Environmental Flow Allocation as a Practical Aspect of IWRM - River Conservation and Management - page 50

could be retained in the river without affecting
agricultural productivity on the Usangu Plains. This
is particularly the case where alternative water
is provided to tail-enders and domestic use. An
estimated 2-5m 3 s − 1 could be retained in the
rivers through improved canal management, while
2-7m 3 s − 1 could be retained through closure of
the canals during the dry season. However, these
are average figures and savings during the late dry
season (October and November) are significantly
lower, averaging below 2m 3 s − 1 . This could provide
a long-term solution to the downstream flow
problem, but will require considerable negotiation
and resources to implement.
Use of groundwater for irrigation. This is unlikely
to provide a solution, since the abstraction of
groundwater is likely to reduce baseflows to the
river. A thorough investigation of the groundwater
connections to the river would be necessary before
such an option could be evaluated.
Engineering the wetland: this comprises many
options, including cutting a channel through
the wetland, or piping water through it, so as
to minimize evapotranspiration losses. Another
option would be to cut a channel through the
geological sill which impounds the water in the
wetland, so as to lower the water level at which
water would flow through into the downstream
river reaches. The costs of such options would be
a reduction in the area and functioning of the
wetland, and these costs are felt to be too serious
to consider at present.
The option of choice for the short term would be to
provide storage and a transfer from the Ndembera
tributary. This will require negotiations with the
downstream rice farmers who currently use water
from the Ndembera. At the same time, the longer-
term solution of more efficient irrigation should
be initiated. The major lessons emerging from this
project are:
Where a river is already over-allocated, short-
term solutions (even if they are compromises)
are required to minimize long-term irreversible
damage.
In such cases, detailed research and a time-
consuming environmental flow assessment is not
the priority. Where there is no flow in a formerly

perennial river, it does not require much expertise
or data to conclude that some flow needs to be
restored. The precise flow rates can be adjusted
later. The adoption of an adaptive management
approach is useful in such cases.
The immediate investigation of different options
for flow restoration provided a pragmatic use of
resources for this project, providing managers with
possible scenarios that were of immediate use.
Whether these will be implemented depends on
many other variables, which are not within the
control of the project team. Eventually, the success
of environmental flows, as with all other water
management issues, becomes a societal choice,
dependent on political, economic and social factors.
Even with the best scientific and monitoring
information available, environmental flows will
only be implemented where there is a general
appreciation of, and support for, the importance of
sustaining flows for conservation, where there is
the political will to use rivers within sustainable
limits, and where there is a recognition of the
long-term economic benefits of protecting water
resources, rather than the short-termmaximization
of consumptive uses.

Conclusions

These case studies have shown that environmental
flow allocation has to be considered on a river-
specific basis, even when the circumstances seem
very similar. Other case studies, on the Rio Conchos
in Mexico, the Ganga in India and the Sao
Francisco in Brazil, have demonstrated that:
Trans-border river issues, which are often seen
as complicating the management procedures, can
also be used to facilitate the implementation
of environmental flows. For example, the Rio
Conchos flows into the Rio Grande, which
forms part of the US/Mexico border. There is
a treaty obligation for Mexico to provide flows
into the Rio Grande, and such flows can be
distributed according to the environmental flow
recommendations.
Large rivers also require environmental flows,
but assessment methods (developed for small