Timescale dependence of environmental and plant‐mediated controls on CH 4 flux in a temperate fen

This study examined daily, seasonal, and interannual variations in CH 4 emissions at a temperate peatland over a 5‐year period. We measured net ecosystem CO 2 exchange (NEE), CH 4 flux, water table depth, peat temperature, and meteorological parameters weekly from the summers (1 May to 31 August) of...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of Geophysical Research 2007-03, Vol.112 (G1)
Hauptverfasser: Treat, Claire C., Bubier, Jill L., Varner, Ruth K., Crill, Patrick M.
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:This study examined daily, seasonal, and interannual variations in CH 4 emissions at a temperate peatland over a 5‐year period. We measured net ecosystem CO 2 exchange (NEE), CH 4 flux, water table depth, peat temperature, and meteorological parameters weekly from the summers (1 May to 31 August) of 2000 through 2004 at Sallie's Fen in southeastern New Hampshire, United States. Significant interannual differences, driven by high variability of large individual CH 4 fluxes (ranging from 8.7 to 3833.1 mg CH 4 m −2 d −1 ) occurring in the late summer, corresponded with a decline in water table level and an increase in air and peat temperature. Monthly timescale yielded the strongest correlations between CH 4 fluxes and peat and air temperature (r 2 = 0.78 and 0.74, respectively) and water table depth (WTD) (r 2 = 0.53). Compared to daily and seasonal timescales, the monthly timescale was the best timescale to predict CH 4 fluxes using a stepwise multiple regression (r 2 = 0.81). Species composition affected relationships between CH 4 fluxes and measures of plant productivity, with sedge collars showing the strongest relationships between CH 4 flux, water table, and temperature. Air temperature was the only variable that was strongly correlated with CH 4 flux at all timescales, while WTD had either a positive or negative correlation depending on timescale and vegetation type. The timescale dependence of controls on CH 4 fluxes has important implications for modeling.
ISSN:0148-0227
2156-2202
DOI:10.1029/2006JG000210