Mekong River Commission

Conventionally, floods and flooding are perceived as geophysical hazards within a common framework of natural disasters that also covers storms and hurricanes, earthquakes, volcanic eruptions, landslides and tsunami. In each case, socioeconomic losses and damage increase exponentially with event magnitude and as a function of civil exposure and vulnerability. Such hazards are perceived as random events that have entirely negative impacts, ignoring the fact that floods also have a positive ecological and socioeconomic function. Great civilizations have developed within flood plains where, on the face of it, exposure and vulnerability have been high. Such societies have included Sumerian Mesopotamia along the Lower Tigris and Euphrates Rivers, ancient Egypt along the Nile, the Harappan culture of the Indus valley in Pakisan and India, the founding cultures of China in the Yellow River Valley and the Angkor civilization of the Lower Mekong Basin itself. Exploiting the benefits and avoiding the risks brought by the annual flood stimulated such societies to put greater efforts into social organization and water management systems, which endorsed them as landmark civilizations.

The use of the term ‘flood’ in large tropical rivers and others such as the Nile refers to a highly predictable seasonal hydrological response to rainfall. The regime comprises two basic seasons, the 'flood' season and the 'low flow' season. Issues of scale mean that ‘flood’ conditions, when flows are considerably above the average for the year as a whole, last for several months. On much smaller rivers a 'flood' lasts only for several days, a duration that is more consistent with definitions of a 'flood' as a short duration geophysical hazard. The annual Mekong flood, in response to the SW Monsoon, does in some years reach peak discharges which cause considerable losses and damage. In other years the 'flood' is benign and benefits accrue. Here the term 'flood' is used to refer to the several months of increased discharge conditions, which may or may not in any given year result in large scale economic loss. The annual Mekong flood has been pivotal to the cultural and socioeconomic evolution of the region and despite the fact that some loss and damage occurs in most years, those that are significant at the regional and national scales have a frequency that is low enough to ensure that over time flood benefits exceed the flood costs by a very considerable margin.

Flood damage arises from a combination of direct losses due to the fact of inundation and secondary losses as a result of the suspension of normal economic activities in the commercial and service sectors which can accumulate long after the end of the event itself and until such time as damage is repaired and stocks and inventory replaced. Assessing these figures in dollar terms reasonably accurately requires detailed surveys of pilot areas the results of which are then applied to the flood affected region on a loss per unit area basis. This is the methodology adopted in each country in the basin and from data available from the relevant National Disaster Management Agencies the losses that are estimated to arise in an average year amount to a regional total of US$ 76 million. The most destructive regional flood conditions of recent decades occurred in 2000 in the south of the basin and in 2008 in the northern parts. By far the larger overall damages occurred in 2000 and amounted to US$811 million, those of 2008 being much less at US$135 million.

The major economic flood benefits arise in the agricultural and fisheries sectors, though by maintaining wetlands the annual flood also ensures that the products and services that such areas provide are maintained from year to year, though these are difficult to value in financial terms. On the basis of the value of production the main agricultural region in the basin is the delta in Viet Nam which contributes 30% of national GDP and is the major source of export rice, Viet Nam being the second largest contributor to the international rice trade.

In Cambodia flood inundated areas of the Delta are also amongst the most fertile regions for agricultural production, with extensive double cropping. Rice occupies 90% of the total cultivated land of which 32% lies in flood prone areas. Because of the fact that only 7% of the land is irrigated it is only in these naturally flooded areas that a second crop is possible on any large scale using receding water around lakes and rivers. In Lao PDR and Thailand the link between agriculture, rice cultivation in particular, and the benefits of the annual flood on the Mekong is less evident since most crops are rain fed or irrigated.

The link between the regional fisheries production and the annual flood is now well established. Current estimates regarding the annual value of the regional aquatic resources exceed US$2.8 billion, though not all of this figure is directly attributable to flood induced processes.

Other flood benefits accrue but they are much more difficult to value in financial terms. This applies in particular to wetlands which provide 'goods', such as fish and aquatic products, and 'services' such as flood attenuation and the retention, recovery and removal of excess nutrients and pollutants.

The theme of the AFR 2008 is not to set out a detailed economic analysis of the benefits and costs of the annual Mekong flood but to compare the two in financial terms based upon the generally available macro-data. These data, such as those with respect to the annual value of the Mekong fishery, are widely accepted not least by the MRC’s Fisheries Programme itself. In addition, the annual value figure for the fishery of US$2.85 billion refers only to that at the first point of sale. In comparison, the estimated average annual costs of flood damage only amount to US$76 million or just 2.5% of the fisheries benefit alone.

The additional figures for the agricultural benefits are harder to assess since the linkage between the annual flood is not as clear as it is with the fishery. Flood recession rice production in Cambodia is reported to account for 32% of national production, valued at US$3.1 billion in 2006 (based a February 2009 international price of US$500/tonnes). The value of the agricultural benefit accruing directly from the flood in Cambodia is therefore US$3.1 billion x 0.32 = US$1 billion.

In the Mekong Delta in Viet Nam agricultural production in 2004 had a reported market value of US$3.5 billion. That this is a benefit of the flood is based on the argument that the9 to 13 millions tonnes of sediment deposited annually across the delta, most of it during the flood season, has over millennia resulted in some of the most productive agricultural land in Southeast Asia, thus establishing the link.

Separating the proportion of the total benefits in any of the sectors that accrue directly from the annual flood is simply not possible, but even if they are half of the figures quoted here with regard the value of the annual Mekong fishery alone, the ‘benefits’ still far outweigh the ‘costs’. The major conclusion to be drawn is that in large tropical monsoonal rivers the annual flood cannot be perceived exclusively in terms of being a geophysical hazard. The comparative benefit and cost figures presented here should not be seen in any way as a direct economic comparison but interpreted as broadly indicative measures of the potential socioeconomic value of flood induced processes.

The 2008 flood season across the Lower Mekong Basin illustrated a common feature of the regional flood hydrology, that is extreme flooding in one part of the basin and average to below average conditions elsewhere. In 2008 it was the northern parts between the Chinese border and Vientiane that witnessed flood levels not seen for almost 50 years, while further downstream in Cambodia and Viet Nam discharges and water levels were average at best.

Volumes of flood runoff, discharges and water levels in the northern part of the region up until mid July had been average but an intensification of the SW Monsoon saw repeated tropical storms and higher flood runoff. The storm situation then intensified dramatically during the first week of August with the passage of tropical storm Kammuri across the northern provinces of Lao PDR. Only satellite indications of the associated 3 to 5 day rainfalls are available (the rain gauge network here being very sparse) but cumulative rainfalls considerably in excess of 500 mm are indicted. Volumes of flood runoff in the large northern Lao tributaries such as the Nam Ou and Nam Khan increased to extreme levels.

Maximum discharges at Luang Prabang and Vientiane/Nong Khai occurred on the 14th and 15th of the month and were in the region of 23,000 cumecs or 50% above the annual average. The associated water levels were comparable to those of 1966, when both city centres were flooded. Some urban flooding again occurred at Luang Prabang since water levels were elevated due to backwater effects on the Nam Khan upstream of the Mekong confluence. At Vientiane large scale urban flooding was avoided principally due to the sandbagging of the river bank adjacent to the central business district, but large areas of suburbs and peri-urban area were flooded to depths locally exceeding 1.5 m.

These conditions soon dissipated downstream of Vientiane/Nong Khai since flood flows into the mainstream from the large left-bank tributaries in Lao PDR, starting with the Nam Ngum, were quite average. This becomes quite clear from the flows at Pakse and Kratie which come close to those for an average flood season overall, while water levels in the Delta at Tan Chau and Chau Doc were below average throughout most of the season.

Apart from these events on the mainstream local, and often seriously damaging, flash floods occurred as they do in most years. In Lao PDR such conditions occurred in Bolikhamxay province during June with over 800 mm in 10 days and as much a 200 mm in 1 day. In the Thai Mekong region during September in the Khon Kaen area heavy rainfall of almost 160 mm. on the 17th caused of flood inundation and damage to property and crops and the deaths of two people. In Cambodia no significant flood damage was reported for the year while in Viet Nam the only events to cause any significant damage occurred in the Upper Sre Prok during May and in the upper Se San Basin during the first week of August. The former saw two lives lost and the latter resulted in some bridges being damaged and about 100 ha of rice paddy lost or seriously damaged. Overall damage from flash floods for the year in Viet Nam amounted to US$1 million. Regionally, however, the number and severity of flash flood episodes was well below average.

The events of August 2008 provided the first real opportunity to assess the performance of the RFMMC’s flood forecasting models and expertise and in general they performed reasonably well, though obviously room for improvement is recognised. The major constraint to accuracy over lead times in excess of two to three days is the emerging conclusion that the storm rainfalls indicated by satellite images are probably too low and that the network available for ‘ground truthing’ needs to be improved in key areas of Lao PDR in particular.

Otherwise the lessons and recommendations put forward in the four country reports generally repeat those of earlier years and focus up such issues as strengthening community based flood risk management and self reliance and building capacity within local communities with regard to flood preparedness and emergency response. A need is also recognized to improve the channels of communication between the local communities and the relevant flood forecasting, mitigation and response agencies. In Viet Nam once again attention is drawn to diverting more financial support for dealing with flash floods in the Central Highlands. Finally, a call is once more made to consider translating the Annual Flood Report into each of the four riparian languages.