Long-term and seasonal variations in CO2: linkages to catchment alkalinitygeneration

Citation
Sa. Norton et al., Long-term and seasonal variations in CO2: linkages to catchment alkalinitygeneration, HYDROL E S, 5(1), 2001, pp. 83-91
Citations number
38
Categorie Soggetti
Earth Sciences
Journal title
HYDROLOGY AND EARTH SYSTEM SCIENCES
ISSN journal
10275606 → ACNP
Volume
5
Issue
1
Year of publication
2001
Pages
83 - 91
Database
ISI
SICI code
1027-5606(200103)5:1<83:LASVIC>2.0.ZU;2-L
Abstract
As atmospheric emissions of S have declined in the Northern Hemisphere, the re has been an expectation of increased pH and alkalinity in streams believ ed to have been acidified by excess S and N. Many streams and lakes have no t recovered. Evidence from East Bear Brook in Maine, USA and modelling with the groundwater acid-base model MAGIC (Cosby et al. 1985a,b) indicate that seasonal and yearly variations in soil PCO2 are adequate to enhance or eve n reverse acid-base (alkalinity) changes anticipated from modest decreases of SO4 in surface waters. Alkalinity is generated in the soil by exchange o f H+ from dissociation of H2CO3, which in turn is derived from the dissolvi ng of soil CO2. The variation in soil PCO2 produces an alkalinity variation of up to 15 mu eq L-1 in stream water. Detecting and relating increases in alkalinity to decreases in stream SO4 are significantly more difficult in the short term because of this effect. For example, modelled alkalinity rec overy at Bear Brook due to a decline of 20 mu eq SO4 L-1 in soil solution i s compensated by a decline from 0.4 to 0.2% for soil air PCO2. This compens ation ability decays over time as base saturation declines. Variable PCO2 h as less effect in more acidic soils. Short-term decreases of PCO2 below the long-term average value produce short-term decreases in alkalinity, wherea s short-term increases in PCO2 produce shortterm alkalization. Trend analys is for detecting recovery of streams and lakes from acidification after red uced atmospheric emissions will require a longer monitoring period for stat istical significance than previously appreciated.