Juba River runs from its headwaters in Ethiopia through Somalia and discharges into the Indian Ocean near Kismayo. Severe droughts recur in Juba River Catchment. This chapter examines the time series of hydrologic droughts and the factors influencing their occurrence in the catchment. The author further demonstrates the application of probability distributions to compute the exceedance probabilities of the hydrologic drought experienced in the catchment in the past, from 2001 to 2014. The fits of the probability distributions are compared to show the best fitting of all distributions. The probability distribution that will reasonably fit well to the data set is adapted to generate graphical plots for forecasting the return periods of drought events in the catchment. Reconstructing past drought episodes and future drought predictions is useful in drought risk management in Baardheere region of Somalia.
TopIntroduction
Probability distributions are used in frequency analysis of hydrological data. A probability distribution is a continuous mathematical expression that determines the probability of a particular event, for example, drought or flood. A probability distribution gives important information about the data, for instance, the flows, how the flow values are changing, whether they are bunched together or spread out, and whether they are symmetrically disposed on the X-axis or not. The distribution also tells the relative frequency of plotted flow values in the population on the X-axis. Knowledge of probability distributions is fundamental to water and environmental professionals (hydrologists, water users, scientists and decision makers) in such tasks as water uses, the operation of water resources and drought risk management. This chapter covers fundamental probability distributions in hydrology: 1. Normal distribution (symmetric), 2. Log-Normal distribution (skew to right), 3. Gumbel (Extreme value type I) distribution (skew to right), 4. Generalized Extreme value (GEV) (any skewness), 5. Weibull distribution (any skewness). Frequency analysis is performed by fitting these probability distribution functions to the discharge variability of Juba River at Baardheere River Gauge Station. The results will be useful in drought risk forecasting and management in Baardheere town, Somalia.
Droughts are a natural phenomenon and can be categorized as meteorological, hydrological and socio-economic droughts. Hydrological drought occurs when low water supply becomes evident, especially in streams after many months of low or no rainfall. The monthly flow sequence of Juba River at Baardheere gauging station was developed using a time series approach to identify past hydrologic droughts events in the period 2001 to 2014. The flow sequence was truncated at a desired truncation level, the mean level. All positive and negative departures from the mean for some selected months were examined using Figure 1. The negative departures from the mean were associated with hydrological droughts; two important parameters were identified from the graph, duration and the anomalies. Generally, the months of February and March were noted to have the greatest drought severity in terms of magnitude of deficit. Major droughts events in the juba river at Baardheere gauge station for the study period occurred in 2002, 2003, 2008 and 2011. In 2011, the severe drought felt in the whole country triggered a famine that resulted in the deaths of about 260,000 people (FAO, 2011); this has made increased emphasis on drought forecasting in the region.
Figure 1. Time series model of hydrologic drought events in Juba River