Predicting the spread of groundwater pollution demands a detailed unde
rstanding of the physical, chemical and microbial processes that contr
ol contaminant mobility in aquifers. Many field studies have been carr
ied out around pollutant sources in an attempt to understand these pro
cesses, but quantitative results are often difficult to obtain because
of the number of assumptions that have to be made about the flow regi
me or the source term which has given rise to the pollution. Models ca
n be constructed with emphases on different processes to describe the
known distribution of contaminants at any one time. However, if these
models are to be used for predictive purposes, or to help remediation,
it is important to identify and quantify individual processes precise
ly by independent or direct methods and not to rely on inference alone
. Laboratory tests suffer from the fact that aquifer material has to b
e sampled and transferred to the laboratory. In the process, the sampl
e may be disturbed physically thus altering its porosity, permeability
and dispersive properties. It may be oxidised, thereby altering its c
hemistry, and the numbers, activity and character of any microbial pop
ulation may change. In situ tracer experiments attempt to overcome the
limitations of the laboratory by maintaining natural conditions, but
at the same time allowing the injection of solute to be accurately def
ined and the hydraulic regime to be well controlled and monitored.Exam
ples are given showing how integrated laboratory and field approaches
have been used to study: (1) organic degradation in a pollution plume
resulting from the disposal of industrial wastes and (2) the role of c
olloids in transporting radionuclides in an intergranular aquifer.