A river flood is a flow in excess of the channel capacity to accommodate the peak discharge, and occurs when the amount of water arriving on land exceeds the capacity of the land to discharge that water by infiltration, surface flow or drainage pipes. Flooding of river valleys and coastal areas is the most frequent of natural hazards and is one of the most significant in terms of death, injuries and long-term social and economic impacts. Flooding regularly claims over 20,000 lives a year and affects 75million people globally. Thus, strategies are put in place to minimize the impacts of these hazards. The 3 broad groups are prediction, mitigation and response. However, the effectiveness of these strategies vary depending on the level of development of the country, the level of cooperation between the authorities, population and private organizations and the severity of the hazard. It is also important to note that effective hazard management cannot rely on just a single strategy and generally, a multi-pronged approach, which is a combination of strategies, is necessary to minimize the severe impacts of the hazard. Prediction
1. Prediction through the calculation of flood recurrence interval a. North Dakota – Accuracy and reliability of data
b. Flood recurrence interval is defined as the probability that a flood of particular magnitude will occur once or more in any given year. It can be used to examine the flood frequency (how often an area will experience flood) and flood magnitude (the size of a flood event). The records of a river’s discharge over the longest time available are ranked according to the discharge volume. The formula (n+1)/R is used to calculate the recurrence interval, where n is the number of discharge levels in the record and R as the rank of discharge. The recurrence intervals then can be plotted against discharge to determine the statistical probability of flood events. c. Using the flood recurrence intervals, a hazard map can be produced, which will aid in mitigation and response strategies. d. However, the statistical prediction of flood recurrence intervals Is only an indication of probability based on past records, and it is known that rivers are constantly changing during to erosion and deposition, thus the statistical prediction will change. Also, the accuracy of the extrapolated values depends on how much data is available, and most data tend to be relatively recent and cover only a few decades. These factors lead to the inaccuracy of flood recurrence interval values. For example, North Dakota had two 250 year floods within 110 years. e. Thus, prediction through flood recurrence interval is only reliable if the flood records are longer and fuller, and if the records are constantly updated as events happen. 2. Forecasting
f. Bangladash [Rmb: NWS] – Level of development
g. Flood forecasting uses satellite readings and radar to predict the occurrence of floods. This method of prediction is more accurate than using the recurrence interval method to predict the occurrence of floods. Forecasting can be split into short-term and long-term forecasting. Short-term forecasting is based on the atmospheric circulation pattern monitored through satellites. It has a higher accuracy of around 24h to 72h lead time, and is useful for emergency actions. Long-term forecasting is based on rainfall run-off modeling for different scenarios using historical flow data, and the likelihood of the river flooding is determined based on the current rainfall conditions and weather observations. h. Flood forecasting is more common in DCs than LDCs, as DCs have the availability of financial resources and technical expertise. For example, in US, river gauges to track water levels, extensive radar networks are used to determine rainfall volume and location, and computer models are sued to predict how the water will flow downstream, just to predict the floods....