Increased rainfall variability and reduced rainfall amount decreases soil CO2 flux in a grassland ecosystem

TitleIncreased rainfall variability and reduced rainfall amount decreases soil CO2 flux in a grassland ecosystem
Publication TypeJournal Article
Year of Publication2005
AuthorsHarper, CW, Blair, JM, Fay, PA, Knapp, AK, Carlisle, JD
JournalGlobal Change Biology
Pagination322 -344
Accession NumberKNZ00939

Predicted climate changes in the US Central Plains include altered precipitation regimes with increased occurrence of growing season droughts and higher frequencies of extreme rainfall events. Changes in the amounts and timing of rainfall events will likely affect ecosystem processes, including those that control C cycling and storage. Soil carbon dioxide (CO2) flux is an important component of C cycling in terrestrial ecosystems, and is strongly influenced by climate. While many studies have assessed the influence of soil water content on soil CO2 flux, few have included experimental manipulation of rainfall amounts in intact ecosystems, and we know of no studies that have explicitly addressed the influence of the timing of rainfall events. In order to determine the responses of soil CO2 flux to altered rainfall timing and amounts, we manipulated rainfall inputs to plots of native tallgrass prairie (Konza Prairie, Kansas, USA) over four growing seasons (1998–2001). Specifically, we altered the amounts and/or timing of growing season rainfall in a factorial combination that included two levels of rainfall amount (100% or 70% of naturally occurring rainfall quantity) and two temporal patterns of rain events (ambient timing or a 50% increase in length of dry intervals between events). The size of individual rain events in the altered timing treatment was adjusted so that the quantity of total growing season rainfall in the ambient and altered timing treatments was the same (i.e. fewer, but larger rainfall events characterized the altered timing treatment). Seasonal mean soil CO2 flux decreased by 8% under reduced rainfall amounts, by 13% under altered rainfall timing, and by 20% when both were combined (P<0.01). These changes in soil CO2 flux were consistent with observed changes in plant productivity, which was also reduced by both reduced rainfall quantity and altered rainfall timing. Soil CO2 flux was related to both soil temperature and soil water content in regression analyses; together they explained as much as 64% of the variability in CO2 flux across dates under ambient rainfall timing, but only 38–48% of the variability under altered rainfall timing, suggesting that other factors (e.g. substrate availability, plant or microbial stress) may limit CO2 flux under a climate regime that includes fewer, larger rainfall events. An analysis of the temperature sensitivity of soil CO2 flux indicated that temperature had a reduced effect (lower correlation and lower Q10 values) under the reduced quantity and altered timing treatments. Recognition that changes in the timing of rainfall events may be as, or more, important than changes in rainfall amount in affecting soil CO2 flux and other components of the carbon cycle highlights the complex nature of ecosystem responses to climate change in North American grasslands.