El-Nino/Southern Oscillation (ENSO): Recent Evolution and Possibilities for Long Range Flow Forecasting in the Brahmaputra-Jamuna River

The El Nino-Southern Oscillation is the dominant pattern of short-term climate variation, and
is therefore of great importance in climate studies. Some recent studies showed the
teleconnection between stream flow and the El-Nino Southern Oscillation (ENSO) of the
equatorial Pacific Ocean. This paper presents an overview of the relationship between ENSO
and stream flow in the Brahmaputra-Jamuna and the potential for wet season flow
forecasting. This seasonal forecast of stream flow is very invaluable to the management of
land and water resources, particularly in Bangladesh to improve the predictability of severe
flooding. Over the years, large investments have been made to build physical infrastructure
for flood protection, but it has been proved that it is not feasible, both economically and
technically, to adopt solely structural mitigation approach. The choice of non-structural
measures in this country focused mainly on flood forecasting because many of the nonstructural
measures including flood plain zoning, compulsory acquisition of flood prone land,
relocation etc have also been proved inappropriate for Bangladesh.
The aim of this research is to find out an effective and long-lead flow forecasting method with
lead time greater than hydrological time scale, using El Nino-Southern Oscillation index.
Some studies indicate that SST can be predicted one to two years in advance using several
ocean/ coupled ocean atmosphere models, therefore the ability to predict flow patterns in
rivers will be highly enhanced if a strong relationship between river discharge and ENSO
exists, and is quantified. With this view, to assess the strength of teleconnection between river
flow and ENSO, at first correlation analyses between ENSO indices of any year and wet
season flow of that year have been done. Here sea surface temperature (SST) has been
used as ENSO index. This correlation analysis demonstrates a noteworthy relationship
between natural variability of average flow of the months July-August-September (JAS) of the
Brahmaputra-Jamuna River with SST of the corresponding months. Then discriminant
prediction approach, also known as “Categoric Prediction” has been used here for the
assessment of long range flood forecasting possibilities. This approach will be able to forecast
the category of flow (high, average or low) using the category of predictor (predicted SST) at
a sufficient lead time. In order to judge the forecast skill, a synoptic parameter “Forecasting
Index” has also been used. This discriminant approach will improve the forecasting lead-time
while the hydrologic forecast through rainfall-runoff modeling could provide a lead time on the
order of the basin response time, which is several days or so. As the Ganges–Brahmaputra
river basin is one of the most populous river basins of the world and is occupied by some
developing countries of the world like Bangladesh, any reduction in the uncertainty about the
flood in the Brahmaputra-Jamuna River would contribute a lot to the improvement in flow
forecasting as well as to the economic development of the country.