Evaluation of the utility of xanthophyll cycle pigment dynamics for assessing upper ocean mixing processes at Station ALOHA

Journal of Plankton Research, Nov 2014

We had the opportunity to evaluate the utility of the chromophyte xanthophyll cycle pigments (XCP), diadinoxanthin (DDX) and diatoxanthin (DTX), for assessing upper ocean mixing processes at Station ALOHA (22.75°N, 158°W) during August–September 2012. De-epoxidation state (DES, [DTX]/[DDX + DTX]) kinetic rate experiments were performed that quantified the conversion of DDX to DTX in the light. DES values ranged from 0.13 to 0.51 (weight:weight), and yielded a first-order rate constant of 0.16 ± 0.02 min−1. In addition, a time-series of chromophyte pigments was collected from the sea-surface in order to quantify the time-dependent rate of change in the ratio of XCP to chlorophyll a (XCP/CHLA). In situ XCP/CHLA values were measured by rapid collection of cells in the mixed layer using a diaphragm pumping system and the xanthophyll cycle activity inhibitor dithiothreitol. Turbulent transfer velocity (TTV) was estimated as the quotient of the XCP/CHLA photoadaptation rate and the XCP/CHLA gradient in the mixed layer. TTVs ranged from 0.3 to 0.5 cm s−1, and agreed to within a factor of ∼2 of the Lagrangian float-derived estimate of vertical mixing at a comparable friction velocity. Coincident measurements of dissolved gases suggest that XCP dynamics may be useful for the interpretation of trace gases species distributions, such as hydrogen and carbon monoxide, in the upper ocean.

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:

https://plankt.oxfordjournals.org/content/36/6/1423.full.pdf

Evaluation of the utility of xanthophyll cycle pigment dynamics for assessing upper ocean mixing processes at Station ALOHA

ROBERT R. BIDIGARE 0 FENINA R. BUTTLER 0 STEPHANIE J. CHRISTENSEN 0 BENEDETTO BARONE DAVID M. KARL SAMUEL T. WILSON 0 HAWAII INSTITUTE OF MARINE BIOLOGY We had the opportunity to evaluate the utility of the chromophyte xanthophyll cycle pigments (XCP), diadinoxanthin (DDX) and diatoxanthin (DTX), for assessing upper ocean mixing processes at Station ALOHA (22.758N, 1588W) during August - September 2012. De-epoxidation state (DES, [DTX]/[DDX DTX]) kinetic rate experiments were performed that quantified the conversion of DDX to DTX in the light. DES values ranged from 0.13 to 0.51 (weight: weight), and yielded a first-order rate constant of 0.16 + 0.02 min21. In addition, a time-series of chromophyte pigments was collected from the sea-surface in order to quantify the time-dependent rate of change in the ratio of XCP to chlorophyll a (XCP/CHLA). In situ XCP/CHLA values were measured by rapid collection of cells in the mixed layer using a diaphragm pumping system and the xanthophyll cycle activity inhibitor dithiothreitol. Turbulent transfer velocity (TTV) was estimated as the quotient of the XCP/CHLA photoadaptation rate and the XCP/CHLA gradient in the mixed layer. TTVs ranged from 0.3 to 0.5 cm s21, and agreed to within a factor of 2 of the Lagrangian floatderived estimate of vertical mixing at a comparable friction velocity. Coincident measurements of dissolved gases suggest that XCP dynamics may be useful for the interpretation of trace gases species distributions, such as hydrogen and carbon monoxide, in the upper ocean. I N T RO D U C T I O N Turbulent mixing in the upper ocean is driven by a variety of airsea interactions including, but not limited to, Ekman transport, evaporative and nighttime cooling that creates vertical convection, wind-driven currents that produce friction velocities, wavecurrent interactions that generate Langmuir circulation cells and wave breaking that produces bubbles and turbulence. Upper ocean mixing processes are important because they provide a link between the atmosphere and the deep ocean, and are responsible for the air sea exchange of freshwater, heat, momentum and gases. The measurement and modeling of boundary layer turbulence in the ocean is challenging because it varies by several orders of magnitude. Consequently, a wide variety of approaches have been used to study upper ocean turbulence including acoustic drifters, thermistor strings, free fall airfoil probes, Doppler profilers, Lagrangian drifters and large-eddy simulations (Gargett, 1997). Turbulent mixing also influences predator prey interactions and particle aggregation, and controls the availability of new nutrients and photosynthetically active radiation (PAR, 400 700 nm) in the euphotic zone (Gargett, 1997). Consequently, turbulence has a profound effect on phytoplankton distributions, abundances, rates of carbon fixation and photoadaptation state. This subject matter resulted in the publication of a number of seminal papers in the 1980s (Cullen and Lewis, 1988; Denman and Gargett, 1983; Lewis et al., 1984) and continues to be an active area of research as evidenced by recent publications and ongoing research efforts. In particular, knowledge of photoadaptation rate kinetics and in situ measurements of photoadaptation state have been used to estimate (or constrain) rates of vertical mixing in the upper ocean (Brunet et al., 2003, 2008; Dusenberry, 1995; Dusenberry et al., 1999; Griffith et al., 2010). Phytoplankton utilize a wide variety of mechanisms for adapting to the variable light field experienced in the upper ocean. The timescales for the pigment-based adaptations range from seconds minutes in the case of xanthophyll cycling (XC), minutes hours for the synthesis of photoprotective carotenoids and hours days for the synthesis of light harvesting pigments (Ferris and Christian, 1991). For certain chromophytic microalgae (diatoms, xanthophytes, raphidophytes, euglenophytes, pelagophytes, prymnesiophytes and dinoflagellates; Brunet et al., 2011), the XC involves the reversible interconversion of diadinoxanthin (DDX) and diatoxanthin (DTX). The de-epoxidation reaction (DDX!DTX) follows first-order kinetics and can be modeled using the following equation (Griffith et al., 2010): kt lnDESt DES1=DESo where DES is the de-epoxidation state and defined as [DTX]/[DTX DDX] (dimensionless), k is the firstorder rate constant (min21), DESo is the initial darkadapted DES value, DESt is the DES value at time t (min) and DES1 is the DES value when pigment changes follow zero-order kinetics (i.e. the light-adapted DES value). Laboratory-based experiments performed with diatom cultures yield rate constants ranging from 0.1 to 0.3 min21 DDX ! DTX (Kashino and Kudoh, 2003; Lohr and Wilhelm, 1999; Olaizola and Yamamoto, 1994). In contrast, the epoxidation reaction (DTX! DDX) is considerably slower with rate constants ranging from 0.01 to 0.06 min21 (Kashino and Kudoh, 2003; Lohr and Wilhelm, 1999). XC activity is an important mechanism responsible for non-photochemical quenching (NPQ) and is correlated with decreases in the efficiency of maximum quantum yield for photosystem II activity (i.e. Fv/Fm) as measured by active fluorescence (Brunet and Lavaud, 2010). Based on a combined approach involving metabolite analysis identification of gene function with the model diatom Phaeodactylum tricornutum, Dambek et al. (Dambek et al., 2012) determined that DDX and fucoxanthin are synthesized from b,b-carotene via the following pathway: carotene ! Zeaxanthin ! Antheraxanthin ! Violaxanthin ! NeoxanthindbFDuiacdoixnaonxtahninthin Lepetit et al. (Lepetit et al., 2010) determined that the marine diatom, Cyclotella meneghiniana, has three different DDXDTX pools that are associated with: (i) the peripheral fucoxanthin chlorophyll protein (FCP) complexes, (ii) the FCP complexes bound to photosystem I (PSI), and (iii) a monogalactosyldiacyl glycerol (MGDG) lipid shield surrounding the peripheral FCP complexes. The first two pools are relatively small and do not vary under changing light conditions. Lepetit et al. (Lepetit et al., 2010) suggest that the peripheral FCP XC pool is involved in NPQ , and the PSI XC pool is involved in light harvesting and the detoxification of triplet chlorophyll (3Chl) and singlet oxygen (1O2). The third pool accounts for the majority of DDX DTX and increases under high light. Lepetit et al. (Lepetit et al., 2010) hypothesize that DTX in the lipid-dissolved XC pool functions as an antioxidant by scavenging 1O2 and peroxylipids. Measurements of xanthophyll cycle pigments (XCP) have been used in field studies for assessing phytoplankton light histories (Bidigare et al., 1987; Claustre et al., 1994; Olaizola et al., 1992) and photoadaptation strategies (Brunet et al., 2008; Kashino et al., 2002), water column stability (Moline, 1998), mixing rates in the upper ocean (Brunet et al., 2003, 2008; Griffith et al., 2010; Welschmeyer and Hoepffner, 1986) and the depth of origin of sinking particulate organic matter (Scharek et al., 1999). During May October of 2012, the Center for Microbial Oceanography: Research and Education (C-MORE, University of Hawaii) conducted a continuous longterm field experiment (HOE-DYLAN: Hawaii Ocean ExperimentDynamics of Light and Nutrients) at Station ALOHA to observe and interpret temporal variability in microbial processes, and the consequences for ecological dynamics and biogeochemical cycling. Special focus was given to time-space coupling because proper scale sampling of the marine environment is an imperative, but generally neglected aspect of marine microbiology. We had the opportunity to investigate time- and depth-dependent changes in phytoplankton pigments during 22 August 11 September 2012 (Cruise ID: KM1219). The primary objective of our research component was to evaluate the utility of XCP dynamics as a chemical tracer of vertical mixing processes. M E T H O D Laboratory investigation of dithiothreitol inhibition Since the interconversion of XCP occurs on timescales of seconds minutes, previous studies have employed the use of the XC activity inhibitor, dithiothreitol, in order to obtain reliable first-order rate constants (Kashino and Kudoh, 2003; Lavaud et al., 2002; Olaizola and Yamamoto, 1994). Prior to the field studies, a laboratory experiment was performed to determine the concentration of dithiothreitol (DTT) required for the inhibition of XC activity. A unialgal culture of Pelagomonas sp., isolated from Station ALOHA by Yoshimi Rii (University of Hawaii), was used in the DTT exposure experiment. The culture was dark-adapted overnight at room temperature. The following morning, 25 mL culture aliquots were dispensed into 75-cm2 cell culture flasks under dim red lighting. One flask (dark control) was immediately filtered onto a 25 mm diameter Whatman GF/F filter (nominal porosity of 0.7 mm) under dim light and frozen at 2208C. DL-Dithiothreitol was obtained from SigmaAldrich and used to prepare a 50 mM stock solution. With the exception of one of the remaining samples (light control), DTT was added to the individual flasks under dim red lighting; the final concentrations ranged from 50 to 495 mM DTT. The samples were allowed to sit in the dark for 30 min. The flasks were then exposed to an irradiance of 300 mmol quanta m22 s21 (400 700 nm) for 30 min at 258C in a Percival model E22L incubator. The samples were immediately filtered onto 25 mm diameter Whatman GF/F filters under dim light and frozen at 2208C. Samples were extracted overnight and analyzed the following day by the high-performance liquid chromatography (HPLC) using the procedure described below. Study site and hydrography Field studies were conducted aboard the R/V Kilo Moana at Station ALOHA (22.758N, 1588W) during KM1219 (22 August 11 September 2012). Conductivity temperature depth ( pressure) measurements were performed with a Sea-Bird SBE 911plus CTD system attached to a rosette of 24 12-L polyvinyl chloride sample bottles. The conductivity sensor was calibrated using discrete sample salinity measurements as described by Bingham and Lukas (Bingham and Lukas, 1996). Mixed layer depths (m) were defined using a density offset of 0.03 kg m23 from a depth of 10 m (de Boyer Montegut et al., 2004). Vertical profiles of downwelling PAR irradiance were obtained every day using a FreeFalling Optical Profiler (Satlantic). A deck sensor by the same manufacturer was mounted on the upper deck of the ship and measured downwelling PAR at the sea surface while the vertical profiles were recorded. Both these measurements were used to compute the percent level of PAR with respect to the value at the sea surface on 1 m depth bins in the water column. Meteorological data were obtained from the Woods Hole Oceanographic InstitutionHawaii Ocean Timeseries Site (WHOTS, http://www.soest.hawaii.edu/whots/) mooring located near Station ALOHA (22.778N, 157.908W). The mooring has been in place for 1-year periods since 2004. Data for this study were acquired during the WHOTS-9 deployment (14 June 2012 14 July 2013). The WHOTS surface buoy is equipped with meteorological instrumentation that measured air and sea-surface temperatures, relative humidity, barometric pressure, wind speed and direction, incoming shortwave and long-wave radiation, and precipitation. Wind speed (m s21) data were measured at a height of 3.3 m during 1100 1500 HST, and were used to calculate values at 15 m (U15, m s21) and corresponding friction velocities (u*, m s21) as described in DAsaro (DAsaro, 2001). Field studies of XCP kinetics DES kinetic rate parameter experiments were performed on 24 and 30 August (KM1219). Sea-surface water samples were collected with an 8-L plastic bucket at 1000 HST (PAR 1280 1300 mmol quanta s21 m22) and transferred to a 20-L clear, UV-opaque, polycarbonate carboy. The carboy was wrapped in a black plastic bag and placed into a Rubbermaid Action Packer storage container. A Biospherical Instruments model QSL-2100 radiometer was used to measure the PAR irradiance inside of the dark incubator and yielded a value of 0 mmol quanta m22 s21. After a dark adaption period of 60 min in the shade, the carboy was placed in direct sunlight and samples were collected at 0, 1, 2, 3, 4, 5, 10, 15, 30 and 45 min (n 10 samples). Single 2.16-L samples were directly dispensed into opaque sample bottles containing DTT to rapidly inhibit XC activity. Samples were filtered onto 25 mm diameter Whatman GF/F filters, flash-frozen in liquid nitrogen and then stored in the 2808C shipboard freezer. Samples were transported to the University of Hawaii on dry ice, and stored at 2808C prior to analysis using the HPLC method described below. The results of these experiments were combined and modeled using Eq. (1). Linear regression analysis and the standard error of the resulting slope were obtained using Minitab 13 statistical software. Phytoplankton pigment time-series measurements A time-series of phytoplankton chlorophyll and carotenoid pigments was collected from the surface waters (5 m) of Station ALOHA during 23 August to 7 September. Seawater was collected at 1200 or 1600 HST using the CTD-rosette system described above. Single 2.16-L samples were filtered onto 25 mm diameter Whatman GF/F filters, flash-frozen in liquid nitrogen and then stored in the 2808C shipboard freezer. Samples were transported to the University of Hawaii on dry ice, and stored at 2808C prior to analysis using the HPLC method described below. A second time-series of phytoplankton pigments was collected from the surface waters of Station ALOHA during 5 and 6 September. Hourly bucket samples were collected during daylight hours (dawn to dusk). Sea-surface irradiance was continuously measured during the cruise using a Biospherical Instruments model QSR-2200 PAR sensor mounted on the mast of the R/V Kilo Moana. Bucket samples were transferred into 2-L opaque sample bottles containing DTT to rapidly inhibit XC activity. Replicate samples were collected on 6 September under low-light (0646 HST, PAR 400 mmol quanta s21 m22, n 4) and high-light (1415 HST, PAR 700 mmol quanta s21 m22, n 4) conditions to assess the variability associated with the HPLC pigment determinations. Single 2.16-L samples were filtered onto 25 mm diameter Whatman GF/F filters, flash-frozen in liquid nitrogen and then stored in the 2808C shipboard freezer. Samples were transported to the University of Hawaii on dry ice, and stored at 2808C prior to analysis using the HPLC method described below. Chromophyte pigment vertical profile measurements Samples for determining vertical profiles of phytoplankton pigments during KM1219 were collected with a Wilden diaphragm pumping system equipped with 75 m of Teflonw tubing that was labeled in 5 m increments. The pumping system was deployed during 1400 1500 HST and 10 samples were collected at 5 m intervals in the upper 50 m. The calculated residence time of seawater in the tubing was 1.2 min. After flushing the tubing for 3 min, samples were directly dispensed into 2-L opaque sample bottles containing DTT to rapidly inhibit XC activity. Single 2.16-L samples were filtered onto 25 mm diameter Whatman GF/F filters, flash-frozen in liquid nitrogen and then stored in the 2808C shipboard freezer. Samples were transported to the University of Hawaii on dry ice, and stored at 2808C prior to analysis using the HPLC method described below. Average downwelling PAR at the sea-surface was measured every 10 min during the diaphragm-pump vertical casts using a LI-COR model LI-192 cosine collector mounted on the top deck of the R/V Kilo Moana. Mixed layer depth (m) was estimated as the mean of the values determined before (1200 CTD cast) and after (1600 CTD cast) a given vertical profiles. Phytoplankton pigment measurements Chlorophyll and carotenoid pigments were analyzed at the University of Hawaii using the HPLC method described by Bidigare et al. (Bidigare et al., 2005). Pigment samples were extracted at 48C in the dark for 24 h in 3 mL of 100% HPLC grade acetone (Fisher Scientific) plus an internal standard (50 mL canthaxanthin). After extraction, samples were centrifuged (5 min at 1500 g) to sediment filter and cellular debris. Samples of a mixture of 0.3 mL HPLC grade water (Fisher Scientific) plus 1.0 mL extract were injected onto a Varian 9012 HPLC system equipped with a Varian 9300 autosampler, a Timberline column heater (268C) and Spherisorb 5 mm ODS2 analytical (4.6 250 mm) column and corresponding guard cartridge. Pigments were detected with a ThermoSeparation UV2000 detector (l 436 nm and 450 nm). A ternary solvent system was employed for HPLC pigment analysis: eluent A (MeOH:0.5 M ammonium acetate, 80:20), eluent B (acetonitrile:water, 87.5:12.5) and eluent C (ethyl acetate). The linear gradient used for pigment separation is a modified version of that originally described by Wright et al. (Wright et al., 1991): t 0 min (90% A, 10% B), t 1 min (100% B), t 11 min (78% B, 22% C), t 27.5 min (10% B, 90% C) and t 29 min (100% B). Eluents A and B contain 0.01% of 2,6-di-tert-butyl-p-cresol (Sigma-Aldrich) to prevent the conversion of chlorophyll a into chlorophyll a allomers. HPLC grade solvents (Fisher Scientific) were used to prepare eluents A, B and C. The eluent flow rate was held constant at 1 mL min21. Pigment peaks were identified by comparing their retention times with those of pigment standards provided by DHI Lab Products (Hrsholm, Denmark; http://c14.dhigroup.com/Product Descriptions/PhytoplanktonPigmentStandards.aspx) and extracts prepared from phytoplankton reference cultures. Pigment concentrations were calculated using internal and external standards, and expressed as concentrations (ng pigment per L of filtered seawater) or pigment ratios (weight:weight). Trace gas measurements To assess the applicability of upper-ocean mixing estimates, vertical profiles of two dissolved gases; hydrogen (H2) and carbon monoxide (CO) were conducted at 10 m intervals within the upper 80 m of the water column. The sampling procedures and analytical methods were identical to those previously described at Station ALOHA (Blomquist et al., 2012; Wilson et al., 2013). Seawater samples were collected using a rosette of 12-L polyvinyl chloride sample bottles and transferred to acid-washed, combusted, glass stoppered 300 mL Wheaton bottles. Dissolved H2 and CO concentrations were determined onboard immediately using headspace equilibration and all samples were analyzed within 1.5 h of collection. The headspace equilibration was conducted using a 50 mL gas-tight, glass syringe (Perfektum) which was contained within a custom-built syringe actuator to ensure accurate and reproducible volumes were used for each sample analysis. The equilibrated headspace was injected into the inlet of the reduced gas analyzer (Peak Laboratories; Mountain View, CA) as previously described (Xie et al., 2002). H2 and CO were chromatographically separated from other gases using a dual column system ( packed Unibeads 1S 60/80 mesh, 3.2 mm diameter and 42 cm length and Molecular Sieve 13X, 60/80 mesh, 3.2 mm diameter and 206 cm length) maintained at 1048C. The sample gas stream subsequently passed through a heated mercuric oxide bed (2508C) to produce mercury vapor which was quantified using an UV absorption photometer located immediately downstream from the reaction site. The retention time for H2 and CO with a carrier gas flow rate of 20 mL min21 was 45 and 112 s, respectively. The system was calibrated using serial dilutions of a 1 ppmv + 1% H2 gas standard and a 5 ppmv + 1% CO gas standard (Scott Marrin). R E S U LT S A N D D I S C U S S I O N Inhibition of XC activity by DTT The results of the Pelagomonas sp. XC activity experiment are shown in Fig. 1. The dark- and light-control flasks yielded DES values of 0.03 and 0.33, respectively. DES values decreased monotonically at inhibitor concentrations ranging from 50 to 200 mM DTT, and remained low and constant at higher DTT concentrations. These findings are similar to those of Lavaud et al. (Lavaud et al., 2002) and Kashino and Kudoh (Kashino and Kudoh, 2003) who reported that XC activities were completely inhibited at inhibitor concentrations of 200 mM DTT. Based on these results, we used a final DTT concentration of 300 mM to ensure a complete inhibition of XC activity in the seawater samples collected during KM1219. XCP kinetics The results of the DES kinetic rate parameter experiments are shown in Fig. 2a. DES values ranged from 0.13 to 0.51 during these experiments. DESo and DES1 values of 0.00 and 0.59, respectively, were obtained from the chromophyte pigment time-series described below. These results were modeled using first-order kinetics (Griffith et al., 2010) and yielded a rate constant of 0.16 + 0.02 min21 (Fig. 2b). This value falls within the range of those determined in laboratory experiments performed with diatoms (0.09 0.28 min21, Table I). Phytoplankton pigment time-series A time-series of phytoplankton pigment concentrations measured at a depth of 5 m during 23 August7 September 2012 is shown in Fig. 3. While time-dependent changes Fig. 1. Inhibition of XC activity in Pelagomonas sp. by dithiothreitol (filled circle dark control flask and open circle light-exposed flasks). Fig. 2. De-epoxidation state (DES) kinetic experiments performed with natural chromophyte populations sampled from surface waters during KM1219: (a) DES vs. time and (b) LN DES ratio vs. time. 0.16 + 0.02 0.090.13 0.100.20 0.250.28 Chromophytes P. tricornutum C. gracilis P. tricornutum This study Olaizola and Yamamoto (1994) Kashino and Kudoh (2003) Lohr and Wilhelm (1999) Fig. 3. Phytoplankton pigment time-series measured at a depth of 5 m during KM1219 (see text for details). in CHLA, chlorophyll c (CHLC), fucoxanthin (FUCO, diatom biomarker), 190-butanoyloxyfucoxanthin (BUT, pelagophyte biomarker) and 190-hexanoyloxyfucoxanthin (HEX, prymnesiophyte biomarker) concentrations were not significant (P . 0.100), Zeaxanthin (ZEAX, cyanobacteria biomarker) concentrations decreased significantly during the course of cruise KM1219 (P , 0.001). The decrease in the latter was associated with a decline in the abundance of Prochlorococcus spp. as measured by flow cytometry (data not shown). These findings suggest that the chromophytic microalgae populations remained relatively constant during KM1219. At Station ALOHA, the dominant chromophytic microalgae that possess a DDX-DTX xanthophyll cycle include the diatoms, pelagophytes and prymnesiophytes (Andersen et al., 1996). The sum of the photosynthetic carotenoids (PSC), FUCO, BUT and HEX is used as a biomarker for chromophytic microalgae, and XCP is defined as the sum of DDX and DTX. In order to adjust for temporal changes in phytoplankton pigment biomass during KM1219, PSC and XCP concentrations are normalized to chlorophyll a (CHLA). The respective uncertainties of the HPLC-determined PSC, XCP, PSC/CHLA and XCP/CHLA values are +5.7, +6.1, +5.6 and +6.0%, and used to construct the error bars in Fig. 3. XCP concentrations increased during 0700 to 1000, remained constant during 1000 to 1500 and decreased during 1500 to 1900 (Fig. 4a). Concentrations of PSC remained relatively constant during the daylight hours. XCP/CHLA values increased from 0700 to 1100 HST, remained constant during 1100 to 1500 HST and decreased from 1500 to 1900 HST (Fig. 4b). It is interesting to note that XCP/CHLA values were similar on 5 and 6 September despite the two-fold difference in PAR. The maximum XCP/CHLA values (0.180 0.182) were measured during 1200 1400 HST, and the minimum XCP/CHLA values (0.075 0.084) were measured during 0700 0800 HST. These latter values are in good agreement with nighttime observations (5 m, 2000 HST) during June September at Station ALOHA during 1989 2011 (0.079 + 0.004, http://hahana.soest. hawaii.edu/hot/hot-dogs/). PSC/CHLA values remained relatively constant during the daylight hours. Rates of photoadaptation (DXCP/CHLA per unit time dP/dt) were estimated via linear regression analysis of the data in Fig. 4b. The slopes of these data averaged 6.7 + 0.5 1026 s21 (Fig. 5) for XCP/CHLA increases and decreases on 5 and 6 September. It should be noted that these photoadaptation rates may be slightly underestimated (morning samples) or overestimated (afternoon samples) since the surface bucket samples were already subjected to natural mixed layer mixing. Despite this uncertainty, the rates reported here are in very good agreement with those determined during the light shift experiments performed by Harris et al. (Harris et al., 2009) with Emiliania huxleyi (dP/dt 5.6 + 1.4 1026 s21; see their Fig. 3c). Chromophyte pigment vertical profiles Five vertical profiles were conducted using the diaphragm pump during 26 August 1 September, and surface PAR 1 September 1484 27 MLD (m) Fig. 4. Chromophyte pigment time-series measured during KM1219: (a) photosynthetic carotenoid concentrations (PSC, open circle: 5 September and filled circle: 6 September) and xanthophyll cycle pigment concentrations (XCP, open square: 5 September and filled square: 6 September) and (b) chlorophyll a-normalized photosynthetic carotenoid concentrations (PSC/CHLA, open circle: 5 September and filled circle: 6 September) and chlorophyll a-normalized xanthophyll cycle pigment concentrations (XCP/CHLA, open square: 5 September and filled square: 6 September). Note that the irradiance on 5 September (dotted line) was approximately two-fold higher than that measured on 6 September (dashed line). Fig. 5. Rates of photoadaptation measured on 5 September (open circle) and 6 September (filled circle) during KM1219. averaged 1383 + 140 mmol quanta m22 s21 (+10%) during these casts (Table II). Mixed layer depths and wind speeds (U15) ranged from 18 to 45 m and 7.4 to 9.1 m s21, respectively. Vertical profiles XCP/CHLA are presented in Fig. 6. Values are expressed as a function of normalized Fig. 6. Pump profiles of normalized XCP-to-CHLA ratios (Z/MLD sample depth (m)/mixed layer depth (m), dimensionless) measured during KM1219. mixed layer depth (i.e. z/MLD, dimensionless in order to facilitate the comparison of data collected on different days). The observed vertical distributions of XCP/CHLA shown are dependent upon both photoadaptation and mixing processes (Cullen and Lewis, 1988; Lewis et al., 1984), and the water motions responsible for XCP dynamics are collectively referred to as a turbulent transfer velocity (TTV). Estimation of turbulent transfer velocities A summary of the wind-forcing and mixing parameters used to estimate turbulent transfer velocities during KM1219 are presented in Table III. TTV (dz/dt, cm s21) was estimated as the quotient of the XCP/CHLA 1026 s21) and the XCP/ photoadaptation rate (6.7 CHLA gradient in the mixed layer (cf. Fig. 6): TTV dP =dtdP =dz Drag coefficientb ade Boyer Monte gut et al. (2004). bLarge and Pond (1981). Quantity or equation 1 September 27 PAR (5 m) 923.7 With the exception of 26 August, the R2 values for the XCP/CHLA gradient regression analyses ranged from 0.667 to 0.856. The average XCP/CHLA value (0.186) measured in the mixed layer on 26 August 2012 exceeded the light-adapted XCP/CHLA values (0.180 0.182) measured during the XCP time-series, and the mixing rate was assumed to be 0 m s21. For wind speeds .7.5 m s21, TTVs averaged 0.41 + 0.10 cm s21 during H-D 9. Based on these latter rates and the observed mixed layer depths, we estimate the mixing timescale (T) to be 2 4 h. The process of photoadaptation of XCP/CHLA is triggered by a change of light intensity over time (dPAR/ dt) that we did not directly take into account in Eq. (2). In that simplified case, we applied a value of dP/dt calculated from the diel variation of XCP/CHLA at the sea surface (cf. Fig. 5), but we wanted to verify that dPAR/dt from those measurements was comparable to dPAR/dt experienced by the cell being displaced in the mixed layer by the water movement. The temporal variation of light during the diel experiment is given by the difference of PAR between successive data points in Fig. 5, divided by the time distance between those points. We call this first quantity dPAR/dt (diel), and its average value is 0.095 + 0.097 mmol quanta m22 s22. The estimates of the temporal variability of light for cells entrained in the motion of the mixed layer, dPAR/dt (ML), was calculated starting from the average vertical gradient of PAR in the mixed layer, dPAR/dz (ML), and TTV: dPAR dPAR dtML dzML We computed dPAR/dz (ML) as the gradient of light between 5 m and MLD using measurements of the vertical attenuation of PAR in the water column (Table IV). The average value of dPAR/dt (ML) is 0.085 + 0.034 mmol quanta m22 s22, considering just the four data points with non-zero TTV. Since dPAR/dt (diel) is very similar to dPAR/dt (ML), we conclude that the value of dP/dt computed from data in Fig. 5 can be applied for the calculation of TTV from the vertical gradients of XCP/CHLA. DAsaro (DAsaro, 2001) used Lagrangian floats to measure vertical velocities for open waters west of Vancouver Island during January 1995. Vertical velocities (sw) and friction velocity (u*) were tightly correlated and ranged from 0.43 to 3.20 cm s21 and 0.0037 to 0.0276 m s21, respectively (i.e. sw (1.35)0.5 u*). DAsaro (DAsaro, 2001) points out that since each float traverses the mixed layer several times during the averaging period of 3.6 h, s2w corresponds approximately to the mixed layer average of the square of the vertical velocity (w2).The DAsaro floats are ballasted to be slightly buoyant in order to prevent escape from the mixed layer, and do not perform well below a wind speed of 8 m s21 (Eric DAsaro, personal communication). This slight buoyancy difference leads to a non-uniform float distribution in mixed layer. The distribution is skewed toward the surface and becomes more uniform with increasing wind stress. Consequently, the floats oversample the more energetic upper portion of the mixed layer and positively bias the value of sw2/u*2 by a factor of 1.27 for the mixed layer depth average (see text and Fig. 6 in DAsaro, 2001 Fig. 7. Vertical distributions of H2 and CO measured on (a) 26 August and (b) 1 September during KM1219. for details). After correcting for Lagrangian float biases, Eric DAsaro ( personal communication) estimates a vertical velocity of 1 cm s21 at a u* of 0.010 m s21 based on his data set. This value agrees to within a factor of 2 with the average TTV of 0.41 + 0.10 cm s21 measured during this study on 28 August 1 September (n 4, u* 0.010 + 0.001 m s21). How mixed is the mixed layer: example of trace gas distributions The timescales of mixing identified from XCP measurements (2 4 h) are similar to the turnover times of H2 and CO in the mixed layer. The residence time for both H2 and CO is in general less than a day and most likely on the order of several hours (Doney et al., 1995; Wilson et al., 2013). Under stratified, low-mixing conditions, vertical profiles of CO decrease exponentially causing sharp near-surface gradients in concentration. This was exhibited on 26 August when the low average wind speed and mixing rate resulted in a shallow mixed layer depth of 18 m and sharp gradients of H2 (0.24 nM m21) and CO (0.13 nM m21) were observed in the mixed layer (Fig. 7). In contrast, the higher average wind speed and mixing rate observed on 1 September yielded a deeper mixed layer depth of 27 m and weaker H2 (0.02 nM m21) and CO (0.06 nM m21) gradients in the mixed layer. The pronounced sub-surface maximum in dissolved H2 observed on 26 August is unusual with respect to its amplitude, exceeding typical surface ocean concentrations by two three-fold compared with previous measurements of dissolved H2 in the Pacific Ocean (Moore et al., 2009; Wilson et al., 2013). The sub-surface maximum was not observed during the remainder of the KM1219 cruise and its location just below the mixed layer depth indicates a biological production source, most likely a byproduct of dinitrogen fixation (Wilson et al., 2013). In contrast to H2 which is produced during the day and night, the production of CO, by the photochemical degradation of chromophoric dissolved organic matter, is a daytime only phenomenon with concentrations decreasing during the night as a result of microbial consumption (Blomquist et al., 2012; Zafiriou et al., 2008). Ultimately, the vertical profiles of CO are in good agreement with the estimates of mixing derived from the XCP measurements for the 26 August and 1 September 2012, and suggest that they can provide an insight into vertical advection over short timescales. C O N C L U S I O N S A N D F U T U R E D I R E C T I O N S The results of this study reveal that the measurement of time-dependent variations in chlorophyll a-normalized XCP concentrations shows promise for investigating upper ocean mixing processes at Station ALOHA. The use of the diaphragm pumping system allowed the rapid collection of seawater samples on much shorter timescales (minutes) than that observed for changes in XCP concentrations (tens-of-minutes). While the conversion of DDX to DTX can provide information on rates of photoadaptation on timescales of seconds-to-minutes, the rate constant determined for this reaction is much too fast to obtain reliable DES depth profiles with the diaphragm pumping system used in this study. Coincident measurements of H2 and CO profiles suggest that XCP dynamics may be useful for the interpretation of trace gas distributions in the mixed layer. Future studies should be conducted in the winter months when wind speeds are higher and mixed layer depths deeper in order to determine whether a predictable relationship exists between TTV and friction velocity (cf. DAsaro, 2001). The latter would permit a reconstruction of turbulent transfer velocities over the last decade using the WHOTS mooring wind speed data, and provide an additional tool for the interpretation of biogeochemical variations at Station ALOHA. AC K N O W L E D G E M E N T S The authors thank Captain and crew of the R/V Kilo Moana for their valuable support during KM1219. We are grateful to Dan Repeta (Woods Hole Oceanographic Institution) for the use of his diaphragm pumping system, Yoshimi Rii (University of Hawaii) for providing the Pelagomonas sp. culture, Ricardo Letelier and Jasmine Nahorniak (Oregon State University) for access to the vertical measurements of spectral irradiance, Susan Brown (University of Hawaii) for the use of the Percival incubator, Robert Weller and Al Plueddemann (Woods Hole Oceanographic Institution) for providing WHOTS mooring wind speed data and finally Roger Lukas (University of Hawaii) and Eric DAsaro (University of Washington) for providing valuable discussions regarding mixing processes in the upper ocean. F U N D I N G This research was supported by National Science Foundation grants EF-0424599 (D.M.K.) and OCE-1153656 (D.M.K. and S.T.W.), National Oceanic and Atmospheric Administration grant NA09OAR4320129, and the Gordon and Betty Moore Foundation (D.M.K.).


This is a preview of a remote PDF: https://plankt.oxfordjournals.org/content/36/6/1423.full.pdf

Robert R. Bidigare, Fenina R. Buttler, Stephanie J. Christensen, Benedetto Barone, David M. Karl, Samuel T. Wilson. Evaluation of the utility of xanthophyll cycle pigment dynamics for assessing upper ocean mixing processes at Station ALOHA, Journal of Plankton Research, 2014, 1423-1433, DOI: 10.1093/plankt/fbu069