Abstract
Abstract. In seasonal flow forecasting applications, one factor which can help predictability is a significant hydrological response time between rainfall and flows. On account of storage influences, large lakes therefore provide a useful test case although, due to the spatial scales involved, there are a number of modelling challenges related to data availability and understanding the individual components in the water balance. Here some possible model structures are investigated using a range of stochastic regression and transfer function techniques with additional insights gained from simple analytical approximations. The methods were evaluated using records for two of the largest lakes in the world – Lake Malawi and Lake Victoria – with forecast skill demonstrated several months ahead using water balance models formulated in terms of net inflows. In both cases slight improvements were obtained for lead times up to 4–5 months from including climate indices in the data assimilation component. The paper concludes with a discussion of the relevance of the results to operational flow forecasting systems for other large lakes.
Highlights
One of the challenges in seasonal flow forecasting is that the lead times of interest often far exceed the hydrological response time of catchments
A common finding is that forecast skill may arise as much from the representation of antecedent conditions as from the meteorological inputs, with the balance depending on factors such as lead times and season, as well as location (e.g. Robertson and Wang, 2012; Greull et al, 2016; Mendoza et al, 2017)
In particular the 1961/62 event for Lake Victoria has previously been investigated in detail with some evidence of a regional shift in climate at that time (e.g. Sutcliffe and Parks, 1999; Nicholson and Selato, 2000)
Summary
One of the challenges in seasonal flow forecasting is that the lead times of interest often far exceed the hydrological response time of catchments. Perhaps the greatest success to date has been in snowmelt forecasting for basins with a significant winter snowpack and typically this has been based on statistical techniques or sampling of historic records for input to hydrological models; for example using an ensemble streamflow prediction approach (Day, 1985; Wood and Schaake, 2008). These techniques have been applied more widely and other more recent developments include the use of seasonal rainfall forecasts, climate indices and ensemble Kalman filter approaches A common finding is that forecast skill may arise as much from the representation of antecedent conditions as from the meteorological inputs, with the balance depending on factors such as lead times and season, as well as location (e.g. Robertson and Wang, 2012; Greull et al, 2016; Mendoza et al, 2017)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
