The water level in Mopti rose 1 cm last week, from 627 cm on 21 September to 628 cm on 22 September, remained three days at that level and declined 5 cm in the next three days to 623 cm on 28 September. The prediction of OPIDIN was that the water level would still further increase some 40 cm to reach its peak in the second half of October. It is still possible that the declining water level last week was temporary and that the water level will increase again the next weeks. That occurs very rare, however, and was only observed in 2005 (when the water level in Mopti declined 6 cm between 26 September and 3 October to increase again 23 cm till 24 October) and in 2010 (when the water level in Mopti declined 5 cm between 19 and 25 October to increase again 8 cm till 6 November).
Assuming that the water level has been at its maximum on 22 September, the flood level will this year not be as high as last year and in 2016 and 2018 (as predicted by OPIDIN), but as high as in 2013, 2014 and 2016. If that were the case, OPIDIN has overestimated the peak flood last month by some 40 cm. This would be the first time that the OPIDIN prediction deviates so much from the actual peak.
The daily water level in Mopti has been measured since 1923. The water level reaches normally its peak in October or November and, on average, for all these years, on 24 October. In only two years the water level was already at its peak in September: in 1982 on 29-9 and in 2017 on 22-9. In both years, the flood level was rather low (551 cm). That fits in the normal pattern. When the flood level in Mopti is low (440 – 600 cm), the peak is usually reached early October. When the flood level is higher, the flood reached its peak later in the season and at a very high flood (> 700 cm in Mopti), the peak in Mopti is reached in November or even early December. Why the water level has reached its peak in Mopti one month earlier than expected this year is still the question. The rainfall has been above average in the Upper Niger Basin and did not stop earlier. Thus this is not the explanation. A further analysis of the daily water level of all station upstream of Mopti would be needed to understand why the water has started to decline already so early this year in Mopti.