Unformatted text preview:

Effects of an experimental drought and recovery on soilemissions of carbon dioxide, methane, nitrous oxide, andnitric oxide in a moist tropical forestERIC A. DAVIDSON*,DANIELC.NEPSTAD*,FRANC¸ OISE YOKO ISHIDAw andPAULO M. BRANDOz,§*The Woods Hole Research Center, 149 Woods Hole Road, Falmouth, MA 02540-1644, USA, wCENA, University of Sao Paulo,Av. Centena´rio, 303 Piracicaba, SP, Brazil, zInstituto de Pesquisa Ambiental da Amazoˆnia, Av. Rui Barbosa, 136 Santare´m, PA,Brazil, §Department of Botany and School of Natural Resources and Environment, University of Florida, 220 Bartram Hall,Gainesville, FL 118526, USAAbstractChanges in precipitation in the Amazon Basin resulting from regional deforestation,global warming, and El Nin˜ o events may affect emissions of carbon dioxide (CO2),methane (CH4), nitrous oxide (N2O), and nitric oxide (NO) from soils. Changes in soilemissions of radiatively important gases could have feedback implications for regionaland global climate. Here, we report the final results of a 5-year, large-scale (1 ha)throughfall exclusion experiment, followed by 1 year of recovery with natural through-fall, conducted in a mature evergreen forest near Santare´m, Brazil. The exclusionmanipulation lowered annual N2O emissions in four out of five treatment years(a natural drought year being the exception), and then recovered during the first yearafter the drought treatment stopped. Similarly, consumption of atmospheric CH4increased under drought treatment, except during a natural drought year, and it alsorecovered to pretreatment values during the first year that natural throughfall waspermitted back on the plot. No treatment effect was detected for NO emissions duringthe first 3 treatment years, but NO emissions increased in the fourth year under theextremely dry conditions of the exclusion plot during a natural drought. Surprisingly,there was no treatment effect on soil CO2efflux in any year. The drought treatmentprovoked significant tree mortality and reduced the allocation of C to stems, butallocation of C to foliage and roots were less affected. Taken together, these resultssuggest that the dominant effect of throughfall exclusion on soil processes during this6-year period was on soil aeration conditions that transiently affected CH4,N2O, and NOproduction and consumption.Keywords: Amazon Basin, Brazil, climate change, CH4,CO2,N2O, nitrogen, NO, soil carbonReceived 28 December 2007 and accepted 5 May 2008IntroductionSome climate models predict that the drought episodesand seasonal water deficits in the eastern and southernAmazon Basin may be more common and more severeas global climatic change proceeds during the 21stcentury (Cox et al., 2004; Li et al., 2006; Malhi et al.,2008). Global warming may also increase the intensityof El Nin˜o Southern Oscillation (ENSO) events (Hansenet al., 2006), which cause severe drought in the easternAmazon Basin (Nepstad et al., 1999). In 1998, a particu-larly severe El Nin˜o episode was associated with pro-longed drought in eastern and northern Amazonia(Nepstad et al., 1999, 2004; Alencar et al., 2006). In 2005,warming of the tropical North Atlantic triggered theworst drought in 40 years across the southern AmazonBasin (Brown et al., 2006; Araga˜o et al., 2007). Tropicalrainfall inhibition by smoke (Rosenfeld, 1999; Andreaeet al., 2004) may exacerbate this general drying trend inthis moist tropical forest region.Reduced precipitation may have important feedbackeffects on climate change by altering soil emissions ofradiatively important gases, such as CO2,CH4,N2OCorrespondence: Eric A. Davidson, tel. 1 1 508 540 9900,fax 1 1 508 540 9700, e-mail: [email protected] Change Biology (2008) 14, 2582–2590, doi: 10.1111/j.1365-2486.2008.01694.xr 2008 The Authors2582 Journal compilation r 2008 Blackwell Publishing Ltd(Forster et al., 2007), and NO (NO is not, itself, a green-house gas, but it is a precursor to the formation oftropospheric ozone, which is a greenhouse gas; Lammel& Gral, 1995). Upland forest soils of the tropics areknown to be important sources of N2O (Matson &Vitousek, 1990) and NO (Davidson & Kingerlee, 1997)and sinks for CH4(Potter et al., 1996). Both primaryproductivity and respiration are high in many tropicalecosystems, resulting in large emissions of CO2fromsoils (Davidson et al., 2000b).Variation in precipitation influences trace gas emis-sions by affecting soil water content and soil aeration,which, in turn, affects microbial processes of productionand consumption of these trace gases (Davidson &Schimel, 1995; Davidson et al., 2000a). Climate changecan also alter root turnover, litterfall, decomposition,and mineralization, which would, in turn, affect theavailability of carbon and nitrogen substrates for tracegas production.We previously reported on the first 3 treatment yearsand 2 pretreatment years of trace gas flux measure-ments in a large-scale (1 ha) throughfall experimentmanipulation conducted in the Tapajo´s National Forest,near Santare´m, Para´, Brazil (Davidson et al., 2004). Here,we report the final results of the entire throughfallexclusion manipulation experiment, including pretreat-ment, 5 years of throughfall exclusion, and 1 year ofposttreatment recovery. The previously reported resultsfor effects on NO, N2O, CH4, and CO2are generallyreconfirmed, although a natural drought that occurredduring the fourth year of the treatment yielded novelresults. Finally, we report on the first year of recoveryafter permitting throughfall to return at natural rates inthe exclusion plot.Materials and methodsThe study area and methods employed here are thesame as those described by Davidson et al. (2004). Wepresent an abbreviated description here.Study areaThe Tapajo´s National Forest, located in east centralAmazonia (2.89681S, 54.95191W), receives 600–3000 mmof rain each year, with a mean of 2000 mm, most of whichfalls during the wet season from January to June (Fig. 1a).The forest is situated on a terrace of Tertiary sedimentscapped by the Belterra Clay Formation (Clapperton,1993). The Oxisol soil (Haplustox) is acidic (pH 4–5), isdominated by kaolinite clay minerals (60–80% clay), andis free of hardpan or iron oxide concretions in the upper12 m; the water table is more than 100 m deep. The foresthas emergent trees up to 55 m in height, with a contin-uous canopy at approximately 30m (Nepstad et al., 2002).Experimental designTwo 1 ha plots were identified from an initial surveyof 20


View Full Document

UA ECOL 596L - Research Paper

Download Research Paper
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Research Paper and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Research Paper 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?