535means of salinity management. Skaggs and van Schilfgaarde ( 1999 ) document 1400
pages of the theory and practice of drainage, with one chapter evaluating options for
control in the Great Plains in North America given to management of dryland
salinity.
El-Ashry and Duda ( 1999 ) define drainage as the preeminent means of managing
salinity in irrigated lands, defining its need for use on 110 M ha of irrigated area.
However, they also define a need to use drainage to intercept groundwater in dry-
land areas of North America, India, Iran and Australia, though no specific examples
are given.
Application of deep drainage to dryland salinity in Western Australia has been
applied within several catchment studies and analysed by Coles et al. ( 1999 ) and
updated by Chandler and Coles ( 2003 ). These reviews report the minimal impact by
open drains, of approximately 2 m depth, where impacts of less than 20–40 m from
the drain were observed from the majority of the 22 case study sites evaluated. The
dominant factors reducing the impact were the low permeability of the soils, limited
gradient for flow and thickness of the aquifer relative to drain depth, contributing
groundwater to the saline areas.
These results are reinforced by a drainage benchmarking study by Stuart-Street
et al. ( 2012 ) at a further eight sites; the authors noting a similar impact of the drains.
More significantly, they note that reduced rainfall in the past decade confounded
their ability to separate the effect of the treatment (drains) and climate (reduced
rainfall and recharge) from observed watertable and soil salinity impacts.
Exceptional impacts (100–200 m radius from the drain) have been described at
sites by Ali et al. ( 2004 ) near Narembeen, where the hydraulic conductivity of the
subsurface was high and the drain depth was at about 3 m, and Kobryn et al. ( 2015 )
near Dumbleyung where parallel drains were installed to between 2-3 m deep and
crop recovery was attributed to watertable control and reduced soil salinity.
Aquifer pumping, siphons and relief wells have also been trialled and shown to
have variable impact. George ( 1990 , 1992 ) studied aquifer properties and the effect
pumping at recharge and discharge sites and concluded while bores were capable of
lowering watertables, and supplying brackish groundwater supplies, their influence
was limited by the same parameters to those above and their economics were sub-
optimum (George et al. 1997 ).
In specific cases, siphons have been used to manage watertables causing saline
seeps on sloping land (< 2 %) and thereby reduce one of the major economic limita-
tions of available and affordable power costs. Seymour and George ( 2004 ) showed
that by designing siphons that attain critical velocity for dissolved gas (the principal
model of failure), water tables were lowered beneath seepage scalds (hillside seeps)
of 1–10 ha by cumulative flows of 1–2 L sec−^1. They also showed drainage to be of
lower environmental impact than the episodic flux of salt from the degrading seeps.
Salinity in Dryland Agricultural Systems: Challenges and Opportunities