Go to AfricaBib home

Go to AfricaBib home Water and Africa Go to database home

bibliographic database

Previous page New search

The free AfricaBib App for Android is available here

Periodical article Periodical article Leiden University catalogue Leiden University catalogue WorldCat catalogue WorldCat
Title:Exploring the frequency of hydroclimate extremes on the River Niger using historical data analysis and Monte Carlo methods
Authors:Siebert, Asher
Ward, M. Neil
Periodical:African Geographical Review (ISSN 1937-6812)
Geographic term:West Africa
Subjects:Niger River
Abstract:Flooding and low-flow events along the Niger River in West Africa have major societal impacts, motivating this paper's historical flow analysis and stochastic simulation of plausible future flow scenarios. The known epochal evolution of mean streamflow since 1950 (wet epoch, dry epoch and moderate recovery) is analyzed, with a focus on climatological flood month at three representative locations, and on associated changes in distribution (e.g. standard deviation and percentiles). The authors especially note the relatively small change in the (high flow) 90th percentile across the three epochs. Following the historical analysis, a statistical simulation framework is developed to explore the sensitivity of threshold-crossing events (TCEs) to plausible assumptions about hydroclimatic evolution over 2010-2040. The flow changes are assumed to reflect varying combinations of the following: systematic global change (GC), natural multidecadal variability (MDV) and interannual variability (IV). Results show that for a 1 in 10 TCE in the historical period, a 20% streamflow trend over 2010-2040 (no change in IV) can produce a factor four difference in average TCE frequency. Adding MDV signals of magnitudes that may be considered conservative for the region, enhances the sensitivity further (>50%). The results highlight how risk management strategies need to integrate GC and MDV impacts. Bibliogr., note, sum. [Journal abstract]