Spatial and temporal patterns of CH₄ and N₂O fluxes in terrestrial ecosystems of North America during 1979–2008: application of a global biogeochemistry model

Description
Continental-scale estimations of terrestrial methane (CH₄) and nitrous oxide (N₂O) fluxes over a long time period are crucial to accurately assess the global balance of greenhouse gases and enhance our understanding and prediction of global climate

Please download to get full document.

View again

of 22
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.
Information
Category:

Small Business & Entrepreneurship

Publish on:

Views: 8 | Pages: 22

Extension: PDF | Download: 0

Share
Tags
Transcript
  Biogeosciences, 7, 2673–2694, 2010www.biogeosciences.net/7/2673/2010/ doi:10.5194/bg-7-2673-2010© Author(s) 2010. CC Attribution 3.0 License. Biogeosciences Spatial and temporal patterns of CH 4  and N 2 O fluxes in terrestrialecosystems of North America during 1979–2008: application of aglobal biogeochemistry model H. Tian 1,2 , X. Xu 1,2 , M. Liu 1,2 , W. Ren 1,2 , C. Zhang 1 , G. Chen 1,2 , and C. Lu 1,21 Ecosystem Dynamics and Global Ecology (EDGE) Laboratory, School of Forestry and Wildlife Sciences, AuburnUniversity, Auburn, AL, 36849, USA 2 International Center for Climate and Global Change Research, Auburn University, Auburn, AL, 36849, USAReceived: 16 March 2010 – Published in Biogeosciences Discuss.: 21 April 2010Revised: 30 August 2010 – Accepted: 1 September 2010 – Published: 10 September 2010 Abstract.  Continental-scale estimations of terrestrialmethane (CH 4 ) and nitrous oxide (N 2 O) fluxes over a longtime period are crucial to accurately assess the global bal-ance of greenhouse gases and enhance our understanding andprediction of global climate change and terrestrial ecosys-tem feedbacks. Using a process-based global biogeochem-ical model, the Dynamic Land Ecosystem Model (DLEM),we quantified simultaneously CH 4  and N 2 O fluxes in NorthAmerica’s terrestrial ecosystems from 1979 to 2008. Dur-ing the past 30 years, approximately 14.69 ± 1.64TgCa − 1 (1Tg=10 12 g) of CH 4 , and 1.94 ± 0.1TgNa − 1 of N 2 Owere released from terrestrial ecosystems in North Amer-ica. At the country level, both the US and Canada actedas CH 4  sources to the atmosphere, but Mexico mainly ox-idized and consumed CH 4  from the atmosphere. Wetlandsin North America contributed predominantly to the regionalCH 4  source, while all other ecosystems acted as sinks for at-mospheric CH 4 , of which forests accounted for 36.8%. Re-garding N 2 O emission in North America, the US, Canada,and Mexico contributed 56.19%, 18.23%, and 25.58%, re-spectively, to the continental source over the past 30 years.Forests and croplands were the two ecosystems that con-tributed most to continental N 2 O emission. The inter-annualvariations of CH 4  and N 2 O fluxes in North America weremainly attributed to year-to-year climatic variability. Whileonly annual precipitation was found to have a significant ef-fect on annual CH 4  flux, both mean annual temperature andannual precipitation were significantly correlated to annual Correspondence to:  H. Tian(tianhan@auburn.edu)N 2 O flux. The regional estimates and spatiotemporal pat-terns of terrestrial ecosystem CH 4  and N 2 O fluxes in NorthAmerica generated in this study provide useful informationfor global change research and policy making. 1 Introduction Methane (CH 4 ) and nitrous oxide (N 2 O) are two potentgreenhouse gases which in sum contribute to more than onefourth of global warming caused by anthropogenic activities(Forster et al., 2007). Although the concentrations of CH 4 and N 2 O in the atmosphere are relatively low, their warm-ing potentials are much higher than that of carbon dioxide(Denman et al., 2007). CH 4  and N 2 O also play significantroles in ozone layer chemistry (Denman et al., 2007; Forsteret al., 2007). Similar to the increase of atmospheric CO 2 concentration, the concentrations of these two gases dramat-ically increased since the Industrial Revolution (Forster etal., 2007; Tueut et al., 2007; Rigby et al., 2008). Althoughthe importance of CH 4  and N 2 O emissions in changing theEarth’s climate has been recognized, scientific communityhas placed large emphasis on the CO 2  problem. Understand-ingandquantifyingCH 4  andN 2 Ofluxesinterrestrialecosys-tems at large spatial scales, therefore, becomes an urgent task for accurately predicting the future climate change (Rigby etal., 2008; Forster et al., 2007; Sheldon and Barnhart, 2009).Terrestrial ecosystems could act as either sources or sinksfor atmospheric CH 4  and N 2 O, depending on the timeand location (Liu, 1996; Potter, 1997; Ridgwell et al.,1999; Chapuis-Lardy et al., 2007; Xu et al., 2008). Glob-ally, natural sources from terrestrial ecosystems contributePublished by Copernicus Publications on behalf of the European Geosciences Union.  2674 H. Tian et al.: Terrestrial fluxes of CH 4  and N 2 O over North Americaapproximately 40% to the CH 4 , and more than half to theN 2 O releases to the atmosphere when removing oceanic con-tribution (Denman et al., 2007). North America, with itslarge land area and high proportion of natural wetland (ap-proximately 30% of the global wetland) (Bridgham et al.,2006; Mitsch and Gosselink, 2007), plays a critical role inglobal carbon cycling (Schimel et al., 2000). However, onlya few studies have investigated CH 4  and N 2 O fluxes overterrestrial ecosystems in North America (Bridgham et al.,2006). For example, Zhuang et al. (2004) estimated thatsoils in Canada and Alaska emitted 7.1 and 3.8TgCH 4  a − 1 ,respectively, during the 1990s. Bridgham et al. (2006) es-timated that CH 4  emission in North America’s wetlands is9TgCH 4  a − 1 . Using a satellite-derived modeling approach,Potter et al. (2006) estimated that the CH 4  emission from thenatural wetlands in the conterminous US is 5.5TgCH 4  a − 1 .Several studies also reported the fluxes of N 2 O in terrestrialecosystems at global and regional scales using empirical ap-proaches (Xu et al., 2008). While these studies improvedour understanding of CH 4  and N 2 O fluxes in North America,accurate estimations of terrestrial ecosystem CH 4  and N 2 Ofluxes in the entire continent over a long time period are stillneeded (Wofsy and Harriss, 2002).Many factors can influence CH 4  and N 2 O fluxes in terres-trial ecosystems at site and regional levels, such as elevatedCO 2  (Hutchin et al., 1995; Schrope et al., 1999; Phillips etal., 2001a, 2001b), tropospheric ozone pollution (Morsky etal., 2008), nitrogen input (Ding et al., 2004), climate change(Goldberg and Gebauer, 2009) and land cover change (Willi-son et al., 1995; Huang et al., 2010). However, most previousprocess-based modeling efforts did not take into account theconcurrent effects of multiple global change factors (Potter,1997; Cao et al., 1998; Walter et al., 2001; Zhuang et al.,2007, 2004). Large uncertainty still exists in the magnitudes,spatial and temporal patterns of CH 4  and N 2 O fluxes at largescales (Kort et al., 2008; Christensen et al., 1996; Zhuang etal., 2004; Bridgham et al., 2006; Potter et al., 2006).Recently, we developed a process-based biogeochemistrymodel, the Dynamic Land Ecosystem Model (DLEM), tosimulate biogeochemical cycling of carbon, nitrogen and wa-ter in the land ecosystems. The DLEM considers multiplefactors including climate, atmospheric compositions (CO 2 ,O 3 ), precipitation chemistry (nitrogen composition), natu-ral disturbances (fire, insect/disease, hurricane, etc), land-use/land-cover change, and land management (harvest, ro-tation, fertilization, irrigation, etc.) (Tian et al., 2005, 2008,2010; Ren et al., 2007a, 2007b, 2009; Zhang et al., 2007,2008; Lu, 2009; Liu et al., 2008; Chen et al., 2006; Xu,2010). This model has been successfully applied to simulatethe effects of multiple environmental factors on carbon andwater cycles in China (Ren et al., 2007a, 2007b; Lu, 2009;Liu et al., 2008; Chen et al., 2006; Xu, 2010) and USA (Tianet al., 2008, 2010; Zhang et al., 2007, 2008).In this study, we enhanced the model’s capability by ad-dressing the biogeochemical processes of CH 4  and N 2 O andsimulated CH 4  and N 2 O fluxes over terrestrial ecosystems inNorth America from 1979 to 2008. The objectives of thisstudy are: (1) to develop the CH 4  and N 2 O modules in theframework of an extant process-based model, DLEM; (2) tocompare modeled results with field observations and otherregional estimates; (3) to estimate CH 4  and N 2 O fluxes inNorth America’s terrestrial ecosystems from 1979 to 2008;and (4) to quantify the contributions of individual countriesand biomes to regional CH 4  and N 2 O fluxes in North Amer-ica. 2 Methodology2.1 The DLEM model and its trace gas modules The Dynamic Land Ecosystem Model (DLEM) couples ma- jor biogeochemical cycles, hydrological cycles, and vegeta-tion dynamics to make daily, spatially-explicit estimates of carbon, nitrogen, and water fluxes and pool sizes (C andN) in terrestrial ecosystems. There are five core compo-nents in the DLEM: (1) biophysics, (2) plant physiology,(3) soilbiogeochemistry, (4) dynamicvegetation, and (5)dis-turbance, land use and management. Briefly, the biophysicscomponent simulates the instantaneous fluxes of energy, wa-ter, and momentum within land ecosystems and their ex-changes with the surrounding environment. The plant phys-iology component simulates major physiological processes,such as plant phenology, C and N assimilation, respiration,allocation, and turnover. The soil biogeochemistry compo-nent simulates the dynamics of nutrient compositions andmajor microbial processes. The biogeochemical processes,including the nutrient mineralization/immobilization, nitrifi-cation/denitrification, decomposition, and methane produc-tion/oxidation are considered in this component. The dy-namic vegetation component simulates the structural dynam-ics of vegetation caused by natural and human disturbances.Two processes are considered: the biogeography redistribu-tion when climate change occurs, and the recovery and suc-cession of vegetation after disturbances. Like most dynamicglobalvegetationmodels, theDLEMbuildsontheconceptof plant functional types (PFT) to describe vegetation attributes.The disturbances, land use and management component sim-ulatescroplandconversion, reforestationaftercroplandaban-donment, and forest management practices such as harvest,thinning, fertilization and prescribed fires.The interactions and feedbacks of various processesamong core components are simulated as controls or materialflows (Fig. 1). The biophysics component yields influenceson plant physiology component through the effects of wa-ter, temperature and radiation, and on soil biogeochemistrycomponent through the effects of soil moisture and temper-ature; the plant physiology component yields influences onthe biophysics component through changes in leaf area in-dex (LAI), canopy conductance, and transpiration, on the soilBiogeosciences, 7, 2673–2694, 2010 www.biogeosciences.net/7/2673/2010/   H. Tian et al.: Terrestrial fluxes of CH 4  and N 2 O over North America 2675 Fig. 1.  Conceptual model of the Dynamic Land Ecosystem Model (DLEM) (Five core components are included in the DLEM). biogeochemistry component through litter-fall, and on thedynamic vegetation component through biomass growth; thedynamic vegetation component yields influences on the plantphysiology and soil biogeochemistry components throughshifts of plant function type (PFT); the soil biogeochemistrycomponent yields influences on the dynamics vegetation andplant physiology components through nutrient flow; distur-bances, land use and management component yields influ-ences on the other four components through changes in landcover type, PFT and nutrient and water flow (Fig. 1).Meanwhile, the DLEM uses climate data from regionalclimate and atmosphere chemistry component which couldbe a climate model or input data. The DLEM outputs in-cluding ecosystem carbon and nitrogen pools and fluxes(e.g. greenhouse gases) will enter the atmosphere; and thewater output and associated nutrients from the DLEM willenter water transport module and flow into lake, river andocean. All the components are also linked together by waterand energy fluxes (Fig. 1).The DLEM emphasizes the modeling and simulationof managed ecosystems including agricultural ecosystems,plantation forests and pastures. The spatial data sets of land management, such as irrigation, fertilization, rotation,and harvest can be used as input information for simulat-ing influences of land management on the structure andfunctioning of ecosystems. This model has been calibratedagainst various field data from US Long-Term EcologicalResearch (LTER) network, AmeriFlux network, and the Chi-nese Ecological Research Network (CERN) which cover var-ious ecosystems, including forests, grasslands, shrub, tundra,desert, wetland, and croplands. The major carbon, nitrogenand water variables have been validated with observationaldata. The simulated results have been compared with inde-pendent field data and satellite products. The DLEM oper-ates at a daily time step and at varied spatial resolutions, frommeters to kilometers, from regional to global. The additionalinformation on the processes, interactions and feedbacks inthe DLEM and associated input/output data (Fig. 1) can befound in our previous studies (Tian et al., 2005, 2008, 2010;Ren et al., 2007a, 2007b, 2009; Zhang et al., 2007, 2008; Lu,2009; Liu et al., 2008; Chen et al., 2006).In this paper, we provide a detailed description of theCH 4  and N 2 O modules with an emphasis on major processesthat control fluxes of CH 4  and N 2 O in terrestrial ecosystems(Fig. 2). 2.1.1 The CH 4  module The CH 4  exchanges between ecosystems and the atmo-sphere are a combination of CH 4  production, oxidation,and transportation from soil pore water to the atmosphere.The DLEM only considers CH 4  production from dissolvedwww.biogeosciences.net/7/2673/2010/ Biogeosciences, 7, 2673–2694, 2010  2676 H. Tian et al.: Terrestrial fluxes of CH 4  and N 2 O over North America Fig. 2.  Modules of CH 4  and N 2 O in the Dynamic Land Ecosys-tem Model (DLEM) (CH 4  production, oxidation, and transport areconsidered in the CH 4  module; nitrification and denitrification areconsidered in the N 2 O module). organic carbon (DOC), which is indirectly controlled by en-vironmental factors including soil pH, temperature and soilmoisture content. The production of DOC mainly comesfrom two sources: allocation of gross primary production(GPP) and decomposition of litter-fall and soil organic mat-ter. The accumulated DOC is either used as substrate formethane, leaves system as leachate, or enters the atmosphereas CO 2  via decomposition. CH 4  oxidation, including the ox-idation during CH 4  transport to the atmosphere, CH 4  oxida-tion in the soil pore water, and atmospheric CH 4  oxidation onthe soil surface, is determined by CH 4  concentrations in theair or soil pore water, as well as soil moisture, pH, and tem-perature. Most CH 4 -related biogeochemical reactions in theDLEM are described by using the Michaelis-Menten equa-tion with two coefficients: maximum reaction rate and half saturation coefficient. Three pathways for CH 4  transportfrom soil to the atmosphere include ebullition, diffusion, andplant-mediated transport. It is assumed that methane-relatedbiogeochemical processes only occur in the top 50cm of soilprofile. The net CH 4  flux between the atmosphere and soil isdetermined by the following equation: F  CH 4  = F  P + F  D + F  E − F  air ,  oxid − F  trans ,  oxid  (1)where  F  CH 4  is the flux of CH 4  between soil and the atmo-sphere (gCm − 2 d − 1 );  F  P  is plant-mediated transport fromsoil pore water to the atmosphere (gCm − 2 d − 1 );  F  D  is thediffusive flux of CH 4  from water surface to the atmosphere(gCm − 2 d − 1 );  F  E  is the ebullitive CH 4  emission to the at-mosphere;  F  air ,  oxid  is the rate of atmospheric methane oxi-dation (gCm − 2 d − 1 );  F  trans ,  oxid  is the oxidized CH 4  duringplant-mediated transport (gCm − 2 d − 1 ).The concentration of CH 4  in the soil pore water was gov-erned by the following equations: d  [CH 4 ] dt  = f ( [CH 4 ] )  (2) = CH 4 prod  − F  P H  − F  D H  − F  E H  − CH 4 soil ,  oxid where [CH 4 ] is the concentration of CH 4  in water (gCm − 3 );CH 4 prod  is the production of CH 4  in soil pore water(gCm − 3 d − 1 ); CH 4 soil ,  oxid  is the oxidation rate of CH 4  in soilporewater(gCm − 3 d − 1 ); H   isthesoildepthofthefirstlayerfor methane production and oxidation. CH 4  production The production of CH 4  in soil pore water is controlled by theconcentration of DOC and environmental factors (Eq. 2),CH 4 prod  = V  prod ,  max ×[ DOC ][ DOC ]+ Km prod (3) × f (T  soil )  × f( pH ) × f  prod ( vwc ) where  V  prod ,  max  is the maximum rate of CH 4  produc-tion (gCm − 3 d − 1 ), [DOC] is the concentration of DOC(gCm − 3 );  Km prod  is the half-saturation coefficient of CH 4 production (gCm − 3 );  f(T  soil ) is a multiplier that describesthe effect of soil temperature on CH 4  production and oxida-tion;  f  (pH) is a multiplier that describes the effect of soil pHon CH 4  production and oxidation;  f  prod (vwc) is a multiplierthat describes the effect of soil moisture on CH 4  production. CH 4  oxidation Three pathways are considered in the DLEM for CH 4  oxi-dation: (1) atmospheric CH 4  oxidation, also called the diffu-sion processes of CH 4  from the atmosphere to the soil porewater, mainly simulates the oxidation of atmospheric CH 4 in the soil pore water; (2) the process of CH 4  oxidation inthe soil pore water mainly simulates the oxidation of CH 4 which is dissolved in water or accumulated in soil poros-ity; and (3) the process of CH 4  oxidation occurs during theplant-mediated transport of CH 4  from soil pore water to theatmosphere. The DLEM assumes that the process of CH 4 oxidation in soil pore water includes the CH 4  oxidation dur-ing ebullition and diffusion because these two processes onlyoccur in water. Atmospheric CH 4  oxidation Oxidation of atmospheric CH 4  is estimated as: F  air ,  oxid = V  air ,  oxid ,  max × [Atm CH 4 ][Atm CH 4 ]  + Km air ,  oxid (4) × f (T  soil )  × f( pH ) × f  oxid ( vwc ) where  V  air ,  oxid ,  max  is the maximum oxidation rate of at-mospheric CH 4  (gCm − 2 d − 1 );  km air ,  oxid  is the half satu-ration coefficient of atmospheric CH 4  oxidation (gCm − 3 );Biogeosciences, 7, 2673–2694, 2010 www.biogeosciences.net/7/2673/2010/   H. Tian et al.: Terrestrial fluxes of CH 4  and N 2 O over North America 2677[Atm CH 4 ] is the atmospheric CH 4  concentration (gCm − 3 ); f  oxid (vwc) is a multiplier that describes the effect of soilmoisture on atmospheric CH 4  oxidation. Because the atmo-spheric CH 4  oxidation is mainly carried out by soil methan-otrophy, and low soil organic matter means lower soil micro-bial biomass (Conrad, 1996), the DLEM assumes that thereis no atmospheric CH 4  oxidation when soil organic matter isless than 10gCm − 2 . CH 4  oxidation during plant-mediated transport Duringtheprocessofplant-mediatedCH 4  transportfromsoilto the atmosphere, portions of CH 4  will be oxidized at therate of: F  trans ,  oxid = min (5)  V  trans ,  oxid ,  max × F  P F  P + Km trans ,  oxid × f (T  air ), F  P  where  F  trans ,  oxid  is the oxidation rate of CH 4  during plant-mediated transport (gCm − 2 d − 1 ); V  trans ,  oxid ,  max  is the max-imum rate of CH 4  oxidation (gCm − 2 d − 1 );  Km trans ,  oxid  isthe half saturation coefficient of soil CH 4  oxidation duringtransportation (gCm − 2 );  T  air  is the air temperature;  f(T  air ) is a multiplier that represents the effect of air temperature onthe oxidation of CH 4  during plant-mediated transport. Soil pore water CH 4  oxidation The accumulated CH 4  in soil pore water is oxidized at therate of:CH 4 soil ,  oxid  = min  V  soil ,  oxid ,  max  (6) × [CH 4 ][CH 4 ]  + Km soil ,  oxid × f (T  soil ) × f( pH ) × f  oxid ( vwc ),  [CH 4 ] ) where  V  soil ,  oxid ,  max  and  Km soil ,  oxid  are maximum soil porewater CH 4  oxidation rate (gCm − 3 d − 1 ) and half saturationcoefficient of CH 4  oxidation in soil pore water (gCm − 3 ),respectively; [CH 4 ] is the concentration of CH 4  in soil porewater (gCm − 3 ). CH 4  transport In this model, ebullition, diffusion and plant-mediated trans-port, are considered the three pathways by which CH 4  can betransported from soil pore water to the atmosphere. Ebullition The ebullition transport of CH 4  from water to the atmosphereis estimated as: F  E = max  (( [CH 4 ]  − 6 ),  0 )  × H   (7)where  F  E  is the flux of CH 4  from water to the atmospherevia ebullition (gCm − 2 d − 1 ); 6 is the threshold value abovewhich the dissolved CH 4  will form bubbles and leave water(gCm − 3 ), and is equals to 0.5molCH 4  m − 3 (Walter et al.,2001). Because this process occurs in very short time (Walteret al., 2001; Zhuang et al., 2004), the DLEM assumes that allthe dissolved CH 4  above this threshold value will leave watervia bubbles in one day. Plant-mediated transport The plant-mediated CH 4  emission from water to the atmo-sphere is estimated as: F  P = V  plant ,  trans ×  ( [CH 4 ]  −  [CH 4 ] max )  (8) × min   GPPGPP max ,  1  [CH 4 ] max  =  [Atm CH 4 ]  × β  (9)where  F  P  is the CH 4  transport via vascular plant (g Cm − 2 d − 1 );  V  plant ,  trans  is the transport coefficient of CH 4 transportation through plant (md − 1 ), which is set as 0.68(Kettunen, 2003); [CH 4 ] max  is the maximum CH 4  concentra-tion in soil solution (gCm − 3 ); GPP is the gross primary pro-ductivity of current day (gCm − 2 d − 1 ); GPP max  is the maxi-mumdailyGPP(gCm − 2 d − 1 ), whichissetas5inthisstudy; β  is the Bunsen solubility coefficient (0.035mlml − 1 ) (Ya-mamoto et al., 1976). Since there is no report on the plant-mediated transport of CH 4  by woody plant, the DLEM as-sumes that the plant-mediated transport only occurs in herba-ceous biomes;  F  P  is set to zero for all woody ecosystems. Diffusion TheDLEMtreatsthetop0.5mofthesoilprofileasonelayer,and the CH4 generated under water’s surface is assumed tohave a fast diffusion rate to water’s surface. The diffusionestimated here is the exchange of CH 4  between the watersurface and the atmosphere. F  D = V  exchange ×  ( [CH 4 ]  −  [CH 4 ] max )  (10)where  V  exhange  is the exchange coefficient of CH 4  throughthe interface of soil pore water and the atmosphere (md − 1 );it is set as 0.3md − 1 (Happell and Chanton, 1995). Environmental factors affecting methane processes To simulate the environmental effects on methane produc-tion, oxidation and transport, the DLEM considers three en-vironmental factors: soil pH, soil moisture, and temperature.These three factors have been considered as the most impor-tant external factors on CH 4  production, consumption, andtransport (Cao et al., 1995; Huang et al., 1998; Mer andRoger, 2001; Zhuang et al., 2004). The line graphs showingthe environmental controls on CH 4  production and consump-tion could be found in the Fig. 3.In the DLEM, the effect of soil pH on methane productionand oxidation ( f  (pH)) is calculated as a bell shape curve,www.biogeosciences.net/7/2673/2010/ Biogeosciences, 7, 2673–2694, 2010
Related Search
Similar documents
View more...
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