With acid rain as an increasing environmental problem in China there is an
urgent need for evaluation and improvement of soil acidification models. We
have evaluated the applicability of a range of simple models describing al
uminum release for a common Chinese forest soil type. Data from field inves
tigations, laboratory experiments and soil extractions are compared and use
d to test several equilibrium models. The commonly used equilibrium with gi
bbsite (Al(OH)(3)) and an ion exchange model did not satisfactorily describ
e the observed data, neither did equilibrium reactions with other minerals.
Better fit was obtained assuming equilibrium between dissolved Al3+ and al
uminum extracted by pyrophosphate or copper chloride, often assumed to repr
esent organically bound aluminum pools in the soil. The sizes of easily ext
ractable aluminum pools were similar to values observed in Europe and North
America. The smallest pools correspond to a few decades of the aluminum fl
ux in soil water. In the laboratory experiments aluminum is re-supplied to
easily extractable pools from more strongly bound phases, but it is not cle
ar to what extent this occurs in the field. At a given pH, the aluminum con
centrations were higher in samples collected in the field compared to the l
aboratory experiments, showing the importance of soil structure and water f
low paths in the field. Our results indicate that it is possible to model a
luminum chemistry in soil water from Chinese subtropical forest soils bette
r than using gibbsite equilibrium. However, we are far from having a reliab
le, mechanistic model. This reduces the confidence in long-term model resul
ts. (C) 1999 Elsevier Science B.V. All rights reserved.