Premium
Variations in dissolved CO 2 and CH 4 in a first‐order stream and catchment: an investigation of soil–stream linkages
Author(s) -
Hope Diane,
Palmer Sheila M.,
Billett Michael F.,
Dawson Julian J. C.
Publication year - 2004
Publication title -
hydrological processes
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.222
H-Index - 161
eISSN - 1099-1085
pISSN - 0885-6087
DOI - 10.1002/hyp.5657
Subject(s) - peat , soil water , riparian zone , hydrology (agriculture) , environmental science , drainage basin , soil science , geology , ecology , geotechnical engineering , cartography , habitat , geography , biology
Abstract Spatial and seasonal variations in CO 2 and CH 4 concentrations in streamwater and adjacent soils were studied at three sites on Brocky Burn, a headwater stream draining a peatland catchment in upland Britain. Concentrations of both gases in the soil atmosphere were significantly higher in peat and riparian soils than in mineral soils. Peat and riparian soil CO 2 concentrations varied seasonally, showing a positive correlation with air and soil temperature. Streamwater CO 2 concentrations at the upper sampling site, which mostly drained deep peats, varied from 2·8 to 9·8 mg l −1 (2·5 to 11·9 times atmospheric saturation) and decreased markedly downstream. Temperature‐related seasonal variations in peat and riparian soil CO 2 were reflected in the stream at the upper site, where 77% of biweekly variation was explained by an autoregressive model based on: (i) a negative log‐linear relationship with stream flow; (ii) a positive linear relationship with soil CO 2 concentrations in the shallow riparian wells; and (iii) a negative linear relationship with soil CO 2 concentrations in the shallow peat wells, with a significant 2‐week lag term. These relationships changed markedly downstream, with an apparent decrease in the soil–stream linkage and a switch to a positive relationship between stream flow and stream CO 2 . Streamwater CH 4 concentrations also declined sharply downstream, but were much lower (<0·01 to 0·12 mg l −1 ) than those of CO 2 and showed no seasonal variation, nor any relationship with soil atmospheric CH 4 concentrations. However, stream CH 4 was significantly correlated with stream flow at the upper site, which explained 57% of biweekly variations in dissolved concentrations. We conclude that stream CO 2 can be a useful integrative measure of whole catchment respiration, but only at sites where the soil–stream linkage is strong. Copyright © 2004 John Wiley & Sons, Ltd.