How likely is flooding now?
This is the second part of my article entitled “How likely is flooding this year?”, which was published in the issue of 'The Daily Star' on August 07 of 2010. While the first part was prepared based on the on-going La Niña event at that time, the second part is prepared on the basis of further strengthening of that on-going La Niña. Although, the first part was published in August, I actually prepared it in the month of May 2010. It was originally intended to be published earlier. But it was not done when some other water experts also commented that the possibility of a flood in 2010 (at least at a moderate level) was nil. Many people believe that advance forecasts with a lead time of about 3-6 months are not doable. Others feel that all these forecasting techniques are in the experimental stage and it is extremely difficult to get reasonable skill for designing a real-time response plan based on these forecasts. Based on the concerns of these probabilistic forecasts for flood hazards mitigation in Bangladesh, my response is clear here that the beauty of these forecasts lies in making skilful products for at least 3-6 months (1-2 seasons) in advance. This El Niño/La Niña-based forecasting method can produce forecasts in a hot/dry spring about a rainy summer. It can also give an idea about the onset of flooding (early or late). These products are successfully used widely in many developing and developed countries in the world. I mentioned in my first article that some readers' might be surprised to read any news on flooding when the country was extremely dry with a severe shortage of rainfall. I am now confident that they are now ready to accept the reality of a flood, which has knocked at the door. So, the El Niño/La Niña-based forecasts, which produce flood forecasts for 3-6 months in advance, are achievable in Bangladesh. Therefore, the objective of this article (Part II) is to emphasise the role of seasonal forecasts for hazards management in Bangladesh. I am just trying to recall that, based on El Niño/La Niña climate cycle, I produced some forecasts in April-May of 2010 (when the country was dry) on a forthcoming floods in late September-October. Currently Bangladesh is flooded and the situation was deteriorating every day. These findings clearly demonstrate that the seasonal forecasts prepared earlier at the beginning of the season (3-6 months before) are fully instrumental. This also summarises that we can produce flood forecast for about 3-6 months in advance with a reasonable accuracy. However, there is a limitation here while the conventional short-range deterministic forecasts are mostly accurate as these are on shorter time scales, the probabilistic seasonal forecasts have some uncertainties as these are on longer time scales. So, when you read a probabilistic flood forecasts in three tercile format like 40 (upper):40(middle):20(lower), it means the possibility (or probability) of higher-than-normal flooding is 40-percentage (upper tercile), the possibility for normal flooding is 40-percentage (middle tercile), and the possibility for lower-than-normal flooding is 20-percentage (lower tercile). So, even we forecast 40-percentage possibility for a higher-than-normal flooding, there is still a 20-percentage possibility for a lower-than-normal flooding. As compared to deterministic forecasts, this is the kind of limitation probabilistic forecasts have. On the contrary, while the deterministic method can produce forecasts only during the occurrence of an event [for example, the 'Flood Forecasting and Warning Center' of the Bangladesh Water Development Board (BWDB) produces information on daily rise/fall of river water-level in a flooding season], the probabilistic method can produce forecasts in a hot/dry spring about a rainy summer and monsoon flooding (3-6 months in advance). It is very important to note here that once the seasonal forecasts are prepared, these forecasts need to be monitored based on oceanic and atmospheric anomalies. So, now the question is how does the flooding picture look now for the season September-October-November (SON) of 2010? Has anything changed since August of 2010? The answer is yes; according to the Climate Prediction Center (CPC) (USA), La Niña strengthened during August 2010, as negative sea surface temperature (SST) anomalies reached at least -1°C across most of the equatorial Pacific Ocean by the end of the month. As La Niña has further strengthened, so current forecast for flooding for September-October-November (SON) of 2010 can slightly be changed from August-September-October (ASO) of 2010. Please note that, on three tercile format, the ASO-forecast was presented as 40 (upper): 40 (middle): 20 (lower). Now with the strengthening of La Niña, the SON-forecast can be presented as 50 (upper): 40 (middle): 10 (lower), which means that there is 50-percentage probability for higher-than-normal flooding, 40-percent probability for normal flooding, and 10-percentage probability for lower-than-normal flooding. Finally, as these types of seasonal products are used widely and successfully in one-quarter of the globe, Bangladesh can benefit by utilising these seasonal forecasts for flood hazards management.
For all latest news, follow The Daily Star's Google News channel.
Comments