review of related literature

A flood occurs when more water than a river can carry is introduced. When a water course is too small to contain the flow, the water overflows the banks. The result of this overflow may produce Nuisance flooding, Damaging flooding or Devastating flooding.
Nuisance floods may result in inconveniences, such as wet feet and tyre spray. Damaging floods go on to soak flooring, carpeting and first floor furniture. Devastating flood can wash off buildings and vehicles downstream as well as take lives.
Floods are environmental hazards that occur regularly every year in different parts of the country (NEST 1991). Flood often occur when runoffs exceed the carrying capacity of the receiving water body, (Tunde 1987). When runoffs from the river tributaries are directed along a river bank or shoreline, flooding often results.
The impact of floods on the built environment is overwhelmingly devastating. Agricultural products are constantly washed away by flood. Urban roads are sometimes rendered impassable especially during heavy down pour. Floods can disrupt academic programmes by either trapping scholars or not allowing them gain access into their school premises. Flood also conveys water-borne diseases or areas that play host to it. Flood reduces the market value of a launched poverty and reduces the salability of housing unit (Ritter 1990).
In his work however, Author (1978), suggested that one of the best way to check flooding is to keep an area’s drainage system open and to control land use development in flood plains. According to Onyeali (1978), the benefits of floods far weighs below the negative costs, thus flooding is both unwanted and unwholesome natural disaster as far as environmental standards are concerned.
Yoshino (1993) states that flooding is a common extreme event that poses challenges for the insurance industry and the public (in planning and protecting infrastructure). Although humans possess considerable adaptive management capacity to deal with floods, flooding could be a more problem with climate change, even if average precipitation decreases. Although rain causes flooding, large rain storm do not always create flooding. The size of a flood depends not only on the amount of rainfall, but also on the conditions within the watershed before and during the storm. When rain falls on the very wet watershed that is unable to absorb more water, or a very large amount of rainfalls on a dry watershed faster than it can be absorbed, the water runs off.
Bates et al (1995) used a stochastic weather generator coupled with two daily rainfall runoff models of annual maximum monthly runoff series of climating settings. In five cases, the series were noticeably higher for a changed climate than for the present day. The size of projected changes were found to depend on the rainfall-runoff model chosen.
According to Frank and Siever (1978), floods are an extreme case of an increased discharge that results from a short- term balance between input and output. As the discharge increases, the flow velocity in the channel increases and the water gradually fills the channel. Ibe (1990), has found that large scale protective engineering measures for controlling flood hazards are impractical in the coasts because of high costs to the area. Instead, low costs, low technology, but effective measures such as permeable non-concrete floating break waters, artificial raising of beach, elevations, timber goins and so forth are considered to be more sensible. He also noted that “fortunately outside the urbanized centres, the coasts are in pristine condition and largely uninhabited”. Where coasts are deemed highly vulnerable, total ban of new development is absolutely necessary.
However, to stimulate river flows, Gellen (1991), used a daily rainfall-runoff model assuming an increase in rainfall during winter. He found more frequent floods, with flows remaining above high threshold for longer period. The mean annual flood peak increased between 2-10% under the scenario used, with the greatest increase in the most responsive catchments. Button et al (1992), used the same model and scenario to estimate the possible changes in flood frequency in a small catchments and found the same result; the mean annual flood increased by 10%. Liettenmaier and Gan (1990), found an increase in rain generated floods, with a consequent increase in the occurrence of a particular discharge being exceeded.
In an examination, Knox (1993), examined a 7,000-year geological record of overbank floods for upper Mississippi River tributaries and found that changes in mean annual temperature of about 1-20c of less than 10-20% can cause large and abrupt adjustments in the magnitude and frequencies of flood kwadijk and Middelkoop (1994), investigated potential changes in flood risk and discovered large changes in the frequency of a given threshold events; an increase in precipitation and a rise in temperature would lead to a major increase in both flood frequencies and risk inundation. Liu simulated increase in variability of floods over time and in frequencies in both northern and southern China.
Beran and Arnell (1995) have shown that, assuming statistical properties similar to those of British rivers, a 10% increase in the mean- with no change in the year to year variability of floods – would result in a current 10- year flood occurring on average of every 7 years. Although the potential impact of climate change on the occurrence of flood disaster has been alluded to frequency in popular accounts of global warning, there have been very few studies addressing the issues explicitly. This is largely because it is very difficult to define credible scenarios for changes in flood-producing climatic events (Weijers and Vellinga 1995, Beran and Arnel 1995).
According to Dahmand Molles, (1992), small changes in precipitation produce large increases in flow variability because runoff response to precipitation is nonlinear. Fisher and Grimm, (1991), states that the contribution of each controlling process to the system’s state depends on the temporal pattern of flooding rather than on the total annual discharge. In years with flood evenly distributed in time, drought effects are rare and biotic interactions become more important and drought conditions characterized a substantial portion of the year. Changes in long-term processes are also important. Catchments derived fluxes of nitrogen to desert streams are higher in floods that occur after several years of low precipitation (Grimm 1992), presumably owing to accumulation of nitrates in soils during drought. However, reduction in stream flow is very likely to reduce the productivity of large flood plain rivers and low gradient streams dependent on periodic flowing. Inundation of flood plains provides expanded food-rich habitat and sources of organisms and organic matter for river ecosystems (Welcomme 1979, Junk et al 1989; Meyer 1990). Fish yields