AQUATIC MICROBIAL AND MOLECULAR ECOLGY - 2021 LABORATORY MANUAL Department of Biology University of Southern Denmark
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
AQUATIC MICROBIAL AND MOLECULAR ECOLGY 2021 LABORATORY MANUAL Department of Biology University of Southern Denmark
Workplan for AMME 2021 (AT-Alexander Treusch, BT- Bo Thamdrup, DEC - Don E Canfield, RNG – Ronnie N Glud) Sun 15.08: 14:00 – 18:00 Excursion to field sites and collection of sediment & water RNG+AT 18:00 -- Icebreaker barbecue (ALL?) Mon 16.08 8.30 – 9:00 Introduction – RNG 9:15 – 10:00 Lecture; The role of prokaryote diversity - AT 10:30 – 12:00 Laboratory work 13:00 – 17:00 Laboratory work Tues 17.08 8:30 – 10:00 Lecture; Structure & growth of microbial populations - AT 10:30 – 12:00 Laboratory work 13:00 – 17:00 Laboratory work Wed 18.08 8:30 – 10:00 Lecture; Microbial energetics - RNG 10:30 – 12:00 Laboratory work 13:00 – 17:00 Laboratory work Thur 19.08 8:30 – 10:00 Lecture; Oxygen cycling & dynamics – RNG 10:30 – 12:00 Oral student presentations 13:00 – 17:00 Laboratory work Fri 20.08 8:30 – 10:00 Lecture; The biogeochemistry of nitrogen - BT 10:30 – 12:00 Laboratory work 13:00 – 17:00 Laboratory work Sat 21.08 8:30 – 10:00 Lecture; The biogeochemistry of methane - BT 10:30 – 12:00 Oral student presentations 13:00 – 17:00 Laboratory work Sun 22.08 Off or/and additional sediment sampling Mon 23.08 8:30 – 10:00 Lecture; Carbon cycling - RNG 10:30 – 12:00 Laboratory work 13:00 – 17:00 Laboratory work 2
Tues 24.08 8:30 – 10:00 Lecture; The biogeochemistry of Sulfur - DEC 10:30 – 12:00 Laboratory work 13:00 – 17:00 Laboratory work Wed 25.08 8:30 – 10:00 Lecture; The biogeochemistry of Iron and manganese - DEC 10:30 – 12:00 Laboratory work 13:00 – 17:00 Laboratory work Thur 26.08 8:30 – 10:00 Lecture; Molecular microbial ecology - AT 10:30 – 12:00 Laboratory work 13:00 – 17:00 Laboratory work Fri 27.08 8:30 – 12:00 Student Data presentation 13:00 – 16:00 Report writing & discussions Sat 28.08 8:30 – 12:00 Report writing & discussions 13:00 – 16:00 Report writing & discussions 18.00 -- Joint dinner Sun 29.08 8:30 – 12:00 Report writing & discussions 13:00 – 13:15 Group photo 13:15 – 16:00 Report writing & discussions 3
EACH TEAM’S TIME SCHEDULE FOR THE VARIOUS EXERCISES The codes refer to chapters in the ”Procedures” section of each exercise. KF indicates Kærby Fed, FS indicates Fællesstrand, Date Team 1 Team 2 Team 3 Team 4 Team 5 KF KF KF FS FS Mon 16.08 1A 1A 1A 1A 1A 10:30-17 7 7 7 7 7 8ABC 8ABC 8ABC 8ABC 8ABC Tue 17.08 1B 1B 1B 1B 1B 10:30-17 2ABCD 4AB 6AB 4AB 5.1 & 5.2 Wed 18.08 1C 1C 1CDI 1C 1C 10:30-17 4AB 2ABCD 4CDE 6AB Thu 19.08 6AB 2E 1DII 5.1 & 5.2 1DI 13-17 2ABCD Fri 20.08 4CDE 1DI 2E 2ABCD 1DII 10:30-17 4AB 3AB Sat 21.08 1DI 1DII 3AB 2E 4AB 13-17 5.1A & 5.2 6AB Mon 23.08 1DII 4CDE 5.1 & 5.2 3AB 2ABCD 10:30-17 Tue 24.08 5.1 & 5.2 3AB 4CDE 1DI 2E 10:30-17 Wed 25.08 3AB 6AB 1DII 4CDE 10:30-17 Demo Demo Demo Demo Demo Thu 26.08 8DE 8DE 8DE 8DE 8DE 10:30-17 Fri 27.08 Student data presentation & Report writing Sat 28.08 Report writing Sun 29.08 Report writing 4
CONTENTS General introduction ................................................................................................................6 EXERCISE 1: Determination of sediment characteristics (density, water content, organic content and chlorophyll a content) .................................................................11 EXERCISE 2: Measurement of sediment oxygen, carbon dioxide and dissolved inorganic nitrogen exchange in light and darkness ........................................15 EXERCISE 3: Determination of O2 microdistribution and diffusive O2 exchange in surface sediments ...........................................................................................20 EXERCISE 4: Measurement of sulfate reduction by the 35S technique .................................24 EXERCISE 5.1: Measurement of denitrification in the sediment by the 15N isotope- pairing technique ............................................................................................29 EXERCISE 5.2: Measurement of potential nitrification (ammonium oxidation) in sediment slurry. ..............................................................................................33 EXERCISE 6: Determination of vertical TCO2, DIN (NH4+ and NO2-+NO3-) and SO42- profiles in the sediment ..................................................................................35 EXERCISE 7: Determination of particulate iron profiles in the sediment .............................37 EXERCISE 8: Molecular analysis of microbial community structure and function ..............40 Reference list ........................................................................................................................48 Appendix 1: Sediment collection ............................................................................................50 Appendix 2: Tables for exercise1,2,4,6,7…………………………………………………..51 Appendix 3: How to do tricky calculations.............................................................................62 Appendix 4: Safety regulations for laboratory work...............................................................64 Appendix 5: Manual for report writing ...................................................................................66 5
Introduction GENERAL INTRODUCTION Organic matter enters coastal sediments either by sedimentation of phytoplankton and terrestrial derived particles (sewage and rivers) or via primary production by benthic microalgae (diatoms and cyanobacteria) and macrophytes (algae and seagrasses). Organic particles located at the sediment surface will gradually be buried within the sediment by continued sedimentation and faunal particle reworking (bioturbation). The rate by which the burial occurs depends on the sedimentation/erosion regime and the density of benthic macrofauna at any specific location. Organic matter in sediments is degraded (mineralized) by an array of aerobic and anaerobic microbial processes with a concurrent release of inorganic nutrients. The actual rates of decay depend primarily on organic matter quality (i.e., the content of protein, cellulose, lignin etc.), age (decomposition stage) and temperature (season). The chemical composition of organic matter in marine environments can be generalized by the following formula: (CH2O)x(NH3)y(H3PO4)z where x, y and z vary depending on the origin and age of the material. Decomposition in the upper part of the sediment occurs under aerobic conditions with oxygen as electron acceptor by the following stoichiometry: (CH2O)x(NH3)y(H3PO4)z + xO2 + yH2O → xHCO3- +yNH4+ + zPO43- + (x+3z)H+ + yOH- As the oxic (oxygen containing) zone in coastal sediments usually is limited to the upper 1-5 mm, a large fraction of the organic matter will be buried more or less undecomposed into anoxic layers. In these layers the large and normally complex organic molecules will be split into smaller and water-soluble units under the production of energy by hydrolyzing and fermenting bacteria. These smaller units will then be degraded completely by an array of microorganisms using a variety of oxidized inorganic compounds as electron acceptors. These processes generally occur in the following sequence with depth in the sediment: Mn4+ ≈ NO3-, Fe3+, SO42- and CO2 respiration (Fig. 1). The actual sequence is determined by the ability of each electron acceptor to receive electrons, and thus the energy output per degraded organic carbon atom. Most of the major microbial processes are carried out by specific groups of organisms, each of which is specialized in one or a few processes only. No single organism can catalyze all the different processes. Oxygen-respiring bacteria may be facultative aerobes that grow by denitrification or fermentation when oxygen is absent. Most manganese- and iron-reducing bacteria, on the other hand, as well as all sulfate-reducing bacteria and methanogenic archaea are strict anaerobes and are specialized in their use of electron acceptor. Furthermore, the strict anaerobes cannot oxidize complex organic matter but rely on fermenting bacteria to produce their substrates, of which acetate, other short-chain fatty acids, and hydrogen are the most important. The sequence of anaerobic processes observed in the sediment (Fig. 1) is the 6
Introduction result of competition between the different respiratory groups for these common substrates. The transition from one electron acceptor to the other downwards in the sediment will occur when the most favorable is exhausted. ”When the best is gone, one has to accept something less good.” Nevertheless, the usually observed decreasing degradation rate with depth in sediments is not caused by the less efficient electron acceptors in the deeper layers, but rather by decreasing quality of the organic matter (lability or degradability) with depth. However, the variety of organic substrates which can be degraded seems to be narrower when moving through the sequence of electron acceptors from O2 to CO2. RESP Concentration (arbitrary scale) Oxygen O2 Mn 4+ Mangan. - NO3 Nitrate 3+ Fe Iron Depth in sediment 2- SO4 Sulfate - HCO3 Methanogen. Figure 1. Idealized depth distribution of electron acceptors in marine sediment. The sequence of respiration processes is indicated along the left vertical axis. The quantitative significance of each of the various aerobic and anaerobic processes for the total decomposition in any sediment is determined by several local conditions, such as sedimentation rate, bioturbation rate, temperature etc. Oxygen respiration and sulfate reduction are usually considered to contribute by up to 50% each. However, the latter process can be responsible for up to almost 100% in organic-rich sediments. A process like denitrification is usually of limited quantitative significance for mineralization (few percent) due to low availability of NO3-. Although sulfate reduction is one of the least favorable respiration processes, the high concentration of SO42- in seawater (300-1000 times higher than 7
Introduction O2) is responsible for its deep vertical distribution and thus quantitative importance. Inorganic C, N and P compounds (HCO3-, NH4+ and PO43-), which are released by the mineralization processes, will accumulate in the sediment porewater and gradually be transported upwards by molecular diffusion (or bioturbation). Eventually they will all end up in the overlying water column and once again act as substrate for primary producers (plants). A large fraction of sediment oxygen uptake is not caused by aerobic respiration (decomposition), but rather by reoxidation of reduced inorganic metabolites (e.g., H2S and NH4+). About 50% or more of the sediment oxygen uptake is usually consumed by reoxidation, which occurs when reduced compounds diffuse from deeper layers and up to the oxic sediment surface. Reoxidation can be a pure chemical process, but usually it is mediated by microorganisms. The oxidation of reduced compounds releases energy, which drives the metabolism of chemoautotrophic Bacteria and Archaea. The most important of these processes in sediments are nitrification and sulfide oxidation. Although the latter process is quantitatively most important in terms of oxygen uptake, only the former process will be dealt with here, since it is very important for nitrogen cycling in sediments. Nitrification is generally a two-step process driven by aerobic Bacteria and Archaea. First NH4+ that is released during ammonification of the organic matreial if oxidized to NO2- that subsequently is oxidoixed to NO3-. The produced NO3- will diffuse either downwards into the sediment or upwards to the overlying water column. If downwards, NO3- will typically be denitrified to N2 in anoxic layers and thus become unavailable for most living organisms. If upwards, NO3- will be an essential nutrient for primary producers. About 30-50% of all NH4+ being produced in sediments will normally be lost by coupled nitrification-denitrification. Despite its limited influence on the carbon cycling, denitrification is very important for the nitrogen cycling in sediments. Until very recently it was difficult to obtain reliable information about the nature of the microorganisms responsible for the chemical processes in sediments. Unlike animals and plants, most bacteria and archaea cannot be identified simply by their size and shape. Only a minority of the microorganisms present in a particular environment can be cultivated using currently available techniques. This makes a comprehensive description of bacterial diversity in the environment using classic microbiologic methods impossible. The solution to this problem is the use of molecular sequence information to investigate microbial diversity and identify prokaryotic species, an approach known as “molecular microbial ecology approach”. The sequence of DNA bases in a particular region of the genome is characteristic of a particular type of microbe and can be used for identification. The most commonly used genes are those encoding for the ribosomal RNA molecules, which are structural components of the ribosomes, particularly the 16S rRNA of the small subunit. This is certain to be present in all bacterial cells whatever their type of energy metabolism or ecological niche. The purpose of these laboratory exercises is to quantify processes driven by the most important microorganisms and their diversity in 2 different coastal sediments. Sediment will be obtained from a relatively organic-rich site, Kærby Fed (KF) in Odense Fjord, and from an organic-poor site, Fællesstrand (FS) at Fyns Hoved. You will be split into 5 teams - team 1-3 will work with Kærby Fed sediment and team 3 and 4 with Fællesstrand sediment. During the 8
Introduction exercises, oxygen, carbon dioxide and dissolved inorganic nitrogen (DIN) exchange caused by microalgal primary production and heterotrophic respiration in the sediment will be measured (exercise 2). Subsequently, the small-scale dynamics of oxygen near the sediment- water interface in light and darkness will be illustrated and related to microalgal biomass (exercise 1 and 3). The rates of oxygen uptake and carbon dioxide release in the dark will be related to rates of sulfate reduction (exercise 4) and denitrification (exercise 5A) in order to determine the quantitative role of aerobic respiration, denitrification and sulfate reduction (see Appendix 3). The distribution of oxidized and reduced particulate iron pools (Fe3+ and Fe2+) will be examined (exercise 7) and compared to show the influence of processes related to redox dynamics of iron. The oxidized iron pool will also be used to quantify the role of iron reduction for the total sediment metabolism (see Appendix 3). Exercise 6 (porewater profiles of TCO2, dissolved inorganic nitrogen and SO42-) and exercise 1 (porosity and organic content) will contribute with essential data for the final calculation and interpretation of results obtained in the other exercises. Potential nitrification will be determined in exercise 5B and related to the results from exercise 8, where the distribution of marker genes for ammonia oxidation will be quantified. Exercise 8 illustrates how molecular techniques based on the analysis of environmental DNA can be used to obtain information about microbial communities, avoiding the need to culture the microbes. We will be using quantitative PCR (qPCR) to determine the metabolic potential of the community for different processes in the nitrogen cycle, namely nitrification and denitrification. By quantifying the copy numbers of the functional marker genes for nitrification (ammonia monooxygenase; amoA gene) and denitrification (nitrite reductase; nirS gene) in parallel to the 16S rRNA gene we will be able to estimate what percentage of the community has the capability for performing these processes. In order to be able to run this analysis, we will extract metagenomic DNA (also called community DNA) from sediment samples. Here we will use a special method that avoids the co-extraction of contaminating substances like humic acids, resulting in reasonably pure metagenomic DNA usable for downstream molecular work. The marker genes will subsequently be amplified by PCR and the amount of PCR product will be determined during each amplification cycle using a double-stranded DNA binding fluorophore and a detection system in the qPCR machine. Using this information will allow us to calculate back to the copy numbers of the marker genes in the original sample and compare them with the potential nitrification measurements from exercise 5,2 and the potential denitrification rates from exercise 5,1. You will find a precise time schedule for the work of each team during the various exercises in the front of this manual (Appendix 1). The laboratory setup for the experimental work has been done in advance, and each exercise has a specific location in the laboratory and is indicated by its number. Please follow the instructions and avoid moving equipment around in the laboratory without informing the instructors. The sediment cores to be used during the course will be sampled by the students on selected dates under guidance of the teachers. Make your own choice of when to participate in this essential part of the course, but everyone must go at least one time. Please read the safety 9
Introduction instruction sheets (Appendix 4) – particularly the instructions on safe handling of radioactivity before performing exercise 4. All data obtained in exercise 1, 2, 4, 6 and 7 should be filled into the appropriate tables found in Appendix 2. They must be sent by e-mail to the teachers when completed. At the end, the results obtained by all teams will be available at the course web-site for your use when writing the final report. Reports should be submitted as team-reports. All members of the teams must participate on equal terms during laboratory work and result treatment as well as during report planning, writing (guidelines for writing a report is given in Appendix 5) and oral presentation during the final synthesis. 10
Exercise 1 EXERCISE 1: Determination of sediment characteristics (density, water content, organic content and chlorophyll a content) Water content, density, organic content, and chlorophyll a content of sediments are important supporting parameters to fully understand many aspects of microbial ecology in sedimentary environments. When related to sediment density, the water content describes the volume of pore spaces in the sediment (porosity, φ = volume water/total volume). This parameter is usually used when microbial rates have to be transformed from per weight unit to per unit volume of sediment. Raw data from laboratory experiments are, of practical reasons, usually determined per unit weight, but they have a better ecological meaning when presented per unit volume. Organic content (loss on ignition) provides an estimate of the potential substrate for decomposing bacteria. There is, however, rarely a simple relationship between organic content and sediment metabolism. Sediment metabolism is more related to the quality and degradability of the organic matter. Most of the organic matter determined as loss on ignition is actually composed of refractory organic compounds of low degradability. It is important to discriminate between the “organic matter content” and the “organic carbon content” which sometime is mixed up in the literature. Chlorophyll a content is a measure of microalgal biomass (diatoms and cyanobacteria) and should be related to benthic primary production as measured in Exercise 2. The most important input of labile organic matter at the two study locations is from benthic microalgae. The biomass of benthic microalgae will here be quantified as chlorophyll a content in the uppermost sediment layers. Chlorophyll a can, however, be relatively persistent to degradation and pigments present in deeper layers are not expected to contribute to the photosynthetic activity. I. Materials: 2 sediment cores from the study sites (5 cm i.d.) Slicing plates + stands + rulers + pistons Alu-trays + alu-foil 4 cut-off syringes (5 ml) 8 centrifuge tubes (15 ml plast) 90% ethanol 2 M HCl Plast spoons Plast cuvettes Pasteur pipets Balance Centrifuge Spectrophotometer Oven, trays from oven 16 crucibles Muffle furnace. Desiccator + separator plates 11
Exercise 1 Vortex Pencils II. Procedures. A. Slicing cores and determination of density Slice two sediment cores, which has been taken with 5 cm i.d. core liners at the selected locality, one at the time in the following intervals: 0-1, 1-2, 2-3, 3-4, 4-5, 5-6, 6-8, 8-10 cm. Place the piston in the core liner (tight, to avoid loss of porewater) instead of the lower rubber stopper (the upper stopper has to remain tightly in place). The core is then fixed on the stand and the upper stopper is removed (Fig. 2). The first 6 slices are cut in 1 cm intervals and the remaining 2 slices in 2 cm intervals by the use of the ruler and cutting plate. The slicing is done by carefully pushing the sediment core inside the liner up to the chosen thickness as indicated by the ruler. Then the cutting plate is pushed along the upper edge of the core liner leaving the intact sediment slice on the cutting plate (use a finger to prevent the slice to slide of during cutting). The sediment slice is then transferred into a pre-weighed and -labeled (with a pencil) alu-tray, which is kept ready. Before the next slice is pushed up, the cutting plate should be cleaned with a paper towel. Cut 4 slices at the time, and handle these as described below - before cutting the next 4 slices. This will minimize evaporation, and thus reduce errors in the water content results. The density of the sediment is determined as the weight of a known volume. Fill a pre- weighed and -labeled 5-ml cut-off (i.e., the tip is cut off) plastic syringe with sediment to the 4-ml mark (avoid any air in the syringe) and weigh it again. Density is calculated as follows: d = m/4 (g cm-3), where m is the sediment weight (i.e., final weight minus syringe weight). Weigh the alu-tray + the remaining sediment and place it in an oven at 105ºC. Ruler Cutting plate Sediment Core tube Piston 12
Exercise 1 Figure 2. Schematic drawing of set-up for core slicing B. Water content When the alu-trays have been in the oven for about 12 hours (the teachers will take them out), they are transferred to a desiccator. Weigh the alu-trays with the dried sediment. Water content (β) is calculated from wet weight (ww) and dry weight (dw) as follows (remember to subtract the weight of the alu-tray): β = [(ww - dw)/ww] ⋅ 100 (%) Porosity (φ) is calculated according to: φ = β ⋅ d/100. Transfer subsamples of the dried sediment to pre-weighed and -labeled crucibles (do not fill them more than 2/3). Weigh the crucibles + sediment and placed it in a muffle furnace at 520ºC. C. Organic content When the crucibles have been in the muffle furnace for 10 h (the teachers will take them out), they are transferred to a desiccator. Weigh all crucibles + sediment and dump the sediment in the waste bucket. Organic content or loss on ignition (LOI) is calculated from dry weight (dw) and ash weight (aw) as follows (remember to subtract the weight of the crucibles): LOI = [(dw - aw)/dw] ⋅ 100 (%) D. Chlorophyll a content We use a rather crude spectrophotometric method, which is easy to perform and reliable. It may not give the exact concentration of chlorophyll, but it is excellent for comparisons between sites. Slice two cores into the following depth intervals: 0-0.5, 0.5-1, 1-2 and 2-3 cm. Mix and homogenize the slices. Transfer a subsample (ca. 0.5 g each) from each depth interval into a test tube containing 5 ml of 90% ethanol, which is placed on a balance (remember to zero the balance before adding the sediment) and weigh the sediment sample. Shake or whirly-mix the test tubes for about 30 seconds. Extract the sediment in darkness (5°C) until next day. Shake the test tubes regularly. When the extraction has finished, the test tubes are centrifuged for 10 min at 3000 rpm. Measure the absorbance of the supernatant at 665 and 750 nm before and after acidification by one drop of 2 M HCl and mixing. The chlorophyll a content is then calculated as follows (Castle et al., 2011): Chlorophyll a (μg/g) = 29.6(665o – 665a) v/m Where 665o is the absorbance at 665 nm subtracted by the absorbance at 750 nm before 13
Exercise 1 acidification; 665a is the absorbance at 665 nm subtracted by the absorbance at 750 nm after acidification; v is the volume of ethanol (5 ml); and m is the weight (g) of extracted sediment. Plot the chlorophyll a concentration as a function of depth in the sediment. Compare the concentrations between the two locations. III. Questions and Figures, which should be considered in writing the report 1. Plot the porosity, the LOI and Chl a as a function of the sediment depth for both sites. 2. Compare the two sites and discuss potential differences. 3. Why is it important that there is no air in the syringe during measurement of density? 4. Why should samples always stay in a desiccator after drying or combustion? 5. Why are samples for organic content combusted at 520ºC? 6. Why do we measure absorbance at both 665 and 750 nm for chlorophyll analysis? 7. Give possible sources of errors. 14
Exercise 2 EXERCISE 2: Measurement of sediment oxygen, carbon dioxide and dissolved inorganic nitrogen exchange in light and darkness. Sediment primary production and total heterotrophic metabolism is determined by measuring oxygen and carbon dioxide exchange across the sediment-water interface. At the same time, the exchange of dissolved inorganic nitrogen (DIN = NH4+, NO2- and NO3-) between the sediment and the overlying water is quantified. DIN flux across the sediment-water interface is a net measure of all nitrogen transformations occurring within the sediment. I. Materials: 6 sediment cores from the study site (5 cm i.d.) 50 l seawater from the study site 1 tank with core holder and central magnet Air pumps with tubing and air stones 1 greenhouse lamp Alu-foil 6 magnet collars and 6 transparent lids with O2 patches 60 ml syringe + tubing 0.45 µm disposable syringe filters Computerized O2 measuring set up ”Flow injection/diffusion cell“ analyzer. 2 mM HCO3- standard Thermometer. 12 ml Exetainers (6) Salinity-meter (conductivity). 20 ml scintillation vials (for nutrient samples) + vial holder. Plast test tubes Flow Injection Analyzer (Lachat). Plast bags. Light sensor II. Procedures. A. Sediment cores Sediment cores have been taken with 5 cm i.d. core liners at the location and pre-incubated in the storage tank at the selected temperature (15ºC) in 12 h light : 12 h dark cycles (light will be kept off today until start of the incubation). ). 6 cores have been collected for flux incubation with a sediment dept of 15-20 cm. Wash hands before or wear gloves when handling cores in the incubation tanks. Describe the sediment of each core with respect to colour zonation and any traces of animal activity (i.e. burrows of Nereis diversicolor). This can be done before or after the incubation. Wrap 3 of the cores in alu-foil (dark cores),but leave the top open until initial sampling has been completed. Submerge these cores in the core holder around the magnetic stirring unit of the incubation tank together with the 3 unwrapped cores (light cores) such, that the water surface is at least 2 cm above the upper edge of the core liners. The light cores must be placed directly under the greenhouse lamp. Be sure that 15
Exercise 2 there is continuous aeration in the tank. Measure and note water temperature, light level and salinity (with conductivity detector). B. Dark and light flux incubation After turning the light on, initial (start) samples (20 ml) are taken from the water phase inside each of the 6 cores while still submerged using a 60-ml syringe for analysis of CO2 and DIN (Dissolved Inorganic Nitrogen). Fill a 12-ml Exetainer (carbon dioxide analysis, see later). Fix a syringe filter to the syringe and filter 15 ml sample to a 20-ml scintillation vial (DIN analysis, see later). Store the CO2 samples at 5ºC and freeze the DIN samples immediately. After finishing the initial sampling, place stirring collars in the middle of the water of each core and seal the cores air tight with a transparent lid holding a small patch of oxygen sensitive phorphyrin-derivate fixed to the inside (see later for explanation) on top of the core liner (do it under water) and stick the glass stopper in the hole in the lid and place the black cover on the three dark cores. Note the start time. IMPORTANT: The entire procedure is conducted for one core at a time, i.e., when the start sample has been taken from one core, it is sealed before sampling is started on another core. Terminate the incubation for one core at a time after 2-3 hours and take the final samples for CO2 and DIN as described for the initial samples - BUT with the exception that sampling is done through the small hole in the lid. Note the precise time at the end of incubation for each core. Measure and note temperature and the height of the water phase in the cores. C. Quantifying abundance and biomass of Nereis diversicolor The respiration contribution to O2 uptake and CO2 production by the most common burrowing benthic fauna species, Nereis diversicolor, will be estimated from the abundance and biomass in flux cores. All cores are therefore sieved through a 1 mm mesh to recover the worms. Recovered worms are transferred to petri dishes (one per core). Worms are counted and weighed. The weighing is done by blotting worms briefly on a paper tissue to remove excess water before transferring them one by one to a petri dish containing seawater, which is placed on a balance. Respiration (V, µmol h-1) as a function of individual biomass (M, g wet wt.) for N. diversicolor (15°C) is: VO2 = 2.15 M0.68 Use the average individual biomass (M) and multiply by the density (m-2) of N. diversicolor to get the contribution on an area basis. D. Photochemical oxygen determination Oxygen will be measured by the optode approach. Central for the technique is the ability of oxygen to quench the fluorescence of certain chemical compounds. In other words, O2 can absorb the energy that otherwise would have been sent out as fluorescence from a given fluorophore. There exist numerous configurations, chemistries and procedures to quantify O2 availability by this approach (e.g. Klimant et al. 1997, Kuhl & Polerecky 2008) and many are 16
Exercise 2 custom build and optimized for specific experimental approaches and procedures – some of these will be demonstrated during the course. The system applied for flux measurements is a commercial available system that is sold by the company Pyroscience. The system is called Firesting and has 4 channels that can be applied for simultaneous measurements. Small patches of an immobilized phorphyrin-derivate are fixed to the inside of the lids to the sediment cores. This compound is excited by red light and the O2 sensitive emitted light has an optimum in the near infrared region. The excitation light is supplied to the sensor patch via a fiber placed in the small hole of the lids. The same fiber also guides the O2 sensitive light back to the measuring instrument. The sensors are pre-calibrated and the O2 signal (in % air saturation) over time during the incubation is directly read out on the computer screen and can be noted and/or saved as a text file. Measure temperature and salinity for conversion from % air saturation to concentration in µM (a small table will be provided ). E. CO2 analysis (Hall & Aller 1992) TCO2 will be analyzed on a ”flow injection/diffusion cell“ analyzer. Samples (0.1 ml) containing CO2 are injected into a carrier stream of 45 mM HCl which is pumped into one side of a diffusion cell equipped with a gas permeable membrane (plumbers tape). A receiver stream of 10 mM NaOH is pumped simultaneously to the other side of the membrane. All CO2 from the acid carrier side diffuses into the base receiver side. The amount of CO2 is then detected as a reduction in conductivity (by a conductivity detector) as a result of base neutralization in the receiver stream. The signal is recorded and peak height measured by a computer. Signals will be evaluated by the program Clarity go to “Analysis” and choose “set zero”, then chose “run single” which activates the data collection. The time range (0-20 min) and signal (-1 -10mV) can if needed be changed during a run. When the baseline is stable you are ready to inject standards and samples. For the samples use a 5 ml syringe with a small piece of tubing to take samples from the exetainers. With the valve in the “load” position inject l ml into the sampling port to fill the 0.1 ml sample loop. Now turn the injection port to the ”inject“ position. Leave the syringe in the sampling port. The CO2 peak appears within 1 min, and when the baseline is stable again, inject sample once more by turning the injection port back to “load” and “inject” positions, repeat 3-4 times. Stop run by the stop icon (menu), which opens the data collection site. Check if baseline is okay or change it manually, note peak height. Close date collection in the upper right corner and start the next injections. The concentration of TCO2 is calculated manually based on the 2mM standard. After measurements flush the system with Milli-Q for 15min (place tubing in Milli-Q), stop pump and open clamps, turn off conductivity meter, empty waste containers), refill HCl and NaOH bottles and shut down the computer F. DIN analysis Analysis of NH4+, NO2- + NO3- is done spectrophotometrically on a LACHAT Flow Injection Analyzer (the exact procedure will be explained during the exercise). NH4+ is analyzed according to the Berthelot-reaction (Bower & Holm-Hansen 1980). In a weak basic solution (pH 9.5-11), NH3 reacts with hypochlorite to produce monochloramine, which in turn 17
Exercise 2 provides indophenol blue in the presence of salicylate, nitroprusside-ions (catalyst) and surplus of hypochlorite. The automated NO2- analysis is based on a reaction between the NO2- ion and sulfanilamide under acid conditions. This produces a diazo-compound which is coupled with N-(1-naphthyl)-ethylene-diamine to provide a purple azo-dye (Armstrong et al. 1967). The absorption of the color reaction is determined at 550 nm. The analysis of NO3- is based on a reduction of NO3- to NO2- by a copper-cadmium column followed by the analysis as described above for NO2-. Procedure: Transfer samples (in the appropriate dilution) to the small autosampler cups and place them in the autosampler. From here, they automatically will be mixed with reagents and measured in a spectrophotometer one by one. The concentrations (in μM) are calculated from a relationship based on the peak areas of 5 known standards (NH4+: 0, 10, 20, 30, 40 μM; NO3-: 0, 5.0, 10.0, 20.0 μM). NO2- and NO3- will not be determined separately here because the concentration of NO2- is very low. Results should therefore be presented as NO2- + NO3- Calculation of fluxes: Oxygen exchange per m2 sediment surface per day (JO2, mmol m-2 d-1) is determined from the slope (α) of a linear regression of O2 against time (days) according to: Flux = slope ⋅ water volume / surface area JO2 = α ⋅ π r2 h / π r2 -2 -1 mmol m d = mmol m-3 d-1 ⋅ m3 / m2 This can be reduced to: JO2 = α ⋅ h where the slope α has the units of μM d-1 (μmol dm-3 d-1 = mmol m-3 d-1) and h is the height of the water column above the sediment in m (or cm/100) Carbon dioxide (JCO2) and DIN exchange (JDIN) per m2 sediment surface per day (mmol m-2 d-1) are calculated according to: Flux = change in concentration ⋅ water volume / surface area / time Jx = (Ce – Cs) ⋅ π r2 h / π r2 / t (values) -2 -1 -3 3 2 mmol m d = mmol m ⋅ m / m / d (units) This can be reduced to: Jx = (Ce – Cs) ⋅ h / t where Ce and Cs are end and start concentration in μM (μmol dm-3 = mmol m-3), h is the height of the water column above the sediment in m (or cm/100) and t is incubation time in days (or hours/24). III. Questions & Figures, which should be considered in the report 1. Provide a Table and a diagram of the total exchange rates of O2, DIN and TCO2 for the two sites. 2. Calculate RQ (respiratory quotient) and PQ (production quotient) from the two sites and explain what this tells us about the two sites (see also appendix 3)? 3. What is the contribution of Nereis respiration to the total O2 uptake (TOU) measured in darkness at the two sites. 18
Exercise 2 4. Calculate the CO2/DIN flux ratio for the two sites and explain what it tells us? 5. Which processes within the sediment determine the net flux of DIN? 6. Why is it necessary with stirring during flux incubations? 7. What can be the cause for observed differences in exchange rates among cores from the same site? 8. Give possible sources of errors. 19
Exercise 3 EXERCISE 3: Determination of O2 microdistribution and diffusive O2 exchange in surface sediments Photic surface sediments are characterized by very steep O2 gradients reflecting concurrent production and consumption of O2. Fast microsensors are required to resolve the distribution and the temporal dynamic of O2 in such environments – in this exercise we use Clark-type O2 microelectrodes (Revsbech & Ward 1983). The microprofile approach makes it possible to quantify O2 exchange and penetration in darkness and light, in order to calculate the O2 consumption in darkness and the net photosynthesis in light. Give some assumption it is in principle possible to calculate the gross primary production from such pared datasets. It is, however, also possible to measure the gross photosynthesis directly by the so-called light-dark shift technique (Glud et al. 1992), but that will not be done in the current exercise. I. Materials: 1 sediment core Complete automated microelectrode measuring set up Adjustable light source 2 stands with hooks Light logger Diagram for diurnal down-welling irradiance Air-pump with tubing and air-stone Software for O2 solubility Table for diffusion coefficients Flow-through chamber Aquaria pump II. Procedures: A. Dark O2 penetration and diffusive mediated exchange: Experimental: The microelectrode tip is placed 1-2 mm above the sediment surface. Subsequently the sensor is lowered in steps of 50 µm (0.05 mm), while the O2 concentration at each depth is recorded and stored by the computerized set-up. Note when the sensor enters the sediment. Repeat the measurements until you record low stable signals (i.e 0-3 mV) that remain constant when you move the sensor - this is defined as the anoxic zone. The sensor is moved back to the start position, moved 3-5 mm horizontally and the measuring procedure is repeated. You should measure a minimum of three profiles – but are welcome to measure more. Calculations and theoretical work: The recorded text-files are transferred into Excel and here the first step is to calibrate the recorded signals into µmol L-1. Oxygen sensors have a linear 20
Exercise 3 response and as we know the O2 concentration in the overlying water phase at the given temperature and salinity and the signal size in the anoxic (zero oxygen) sediment layers, it is simple to calibrate each of the recorded data points according to: Calibrated signal (µmol L-1) = Sr (Cw /(Sw-Sa)) + Sa(Cw /(Sw-Sa)) Where Sr is the recorded signal (in mV), Sw the signal in the overlying water phase, Sa the signal in anoxic sediment and Cw is the O2 concentration in the overlying water phase. Define the relative position of the sediment surface. Assign this position to depth “0” and use negative prefix for recordings at any depths below this fixed point (see Fig 3). Plot the respective microprofiles with “Depth” (mm) on the ordinate and O2 concentration µmol L-1) on the abscissa. Determine the average O2 penetration (P) (unit: mm) and calculate the average Diffusive O2 Exchange (DOE – Jdark,up) from the profiles using: DOE (mmol m-2 d-1) = -Do (dC/dZ)*8640 Where Do is the molecular diffusion coefficient (cm2 s-1) at the given conditions (see Table 1) and (dC/dZ) is the slope of the linear O2 concentration gradient in the Diffusive Boundary Layer (DBL) (µmol L-1 mm-1). Figure 3. Two microprofiles measured in sediment cores recovered from a water depth of 1.2 m in Helsingør Harbour, Denmark. The profiles were measured at the exact same spot in darkness and at a irradiance of 600 µmol photons m-2 s-1. The white horizontal bars indicate the gross photosynthetic rates at each depth measured by the light-dark-shift approach (not measured in this exercise). The temperature was 1 ºC and the salinity 12. 21
Exercise 3 B. O2 penetration and diffusive mediated exchange as a function of light: The experimental procedure in II.A is repeated three times at 4 different irradiances (i.e. 10, 50, 100. 300, µmol photons m-2 s-1). Start with the lowest irradiance. Calibrate and plot the respective profiles as above. Determine the average O2 penetration (P) (mm) at the respective light levels. Calculate net photosynthesis (Net-P, mmol m-2 d-1) in the photic zone for each light level using a modified version of the DOE equation above (see also Fig 3B): Net-P = DOE (Jlight,up) = ((Do (dC/dZ)) *8640 III. Questions and figures to be considered in the report: 1. Present one set of profiles including the one measured in darkness and at the four respective light levels for both sites. See Fig 3 for units and layout. 2. Describe and discuss the shape of the profiles. 3. Present the derived O2 penetration depth and the DOE/NetPP as a function of the light intensity for the compiled datasets from the two respective sites. What can be learned from such a figure? 4. How does DOE measured in darkness compare to the total oxygen uptake (TOU) measured during intact core incubations? Discuss potential reasons for any differences. 22
Exercise 3 5. Is the sediment net heterotrophic or autotrophic when integrated over a normal 24h day/light cycle? 6. Discuss if the resolved benthic primary production is important on system level. 23
Exercise 4 EXERCISE 4: Measurement of sulfate reduction by the 35S technique The purpose of this exercise is to determine the rate of sulfate reduction in the sediment. Reduction of sulfate to sulfide is followed by the use of the radioactive isotope 35S. Labeled sulfate in the form of 35S-SO42- is injected into the sediment (Jørgensen 1978). The rate of sulfate reduction can be determined when the ratio between the added amount of labeled sulfate and the amount of sulfur recovered as labeled reduced sulfur compounds is related to the total concentration of sulfate in the sediment. Radioactive sulfide is incorporated into various reduced sulfur pools in sediments. In order to release these pools, we will distill the sediment according to the TRIS technique (Total Reduced Inorganic Sulfur). By this technique, both the acid volatile sulfides and the chromium reducible sulfur will be released simultaneously (Fossing & Jørgensen 1989). Acid volatile sulfides (FeS) are converted to H2S by the addition of hydrochloric acid (HCl). To release the chromium reducible sulfur (FeS2 and S0) in the form of H2S, on the other hand, a reducing agent (chromium II) is needed. The following process describes pyrite (FeS2) reduction: FeS2 + 2 Cr2+ + 4 H+ → Fe2+ + 2 H2S + 2 Cr3+ I. Materials: 2 sediment cores in tubes with 1 injection port per cm. (2.6 cm i.d.) Stand + holder for tubes Cutting plate + ruler + piston 2x6 plastic centrifuge tubes (50 ml) Centrifuge Balance Whirli mixer Spectrophotometer Cuvettes (1 cm) Gloves Scintillation counter 2 x 6 reaction flasks + various for distillation Beaker for dist. H2O. Long test tubes 20 ml scintillation vials. 7 ml scintillation vials. Vial holders 10 ml plast centrifuge tubes Eppendorf tubes Waste container Plastic sheet for covering table 60 ml syringe 5 ml Finnpipette + tips 1 ml Finnpipette + tips 24
Exercise 4 200 μl Finnpipette + tips 50 μl Hamilton syringe Long and short pasteur pipettes Radiotracer, 35S-SO42-, carrier free 12 kBq µl-1 20 % og 5% (w/v) Zinc acetate. 6 M HCl Chromium solution (1 M Cr2+ in 2 M HCl). Scintillation liquid Cline’s reagent (“low Cline”). Silicone oil (anti foam) II. Procedures (This exercise is done in a marked and approved area. Only persons participating in the exercise are allowed in this area. Lab coats will be handed out and gloves must be worn) Fill out Table 8 in appendix 2 as you progress! A. Incubation (Before you start read the radio-safety instruction sheet - Appendix 3) Two sediment cores have been sampled using core liners (2.6 cm i.d.) equipped with silicone sealed injection ports for every cm. Mount the cores on the lab-stand and remove the overlying water. Inject 5 μl of a 35S-SO42- solution (60 kBq) in every injection port down to 10 cm depth by leaving the tracer as a horizontal line while pulling the syringe and needle backwards. Allow the cores to incubate for 4-5 hours in darkness at the same temperature as the flux cores (Exercise 2). Note the time of start and temperature. B. Core sectioning Label and pre-weigh (with lids) 2x6 plastic centrifuge tubes (50-ml). Add 5.0 ml (1 cm slices) or 10.0 ml (2 cm slices) of 20% ZnAc (density 1.03 g/ml) and re-weigh the tubes. Mount the incubated cores on the lab-stand. Cut 6 sections (0-1, 1-2, 2-4, 4-6, 6-8, 8-10 cm) as described in Exercise 1 and transfer them to the centrifuge tubes. Clean the cutting plate with paper towels between each slice. Cap the tubes immediately and mix the content by shaking before weighing the tubes again. Note the time of sectioning. Finally, all samples are stored frozen. Transfer the remaining sediment to the plastic bag labeled radioactive waste. C. Distillation Thaw the frozen samples in the centrifuge tubes and centrifuge them at 3000 rpm for 10 minutes. Transfer 200 μl of the supernatant into a 7-ml scintillation vial (label it on the lid!) for determination of 35S-SO42- activity, and then add 800 μl of distilled water. Scintillation liquid will be added later (see below). Transfer an additional 1 ml supernatant to 1.5 ml Eppendorf tube (a spare sample in case of problems with the original) before decanting the remaining supernatant to the waste container. Weigh the centrifuge tubes with sediment again after removing the supernatant. 25
Exercise 4 While the thawing proceeds, the ZnAc traps for retaining H2S are prepared (H2S is precipitated as ZnS). Add 10 ml of 5% ZnAc into 6 long test tubes together with one very small drop of silicone oil (prevents foaming). Mount the traps in the distillation unit. Mix the sediment in the plastic centrifuge tubes (after centrifugation and decanting the supernatant) and transfer and weigh about 4 g of sediment (msub) into a reaction flask. Transfer then 15 ml of distilled water into each reaction flask and seal all by stoppers. When all reaction flasks are placed on the heating plates, the entire distillation unit is assembled by sealing the ”tops” equipped with long Pasteur pipettes. Turn on the cooling water!!! Remember to check the water flow. Turn on the N2 flow using the small valves above the distillation unit. The flow is appropriate when there is a calm and steady bubble development in the traps. Let the reaction flasks stand with bubbling for 5-10 minutes before adding 8 ml of 6 M HCl (use a plastic syringe) and 16 ml of a chromium solution (1 M Cr2+ into 0.5 M HCl). Turn on the heating plates (level 3). Let the distillation proceed for 45 minutes after boil. Keep an eye on the N2 bubble pattern and the flow of cooling water. The reaction mixture has to boil vigorously. Terminate the distillation by turning off the heating and dismantling the unit (disconnect the connection to the trap) and close the N2 flow. resuspend the content of the traps with an automatic pipet and transfer immediately 5 ml into a 20-ml scintillation vial (only labels on the lid). The remainder is transferred into 20-ml scintillation vials and stored for later use. Add 2 ml scintillation liquid to all 7-ml vials containing supernatant and 10 ml scintillation liquid to the 20-ml vials containing trap material, and mix the solution for about 1 min. Make blanks. The samples are now ready for counting. Ask the teachers about the procedure. The reaction mixture from the reaction flasks is transferred into a waste container. The reaction flasks are rinsed thoroughly with distilled water before the start of a new distillation. Warning. Bottles with N2 are handled by staff members only. Working with open samples must occur on the covered tables. All handling of open samples must be done standing up, partly to avoid inhalation of aerosols, partly to minimize the risk of person contamination. Never use a whirli-mixer for samples in open containers, instead you must mix with an automatic pipette. 35S might be mutagenic. Waste must be collected in a plastic container marked “waste (affald in Danish)”. A solution of 6 M HCl is very corrosive. Vapours are very easily absorbed in the mouth, throat and lungs, where it can cause corrosion and fluid diffusion (dys-pnoea, difficulty in breathing). Chrome chloride may be mutagenic. Waste is collected in a plastic container marked “Cr”. Scintillation fluid is a mixture of organic solvents, and inhalation may cause brain damage. Use the fume hood! All other waste must be collected in the buckets with radioactive marks. D. Calculation of sulfate reduction rate The rate of sulfate reduction (SRR) is calculated according to: 26
Exercise 4 [SO42-]s ⋅ a ⋅ 24 ⋅ 1.06 SRR = ────────────── (nmol SO42- cm-3 d-1) (A + a) ⋅ h where [SO42-]s is sulfate concentration in nmol cm-3 sediment = [SO42-] (mM) ⋅ φ ⋅ 1000 (determined from porewater profiles, see Exercise 6); a is the total radioactivity of sulfide (H235S) in the sample slice (remember that only a subsample of the traps has been counted); A is the total radioactivity of sulfate (35SO42-) in the sample slice; h is incubation time in hours; 1.06 is a correction factor for bacterial isotopic fractionation. Note that A and a have to be transformed to dpm cm-3, based on the sediment weight and porosity (from Exercise 1) and corrected for the amount of sample used in the distillation. Calculation of a: K ⋅ (dpma - dpmB) ⋅ msedc ⋅ d a = ──────────────────── (dpm cm-3) msub ⋅ msed where: K = correction factor to scale to radioactivity in total trap volume (you may have only counted the radioactivity of a part of this volume). msed = weight of sediment slice before centrifugation (g). msedc = weight of sediment after centrifugation and decantation of ZnAc (g). msub = weight of sediment subsample (g). dpma = radioactivity of H235S (dpm). dpmB = background radioactivity (dpm). d = sediment density (g cm-3). - from Exercise 1 Calculation of A: (VZnAc + msed ⋅ ß/100) ⋅ (dpmA - dpmB) ⋅ d A = ──────────────────────────── (dpm cm-3) VA ⋅ msed where: VZnAc = volume of ZnAc (5.0 or 10.0 ml). ß = water content (%). - from Exercise 1 dpmA = radioactivity of 35SO42- (dpm). VA = volume of sampled supernatant (0.2 ml). E. Determination of the reduced sulfide pool Besides the labeled sulfide, the distillation procedure has also released and trapped the total pool of reduced inorganic sulfur in the sediment. The size of this pool can be determined by a spectrophotometric technique. For this purpose, we will use the excess trap material, which was stored in 20-ml scintillation vials. Mix the trap material thoroughly on a Whirli-mixer. Transfer 50 µl (KF) or 200 μl (FS) of each sample (+ a blank consisting of 5% ZnAc) into 27
Exercise 4 centrifuge tubes and dilute with 10 ml of distilled water. Add 800 μl of Cline’s reagent (Cline 1969) to the centrifuge tubes, close lids immediately, shake the tubes and let them rest for 20 minutes. Transfer ca. 3 ml of each sample to a cuvette and read the absorption at 670 nm on a standard spectrophotometer. The absorbance must be below 0.8. If it is higher, the entire procedure is repeated using less trap material. Warning. Cline reagent contains 6 M HCl and is very corrosive. Take good care and use paper cover on the table. The reagent also contains dimethylphenylendiamine (0.4%), which is very easily absorbed through the skin and methaemoglobin is formed in the blood. Dimethylphenylendiamine may have mutagenic and other dangerous effects (not yet completely known). By reaction with sulfide, methylene blue is created, which can be mutagenic. All Cline-work must be done in the fume hood. Gloves and lab coat must be worn and waste must be collected in a plastic container marked “Cline”. The concentration is calculated using a constant factor (μM/abs) based on the absorption coefficient for the actual Cline reagent used here (will be handed out by the teachers). The pool of Total Reduced Inorganic Sulfur (TRIS) can be obtained according to: abss ⋅ F ⋅ b ⋅ msedc ⋅ d [TRIS] = ──────────────── (μmol cm-3) k ⋅ 1000 ⋅ msub ⋅ msed where: abss = absorbance of trap sample. k = absorption coefficient (abs/μM). F = dilution factor. b = trap volume (10 ml). III. Questions and figures to be considered in the report 1. Plot SRR as a function of depth for both sites. 2. Plot TRIS as a function of depth in the sediment for both sites. 3. What are the advantages and disadvantages by incubations using intact cores? 4. Discuss the depth distribution of sulfate reduction rates in the sediment. 5. Compare the profiles of sulfate reduction rates with those of TRIS. 6. Calculated the depth-integrated sulfate reduction (mmol m-2 d-1). 7. Compare the depth-integrated SRR with sediment oxygen uptake (Exercise 2) - How much of the oxygen may be used to oxidize H2S at the two sites? Discuss potential differences. 8. How important is the SRR for the total benthic mineralization at both sites (see also appendix 3)? Discuss potential differences. 9. Give possible sources of errors. 28
Exercise 5 EXERCISE 5.1: Measurement of denitrification in the sediment by the 15N isotope- pairing technique Sediments typically house an active and extremely interesting microbially driven nitrogen cycle. Nitrate is used in denitrification (as well as anammox and nitrate reduction to ammonium) and this process can sometimes account for a significant fraction of carbon mineralization in sediments while also representing an important sink for bioavailable nitrogen. Denitrifiers are facultative anaerobic bacteria which can use nitrate instead of oxygen in anoxic environments. Nitrate comes either from diffusion from the overlying water or from nitrification in the sediments, a chemoautotrophic process where ammonium, liberated to the sediment through organic matter decomposition, is oxidized nitrate. This exercise will focus on denitrification. Our determination of denitrification rates applies the Isotope Pairing Technique (IPT) based on addition of nitrate labeled with the stable isotope 15N and subsequent analysis of the labeled dinitrogen pairs formed (Nielsen 1992). Denitrification rates can in theory be determined from the accumulation of N2 in the water column. In reality, however, this is extremely difficult due to the high background concentrations (about 500 μM) of N2 in water. Given some assumptions, the source of nitrate (overlying water or nitrification) for the denitrification can be derived from the Isotope Pairing Technique. The IPT as applied here assumes that anammox is not active. This assumption typically holds for shallow coastal sediments. I. Materials 3 sediment cores with sediment to about 10 cm from the top of the core 10 mM stock 15NO3- solution. Alu foil and saturated ZnCl Stirring collars Sediment stirring set up similar to exercise 2 Plastic disposable pipets Ruler Glass spatulas 50 ml glass syringe 7 cm isoversinic (gastight) tube to fit syringe 10 ml Exetainers 10 ml plastic vials for nutrient samples 0.45 µm syringe filters 10 ml syringes 1 ml pipet and tips 200 µl pipet and tips tape markers calculator 29
Exercise 5 rack II. Procedures A. Incubation Three sediment cores from your location are taken and the water level is adjusted to about 0.3 cm from the rim. Use a 50-ml syringe with a piece of tubing and try to avoid resuspension of sediment. Then take a water sample for nutrient analysis from the circulating water in the tank and filter it through 0.45 µm filters into a plastic vial. This and subsequent nutrient samples are needed to determine the percentage of labeled nitrate for separation of the sources of nitrate for denitrification (Dw and Dn – see below). Before addition of 15NO3- the height of the water column is measured to calculate the volume in which the added nitrate is diluted. Calculate the volume of your overlying water and add enough stock solution to bring the 15NO3- concentration to 50 μM. Check this calculation with the teachers before adding the 15NO3-. Mix 15NO3- into the water phase up by gentle bubbling with air from a disposable pipette. Notice: 15N is not a radioisotope! Take a subsample for NO3- analysis from each core and filter it into a plastic vial. Remember to label the vials with team number, core number and time (before or after label addition). Note starting time, add a stirring magnet, and close the core with a stopper without trapping air bubbles. Do not press the stopper too hard. Leave the cores with circulation in darkness for about 3 hours. At the end of the incubation, note the time and stop denitrification by dripping ~0.5 ml saturated ZnCl2 solution (a little toxic and corrosive) into the overlying water of the core with a small plastic pipet. With a glass spatula, mix the top 6 cm of sediment and water with gentle vertical and rotating motions to distribute the labeled N2 evenly before sampling. Avoid skimming the surface as gas may escape. Check the new height of the water column and the height of the mixed sediment to determine the volumes in which labeled dinitrogen will be distributed. With a 50 ml syringe with a small piece of isoversinic tubing at the end, first take in a few ml of the suspension and discharge to wash out any gas phase in the syringe, Now, carefully withdraw about 30 ml sample. You will have prepared 6 Exetainers by adding two drops of ZnCl2 solution to each. Transfer the sample to an Exetainer by introducing the tube at the bottom and then draw up as the vial is filled to the rim. This is to minimize loss of dinitrogen to the air. Put on the screw cap and check that no bubbles are trapped. Take two samples from each core. Dinitrogen samples are marked with Team number and core number, location and incubation time. They are saved for later IRMS analysis. B. Analysis of dinitrogen by mass spectrometry (IRMS). The isotopic composition in the produced N2 is determined by isotope ratio mass spectroscopy. Initially, with the lid still on the Exetainers, one cm of sample is replaced with 2 ml helium through hypodermic needles inserted through the rubber seal, and by vigorous shaking, dinitrogen is equilibrated into the He headspace. From the headspace, 0.25 ml is 30
You can also read