Small Business & Entrepreneurship

Methane efflux from marine sediments in passive and active margins: Estimations from bioenergetic reaction-transport simulations

Description
A simplified version of a kinetic-bioenergetic reaction model for anaerobic oxidation of methane (AOM) in marine sediments [Dale, A.W., Regnier, P., Van Cappellen, P., 2006. Bioenergetic controls on anaerobic oxidation of methane (AOM) in coastal
Published
of 16
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Methane efflux from marine sediments in passive and active margins:Estimations from bioenergetic reaction – transport simulations A.W. Dale ⁎ , P. Van Cappellen, D.R. Aguilera, P. Regnier   Department of Earth Sciences-Geochemistry, Utrecht University, P.O. Box 80021, 3508 TA Utrecht, Netherlands Received 8 September 2006; received in revised form 27 July 2007; accepted 26 September 2007Editor: H. ElderfieldAvailable online 2 October 2007 Abstract A simplified version of a kinetic –  bioenergetic reaction model for anaerobic oxidation of methane (AOM) in marine sediments[Dale, A.W., Regnier, P., Van Cappellen, P., 2006. Bioenergetic controls on anaerobic oxidation of methane (AOM) in coastalmarine sediments: a theoretical analysis. Am. J. Sci. 306, 246 – 294.] is used to assess the impact of transport processes on biomassdistributions, AOM rates and methane release fluxes from the sea floor. The model explicitly represents the functional microbialgroups and the kinetic and bioenergetic limitations of the microbial metabolic pathways involved in AOM. Model simulationsillustrate the dominant control exerted by the transport regime on the activity and abundance of AOM communities. Upward fluidflow at active seep systems restricts AOM to a narrow subsurface reaction zone and sustains high rates of methane oxidation. Incontrast, pore-water transport dominated by molecular diffusion leads to deeper and broader zones of AOM, characterized by muchlower rates and biomasses. Under steady-state conditions, less than 1% of the upward dissolved methane flux reaches the water column, irrespective of the transport regime. However, a sudden increase in the advective flux of dissolved methane, for example asa result of the destabilization of methane hydrates, causes a transient efflux of methane from the sediment. The benthic efflux of dissolved methane is due to the slow growth kinetics of the AOM community and lasts on the order of 60 years. This time windowis likely too short to allow for a significant escape of pore-water methane following a large scale gas hydrate dissolution event suchas the one that may have accompanied the Paleocene/Eocene Thermal Maximum (PETM).© 2007 Elsevier B.V. All rights reserved.  Keywords:  geomicrobial reaction networks; anaerobic oxidation of methane; reaction – transport model; methane fluxes; gas hydrates 1. Introduction Anoxic marine sediments constitute by far the largest methane (CH 4 ) reservoir on Earth, either dissolved in theinterstitialwaterorascondensedgashydrates.Onaglobalscale, the latter contain between 10,000 and 22,000Gt of carbon (Gt=10 15 g) (Dickens, 2001), with a mean  13 Cisotopiccompositionof  − 60 ‰ (Kvenvoldenetal.,1993).FluxesofCH 4 totheatmospherefromhydrateshavebeensuggested to be a key forcing of the climate system(Archer and Buffett, 2005) and associated with climaticoscillations in the geological past (Kennett et al., 2000).Modernmarine sediments,however, account for only 3%of the global CH 4  flux to the atmosphere (Reeburgh,2003) because most of the CH 4  is consumed beforereaching the seafloor through anaerobic oxidation of methane (AOM). Despite its global significance, themechanismofAOMisstilluncertain(Boetiusetal.,2000;  Available online at www.sciencedirect.com Earth and Planetary Science Letters 265 (2008) 329 – 344www.elsevier.com/locate/epsl ⁎  Correspondingauthor. Tel.:+3130 253 5075;fax: +31 30253 5302.  E-mail addresses:  dale@geo.uu.nl (A.W. Dale), pvc@geo.uu.nl(P. Van Cappellen), aguilera@geo.uu.nl (D.R. Aguilera),regnier@geo.uu.nl (P. Regnier).0012-821X/$ - see front matter © 2007 Elsevier B.V. All rights reserved.doi:10.1016/j.epsl.2007.09.026  Hinrichs et al., 1999). It is thought to be carried out by asyntrophic association of methane-oxidizing archaea andsulfate-reducing bacteria (Boetius et al., 2000). The net AOM reactionconsumes methane and sulfate (SO 42 − ) and produces sulfide (HS − ) and carbonate alkalinity (HCO 3 − ),according to:CH 4 ð aq Þ  þ  SO 2 − 4  ð aq Þ → HS − ð aq Þ  þ  H 2 O ð l Þ  þ  HCO − 3 ð aq Þ ð 1 Þ The highest AOM rates are found in cold-seepenvironments, where high CH 4  fluxes are sustained byupward advection of pore-water flow (Luff and Wall-mann, 2003). The upward fluid flow may arise from avariety of processes, such as hydrate dewatering induced by tectonic uplift or hydrate dissolution due togeothermal gradients (Suess et al., 1999). The dissolvedCH 4 isthenmostlyconsumedatageochemicalboundarytermed the sulfate – methane transition zone (SMTZ),where SO 42 − concentrations typically fall from seawater values ( ∼ 28 mM) to sub-mM levels within a few cm below the sea floor. Biomass concentrations also showmaximumvaluesinthishighlyactivezoneandmayforma thick mat at the sediment surface (Treude et al., 2003).From a global perspective, however, the bulk of AOMoccurs in passive continental shelf sediments where pore-water solute transport occurs mainly by molecular diffusion. In these systems, the SMTZ extends over agreater depth interval, and is generally located between1 – 10 m below the sediment surface.Despite considerable effortto quantify CH 4  flux fromthe sea floor (Luff and Wallmann, 2003; Tryon andBrown, 2001; Wallmann et al., 2006; Torres et al., 2002;Mau et al., 2006; Haese et al., 2003; Linke et al., 2005;Sommer et al., 2006), the effects of the macroscopictransport regime on the AOM efficiency and CH 4 turnover remain equivocal. Early diagenetic models of CH 4  generally assume a steady-state biomass and ignorethe role of microbial bioenergetics. Most commonly, a bimolecular expression for AOM rate (Luff and Wall-mann, 2003; Haese et al., 2003; Sommer et al., 2006;Wallmann et al., 2006; Van Cappellen and Wang, 1996)is used:Rate  ¼  k   AOM  ½ SO 2 − 4  ½ CH 4  ð 2 Þ where  k   AOM   is an apparent second-order rate constant (concentration − 1 time − 1 ). In the case of AOM, the verysmall energetic yields and low growth rates are likely to play a major role in the response of AOM to changes inthe fluid flow regime and CH 4  transport rates. Theseeffects, however, cannot be accounted for by the aboverate equation.In this work, a model coupling microbial activity andtransport is used to explore the impact of externally-impressedfluidflowonbiomassdistributions,metabolismand, ultimately, on CH 4  fluxes in marine sediments. AreducedversionofthereactionnetworkdevelopedbyDaleet al. (2006) was incorporated into a reactive-transport model software (Aguilera et al., 2005). The modelexplicitly simulates functional biomasses, and accountsfor kinetic and bioenergetic constraints on biomasssynthesis and substrate turnover. The first part of the paper investigates two representative steady-state baselinescenarios with contrasting macroscopic transport regimes;a passive, non-seep, margin in which transport isdominated by molecular diffusion (diffusion-dominatedsystem,DDS), andanactive seep systeminwhich upwardadvection of pore water is the dominant solute transport term(advection-dominatedsystem,ADS).Thesecondpart focuses on the transient evolution of biomass distributionandCH 4 effluxfollowingchangesintheupwardadvectiveCH 4  supply. Finally, the biomass response to fluid flow isused to asses the importance of AOM in delivering largequantities of isotopically-light carbon to the ocean – atmosphere during hydrate dissolution scenarios. 2. Modeling strategy 2.1. Physical aspects The biomass-explicit model simulates the depth profiles of the pore-water concentrations of methane(CH 4 ), sulfate (SO 42 − ) and hydrogen (H 2 ), as well as themicrobial biomasses of sulfate-reducing bacteria (  B SRB ),methanogenic archaea (  B  MET  ) and methane-oxidizingarchaea (  B  MOA ). The biomasses are assumed to beattached to particle surfaces. Three microbially-mediat-ed reactions are considered: hydrogenotrophic sulfatereduction (hySR), hydrogenotrophic methanogenesis(hyME) and AOM (Table 1).The one-dimensional mass-conservation equations for the solutes and biomasses are (Berner, 1980; Boudreau,1997):Solutes :  u A C   j  A t   ¼  AA  x  u d   D S  A C   j  A  x     A  u d  v  d  C   j    A  x  þ  u  R  j   ð 3a Þ Biomasses : 1    u ð Þ A  B i A t   ¼  1    u ð Þ  R i  ð 3b Þ where  x  is depth (positive downwards),  φ  is porosity,  t  is time,  C    j   and  B i  are the time- and depth-dependent  330  A.W. Dale et al. / Earth and Planetary Science Letters 265 (2008) 329  –  344  concentrations of the solutes and biomass groupsconsidered,  D S   is the tortuosity-corrected molecular diffusion coefficient (see Appendix),  v   is the verticalupward fluid flow velocity,  R   j   is the rate of change of   C    j  due to biogeochemical reactions, and  R i  is the net growth or decay rate of   B i  (see Section 2.2). Note that although  v   has a negative sign, all fluid flow velocitiesare reported from here on as absolute values. For the DDS and ADS baseline scenarios,  v   equals 0 and10 cm y − 1 , respectively.In Eqs. (3a) and (3b) bioturbation and bioirrigationare not considered. Biologically-enhanced transport wasfound to be negligible at Hydrate Ridge seeps (Luff andWallmann, 2003), while in the DDS simulations thedepth of the SMTZ is typically well-below the mixedand irrigated sediment layers. Sediment burial is alsoignored because it has a negligible effect on the modeledspatial distribution of biomasses (results not shown).The time-dependent change in  B i  thus only depends onthe net growth rate (Eq. (3b)). Omission of theadditional transport mechanisms does not affect themain conclusions of the paper. Further details of themodel including the calculation of solute fluxes acrossthe sediment  – water interface are given in the Appendix. 2.2. Biogeochemical aspects The biogeochemical reaction network (Table 1) is areduced version of the model developed by Dale et al.(2006). Microbial growth depends on the energy-generating reaction between an electron donor and ac-ceptor (the catabolic reaction), whose rate is a functionof kinetic and bioenergetic factors. The two electrondonors considered here are CH 4  and H 2 . Hydrogen is aubiquitous product of organic matter fermentation and amajor energy source for a wide range of microorganisms(Schink, 1997). Sulfate-reducing bacteria compete withmethanogenic archaea for H 2  (Lovley and Goodwin,1988), whereas their relationship with methane oxidi-zers is synergistic; AOM only occurs when the sulfatereducers consume H 2  to low enough levels for AOM to become thermodynamically viable (Dale et al., 2006;Hoehler et al., 1994). In view of mounting evidence, weassume that reverse bicarbonate methanogenesis (i.e.hyME in Table 1 written from right-to-left) is a feasible pathway for AOM (Krüger et al., 2003; Hallam et al.,2004), although this has been questioned ( Nauhaus et al., 2002). Simultaneous AOM and hyME is thusimpossible, since they have equal and opposite Gibbsenergy yields.The depth-dependent production of H 2  from fermen-tation of organic matter is treated as a forcing function inthe model, given by:H 2  input   ¼  g : exp  a d   x ð 4 Þ where  γ  is the rate of H 2  production at the sediment surface (mM y − 1 , Table A1) and  α  (cm − 1 ) is a depthattenuationcoefficient.Theexponentialfunctionreflectsthe decreasing abundance and reactivity of organicmatter with depth in sediments (Middelburg, 1989).The rate of catabolism is linked to the rate of  biomass synthesis (Dale et al., 2006) via the growthyield (mole C biomass produced per mole electrondonor consumed). The macrochemical equations,which couple catabolism to the microbial growth of each biomass species (anabolism) (Dale et al., 2006), arealso shown in Table 1. Note that a cellular compositionof C 5 H 7 O 2  N, is assumed (Rittmann and McCarty,2001). This cellular composition is close to the repor-ted average composition (C 4 H 6 O 2  N) for chemolithoau-totrophic microorganisms in anoxic environments(McCollom and Amend, 2005). Table 1Microbial redox processes considered in the modelBiomass Catabolic reaction Stoichiometry a  Δ G  o ′  B SRB  Sulfate reduction (hySR)  1 /  2 H 2 (aq) + 1 /  8 SO 42 − (aq) + 1 /  8 H +(aq) → 1 /  8 HS − (aq) + 1 /  2 H 2 O (aq)  − 27.739.5 SO 42 − +0.2NH 4+ +10.3H + +HCO 3 − +40 H 2 → 0.2 C 5 H 7 O 2 N +9.5HS − +40.6H 2 O  B  MET   Methanogenesis (hyME)  1 /  2 H 2 (aq) + 1 /  8 HCO 3 − (aq) + 1 /  8 H +(aq) → 1 /  8 CH 4 (aq) + 3 /  8 H 2 O (aq)  − 23.9110.5HCO 3 − +0.2NH 4+ +10.1H + +40 H 2 → 0.2 C 5 H 7 O 2 N +9.5 CH 4 +31.1H 2 O  B  MOA  Methane oxidation (AOM)  1 /  8 CH 4 (aq) + 3 /  8 H 2 O (aq) → 1 /  2 H 2 (aq) + 1 /  8 HCO 3 − (aq) + 1 /  8 H +(aq)  +23.9110 CH 4 +0.2NH 4+ +27.4H 2 O → 0.2 C 5 H 7 O 2 N +9.2H + +38 H 2 +9HCO 3 − For each process, the upper equation is the catabolic energy-generating reaction. The lower equation corresponds to the macrochemical growth anddefines the modeled stoichiometry required to synthesize 1 C-mol biomass (i.e.  1 /  5 ×C 5 H 7 O 2  N). Only the constituents in bold are explicitly modeled.The standardGibbs energies ( Δ G  o ′ , kJ e-mol − 1 ) of the catabolic reactions are calculated usingthe data in (Amend and Shock, 2001) and corrected for  biologically neutral pH conditions (pH=7.37) at 278 K and 1 bar with the transformation  Δ G  o ′ = Δ G  o ′ −  RT  υ H+  ln[Kw 0.5 ], where  υ H+  is thestoichiometric coefficient of H + in the reaction Amend and Shock (2001) and Kw is the ion product. Pressure effects for the passive and active seepenvironments are neglected in the calculations. a  An explanation of how the growth equations are established is given by Dale et al. (2006).331  A.W. Dale et al. / Earth and Planetary Science Letters 265 (2008) 329  –  344  The net growth rate of a given biomass group  i (mol C g − 1 y − 1 ) is represented in the model by:d  B i d t   ¼  l max ; i d   B i d   F   K  ; i d   F  T  ; i    l e ; i d   B i  ð 5 Þ where  μ max, i  (y − 1 ) is the maximum specific growth rateand  μ e , i  (y − 1 ) the maximum specific decay rate of the biomass group (Jin and Bethke, 2005).  F   K  , i  and  F  T  , i  arethe kinetic and thermodynamic driving forces for microbial growth (see below). Values of   μ max  for hySR, hyME and AOM of 26.1, 23.2 and 18.3 y − 1 ,respectively, were calculated by Dale et al. (2006) fromgeneralized principles of microbial metabolism (Ritt-mann and McCarty, 2001). Biomass decay represents afirst-order loss term due to, among others, cell death,lysis and trans-membrane leakage of metabolites. Valuesof   μ e  are assumed to be 0.1 y − 1 for all species, whichfalls within the range 0.04 – 0.32 y − 1 of  Dale et al.(2006). A central working hypothesis of the present  paper is that the same set of microbial parameter valuesapplies to the ADS and DDS scenarios. This assumptionis necessary because comparable data of microbialgrowth rates in diffusive and cold-seep environmentsare currently lacking.The kinetic and bioenergetic controls on biomassgrowth are included in Eq. (5) through the  F   K  , i  and  F  T  , i terms, which are given by:  F   K  ; i  ¼  E D ½   E   E   D S   i  þ  E D ½  d  E A ½   K   E   A S   i  þ  E A ½ ð 6 Þ  F  T  ; i  ¼  1    exp  D G   NET  v  RT     if   D G   NET  b 0  ð 7a Þ  F  T  ; i  ¼  0 if   D G   NET  N 0  ð 7b Þ and D G   NET   ¼  D G   INSITU   þ D G   BQ  ð 8 Þ In Eq. (6),  K  S  - i ED and  K  S  - i EA are the half-saturationconstantsfortheelectrondonor(E D )andelectronacceptor (E A ) (mol L − 1 ), respectively. In the reaction network considered, the only electron acceptor which may be rate-limitingisSO 42 − .InEqs.(7a)and(7b)and(8), Δ G   NET  isthenet Gibbs energy (kJ e-mol − 1 ) which is channeled intomicrobial growth, while  Δ G   INSITU   (kJ e-mol − 1 ) is the insitu Gibbs energy yield of the corresponding catabolicreactionand Δ G   BQ  (kJ e-mol − 1 ) is thebioenergetic energyminimum (see below).  χ  is the average stoichiometricnumber; equivalent to the number of protons translocatedacross the cell membrane during catabolism and isassumed to be equal to 1 per electron transferred (Jin andBethke, 2005).  F  T  and  F   K   are dimensionless functions andvary between 0 (complete kinetic/thermodynamic limita-tion) and 1 (no kinetic/thermodynamic limitation). Theincorporation of the  F  T   term in Eq. (5) accounts for thelimiting effect on the catabolic reaction rate by theaccumulation of reaction products.For each catabolic reaction,  Δ G   INSITU   is calculatedfrom the activity quotient of the solutes involved in thereaction and the standard Gibbs energy of catabolism, Δ G  o '   (kJ e-mol − 1 , Table 1).  Δ G   BQ  is defined as theminimum conservable energy needed to maintaincontinuous microbial activity (Hoehler, 2004). It isoften quoted to be equal to 15 – 20 kJ per mol per formula reaction, that is, the energy required tosynthesize 1/3 to 1/4 mol ATP (Schink, 1997). However,these very high values of   Δ G   BQ  are measured under optimal laboratory conditions (Schink, 1997), and their extrapolation to field conditions is questionable (LaR-owe and Helgeson, 2007). For the catabolic processesconsidered here,  Δ G   BQ  is estimated to be much lower,on the order of 0.5 kJ e-mol − 1 , based on the modelingwork by Dale et al. (2006). Note that   Δ G   BQ  is definedhere as a positive value and represents a fixed loss of  Δ G   INSITU  , which has a negative value when the reactionis thermodynamically favored. The minimum energylimitation is accounted for in the model such that   F  T  =0if   Δ G   NET  N 0 (Eqs. (7a) and (b)). 3. Model application: steady-state scenarios 3.1. DDS and ADS baseline simulations Model results are in order-of-magnitude agreement with typical integrated AOM rates ( Σ AOM), CH 4  effluxfromsediments and biomassconcentrationsobservedinavariety of DDS and ADS environments (Table 2). In theDDS (Fig. 1a – d), the sediment is characterized by asulfate-reducing zone extending from 0 to 400 cm and amethanogenic zone from 400 to 500 cm. CH 4  is almost entirelyconsumedintheSMTZ(arbitrarilydefinedwhereAOMrate N 0.01%ofmaximumrate)anddoesnotescapethe sediment. The maximum (0.2 nmol cm − 3 d − 1 ) andintegrated (0.04 mmol m − 2 d − 1 ) AOM rates are largelydeterminedbythemagnitudeoftheupwarddiffusiveCH 4 flux. Above and below the SMTZ, fermentative H 2  production drives hySR (0.1 – 0.2 nmol cm − 3 d − 1 ) andhyME ( ∼ 0.1 nmol cm − 3 d − 1 ). The H 2  concentrationexhibits values characteristic of sediments where sulfatereduction (0.1 – 0.2 nM) and methanogenesis (7 nM) arethe dominant microbial metabolic pathways (Lovley andGoodwin, 1988). 332  A.W. Dale et al. / Earth and Planetary Science Letters 265 (2008) 329  –  344  The distributions of   B MOA  and  B SRB  in the SMTZmirror the corresponding rate profiles, with maximumconcentrations of 0.23 and 0.45×10 8 cells cm − 3 ,respectively (Fig. 1d). Methanogenic biomass withintheSMTZisverylowandonlyaccumulates whereSO 42 − is depleted (Dale et al., 2006; Lovley and Goodwin,1988). Total biomass concentration integrated over theSMTZ equals 2.8×10 9 cells cm − 2 (4.4×10 − 6 mol Ccm − 2 ), equivalent to 28% of the total sediment biomass( Σ  B =1.0×10 10 cells cm − 2 or 1.6×10 − 5 mol C cm − 2 ).The calculated mean biomass concentration, [  B ],(2.0×10 7 cells cm − 3 or 3.2×10 − 8 mol C cm − 2 ) is 1 – 2orders-of-magnitude lower than reported values for surface sediments along passive margins (Table 2) because many more microorganisms inhabit thesesediments than those considered here. This discrepancydoes not occur for the ADS (see below), because sulfatereducers and methane oxidizers dominate the biomass inseep environments.The results for the baseline ADS scenario (Fig. 1e – h)are a striking contrast to the DDS due to the upwardfluid flow of 10 cm y − 1 imposed at the lower boundary.Almost all sulfate reduction is coupled to AOM, andSO 42 − drops to 1 mM at 3 cm below the sediment surface, in line with observations at Hydrate Ridge seeps(Luff and Wallmann, 2003). The vertical advective fluxof dissolved CH 4  effectively forces the SMTZ to be positioned immediately below the sediment  – water interface and exhibits a distinct peak in coupledhySR  – AOM rates at 3 cm. Maximum hySR and Table 2CH 4  efflux rates, integrated AOM rates ( Σ AOM) and biomasses ( Σ  B ) and mean biomass concentration ([  B ]) in passive and active margins comparedto model-predicted values in this work (in bold)Value Active margins site(and reference)Method Value Passive margins site(and reference)MethodCH 4  efflux(mmol m − 2 d − 1 )1.9 – 90 Hydrate Ridge(Torres et al., 2002;Linke et al., 2005)Benthic chamber 0.6 Hydrate Ridge(Sommer et al., 2006)Model No data12.1 Costa Rica MV(Linke et al., 2005)Model8.3 Dvurechenskii MV(Wallmann et al., 2006)Model 0.1 ADS (this study) Model 5.0 × 10 − 4 DDS (this study) Model Σ AOM(mmol m − 2 d − 1 )16.3 Kazan MV(Haese et al., 2003)SO 42 −  profile 0.05 Blake Ridge(Borowski et al., 1996)SO 42 −  profile;  35 S5.2 – 140 Hydrate Ridge(Boetius et al., 2000;Luff and Wallmann, 2003;Treude et al., 2003)Model;  35 S 0.1 Black Sea(Jørgensen et al., 2001)SO 42 −  profile16.1 Costa Rica MV(Linke et al., 2005)Model 0.24 Namibian shelf (Fossing et al., 2000)SO 42 −  profile;  35 S31.1 Dvurechenskii MV(Wallmann et al., 2006)Model 0.4White Oak River Estuary(Martens et al., 1998)Model 19.4 ADS (this study) Model 0.04DDS (this study)Model Σ  B (cells cm − 2 )1.5 – 1.8×10 11 Hydrate Ridge(Treude et al., 2003) a  AODC  b 1.0×10 11 Danish coast [B. Cragg, pers. comm.]AODC  b 7.4 × 10 11 ADS (this study) Model 1.0 × 10 10 DDS (this study) Model [  B ](cells cm − 3 )2.1×10 10 Hydrate Ridge(Boetius et al., 2000)FISH  c 0.5 – 2.2×10 8 Surface sediments(Whitman et al., 1998)Arithmetic mean3.8×10 10 Hydrate Ridge(Treude et al., 2003)AODC  b 5.0×10 9 Coastal sediments(Luna et al., 2002)AODC  b 2.7×10 9 Gulf of Mexico(Orcutt et al., 2005)FISH c 2.0 × 10 7 DDS (this study)  Model d 7.4 × 10 10 ADS (this study) d Model (MV = mud volcano). a  Calculated from total aggregates, assuming that 1 aggregate contains 300 cells (Boetius et al., 2000).  b Acridine orange direct counts (AODC). c Fluorescence in situ hybridization. d Meanvalueoverthe top10 cmwherehighestrates ofmicrobialactivityoccur(Boetius etal., 2000;Luff andWallmann,2003;Orcuttet al., 2005).333  A.W. Dale et al. / Earth and Planetary Science Letters 265 (2008) 329  –  344
Search
Similar documents
View more...
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks