic zone depths at least in the open ocean at low and middle latitudes.②The "biological pump" depends on the vertical distribution of chlorophyll-a(VDOC)in the upper water column.On the other hand,this vertical distribution is a result of multiple forces,and it has been a long-standing challenge to determine this vertical distribution of the global oceans based on data from remote sensing data.Here,using data in the South China Sea(SCS)as examples,the interaction of underwater light field and mixed-layer depth(MLD)is used to improve the determination of VDOC.Specifically,four types of VDOC(vertically uniform,Gaussian,exponential,and hyperbolic)were found from 497 profiles obtained from 2001 through 2015.For these observations,MLD from temperature profiles and Zeu from remote sensing reflectance(Rrs)are derived.Analyses of this dataset suggest that the VDOC is always exponential or hyperbolic type on the coast area,with high chlorophyll-a concentration in the surface and decreased with increasing depth.Further,the relationship between Zeu and MLD shows strong indications on VDOC for offshore waters.When Zeu is shallower than MLD,VDOC is generally uniform and without a subsurface chlorophyll-a maximum(SCMs).However,if Zeu is deeper than MLD,generally VDOC is in a Gaussian form with SCMs.The parameters used to model a Gaussian type(background chlorophyll-a(Chls);maximum depth(Zmax),maximum chlorophyll-a(Chlmax),and thickness(W)of SCMs)are further modeled from the measured data.The results indicate the precision and accuracy of the new algorithm of VDOC are improved.After that,the spatial and temporal distribution of the integral chlorophyll-a concentration in the Zeu is obtained.③ Next,we improved and regionalized the Remote sensing Productivity Model.In recent years,the widely-used models for estimating the marine primary productivity in Zeu(IPP)through remote sensing satellite data are the chlorophyll-based VGPM,the carbon-based CbPM,and the phytoplankton absorption coefficient aph(λ)-based AbPM.Firstly,we use the long-time measured data of HOT and BATS stations for more than 20 years to evaluate the performance of the three classic models in capturing the seasonal dynamic distribution of IPP.The results show that the AbPM model has the best accuracy and stability.However,due to the complexity of the marine environment,it is necessary to regionalize the input parameters of the AbPM model.Based on the existing research,we regionalized the Zeu and photosynthetic quantum yield(Φ)in the North Atlantic and North Pacific,then obtained more accurate IPP and their spatiotemporal variation.On the other hand,we combined the KORUS cruise around the Korean peninsula with the high-spatiotemporal-resolution geostationary satellite GOCI datasets to establish a regionalized AbPM.Specifically,we first divided several biooptical zones through a variety of cluster analysis and machine learning methods(SOM)with the biological,physical,and optical parameters as inputs;we also matched the insitu Φ with the bio-optical zones.Then the IPP is calculated from in-situ Φ,PAR,and aph(λ)estimated from GOCI by the AbPM.Finally,the Sobol method is used to analyze the global sensitivity of the input parameters and output variables of this regionalized AbPM.In summary,we have discovered and demonstrated the broad prospects of AbPM in the regional estimation of marine IPP.④ The Critical Depth Hypothesis(CDH),one of the fundamental tenets of plankton ecology,posits that the spring bloom in the North Atlantic Ocean is initiated when the MLD becomes shallower than the critical depth(ZCR),a condition ensures that vertically integrated net rates of phytoplankton productivity exceed community loss rates,and allows phytoplankton to grow and form large blooms.Recent studies have challenged the concept of the CDH in the North Atlantic and other locations that experience spring blooms,and instead,have proposed alternate hypotheses and physical and biological mechanisms that deviate from those on which the CDH was based upon.In this study,we have used a new optical model for estimating the diffuse attenuation coefficients of visible light in the water column,and spatially and temporally enhanced bio-optical datasets from Suomi-VIIRS and BGC-Argo floats to show that current misgivings about the CDH stem from lack of both adequate and accurate measurements of community compensation irradiance(Ic-com),i.e.the light intensity in the water column where photosynthetic and ecosystem community loss processes are in balance,and a key input for calculating ZCR.Variations in ZCR,MLD,and changes in chlorophyll-a concentration,over seasonal,annual,and interannual time scales(2002-2018),across different latitudinal bands in the North Atlantic,lend strong support for the CDH.Finally,both of the long-term observation data of the VIIRS satellite and the BGC-Argo profile data support the mechanism of the North Atlantic spring bloom described in the CDH theory.⑤ The productivity supported by the new nutrients entering Zeu is called New Production(NP),which is an important concept for understanding marine production and carbon export.Researches have already shown that NP can be calculated based on the C:N ratio and the NO3 consumed by phytoplankton during the growing season.NO3 can be obtained from satellite Sea Surface Temperature(SST)and Chlorophyll-a Concentration(Chl).Here,the long-time distributions of NO3 and NP from 2003 to 2018 were calculated through VIIRS and MODIS,respectively.Then we delineated 11 zones corresponding to main ocean currents in the North Pacific and individually analyzed the relationship of NO3 and environmental factors.The results indicated that PAR,sea surface height(SSH),and wind stress show a significant correlation with NO3.It’s also worth to mention that NO3 and NP in North Pacific have a strong response to the strength and periodicity of the ENSO events,but show different patterns in different zones.In summary,the changes in NO3 and NP near the ocean boundary are more dramatic.After the ENSO event,NP was suppressed across the east coast of the North Pacific and the Gulf of Alaska,while NP in the western North Pacific was much higher.