Martens, Dean (USDA-ARS Tucson AZ, Southwest Watershed Research Center, 2000 E. Allen, Tucson, AZ, 85719; Phone: 520 670-6380 x156; Fax: 520 670-5550; Email: dmartens@tucson.ars.ag.gov)

 

Moisture Controls on Trace Gas Fluxes in Semiarid Soils

 

D. Martens *, J. E. T. McLain

 

In semiarid soils, variability in seasonal soil moisture (SM) and temperature (T) can alter ecosystem/atmosphere exchange of CO2 and the trace gases N2O and CH4 potentially acting as a negative or a positive feedback to global warming. The impact of SM inputs (warm summer monsoon vs. cool winter rain) on fluxes of these gases was monitored in three vegetation zones from July 2002 – September 2003 in southeastern Arizona. The soil C content (0-5 cm) in the vegetation zones ranged from 5.2 g C in the bare site, 13.4 g C in the sacaton (Sporobolus wrightii) site to 30 g organic C kg-1 soil under established mesquite (Prosopis velutina). Carbon dioxide and N2O emissions during the 15 month study were highly dependent on available SM and T. During heavy rains of the 2002 monsoon (81% of 2002 rainfall), large differences in soil C content did not correlate with variations in CO2 production, as efflux from the three sites averaged 124.4 ± 1.9 g CO2 m-2. During the fall through spring period, CO2 production efflux for the three sites averaged 93.5 ± 14.7 g m-2. In 2003, limited monsoon rain (59% of 2003 rainfall) CO2 emissions were reduced (compared with 2002) by 19%, 40% and 30% in the mesquite, open, and sacaton sites, respectively to 88.4± 16.3 g m-2 (average 29% reduction). Isotopic analysis of CO2 respired showed that the majority of C respired (50 to 98%) reflected the isotopic signal of the site vegetation. Nitrous oxide emissions during the 2002 monsoon season averaged 21.1 ± 13.4, 2.1 ± 4.4, and 3.9 ± 5.2 ug N2O m-2 h-1 in the mesquite, open, and sacaton sites for an average N2O monsoon flux of 16.2 mg m-2. During the fall through spring period, N2O production efflux for the three sites averaged 33.7 ± 18.6 mg m-2. Limited rainfall during the 2003 monsoon reduced N2O emissions by 47% in the mesquite, but N2O fluxes increased in the sacaton (5%) and open (55%) compared with 2002 for an average monsoon emission of 12.3 ± 6.7 mg m-1> (24% reduction). Following a dry winter and spring 2002 (15 mm rain), premonsoon CH4 consumption in the three vegetation zones was close to zero, but following monsoon precipitation (238 mm rain), the CH4 sink averaged 45.4 ± 15.8 mg m-2. From October through April 2003 across all sites the CH4 sink averaged 146.1 ± 23.6 mg m-2. Methane oxidation was a subsurface process as oxidation rates were measured at the 5-50 cm depth in laboratory incubations suggesting that as the soil surface dried, CH4 oxidation activity shifted deeper in the sandy soils, thus allowing for high net oxidation flux rates while the surface soil was extremely dry.  This study measured ecosystem greenhouse gas potential (GHGP) that averaged 128.5 ± 23.6 g m-2 with the 2002 monsoon with a post monsoon GHGP of 100.1 ± 19.1 g m-2. A 60% reduction in monsoon precipitation in 2003 reduced GHGP to 91.0 ± 18.1 g m-2 (29% reduction) that suggest predicted shifts in annual precipitation patterns from a majority as summer rain to greater winter precipitation may reduce soil CO2 and N2O emissions while promoting CH4 oxidation rates in semiarid zones of the Southwest, potentially acting as a negative feedback for future global warming.