The Ganges, Brahmaputra and Meghna (GBM) river basins occupy about 1.75 x 10 6 km 2 of the Himalayan region. More than half a billion people in Nepal, India, Bhutan and Bangladesh are directly or indirectly dependent on the water resources of the GBM rivers. These river basins are characterized by diversified climatic patterns. Analyses of trends and persistence in precipitation over these river basins are necessary for sound water resources planning. Time series of annual precipitation for each of the 16 meteorological subdivisions covering the three river basins were examined for trends using the Mann-Kendall rank statistic, Student's f-test and regression analysis, and for persistence using first order autocorrelation analysis. Results indicate that precipitation in the Ganges basin is by-and-large stable. Precipitation in one subdivision in the Brahmaputra basin shows a decreasing trend and another shows an increasing trend. One of the three subdivisions in the Meghna basin shows a decreasing trend while another shows an increasing trend. Markovian persistence is not present in the precipitation series in the Ganges basin but it is present in two common subdivisions in the Brahmaputra and Meghna basins. Tendances et persistance des précipitations des bassins des fleuves Gange, Brahmapoutre et MeghnaRésumé Les bassins des fleuves Gange, Brahmapoutre et Meghna (GBM) occupent une surface d'à peu près 1.75 x 10 6 de km 2 dans la région Himalayenne. Plus d'un demi milliard de personnes au Népal, en Inde, au Bhoutan et au Bangladesh dépendent directement ou indirectement des ressources d'eau des fleuves GBM. Les bassins de ces fleuves sont caractérisés par des contextes climatiques variés. Des analyses de tendances et de persistance des précipitations de ces bassins se sont révélées nécessaires en vue de réaliser une planification efficace des ressources en eau. Nous avons étudié les séries chronologiques des précipitations annuelles de chacune des seize sous-divisions météorologiques couvrant les trois bassins fluviaux en utilisant la statistique de Mann-Kendall, le test t de Student et l'analyse de régression ainsi que l'auto-corrélation du premier ordre pour les problèmes de persistance. Les résultats indiquent que les précipitations du bassin du Gange sont relativement stables. Les précipitations de l'une des sous-divisions du bassin du Brahmapoutre présentent une tendance décroissante alors que celles d'une autre sous-division présentent une tendance croissante. Il en est de même de deux des trois sous-divisions du bassin du Meghna. La série des précipitations du bassin du Gange ne montre aucune persistance Markovienne que l'on peut au contraire mettre en évidence sur deux sous-divisions communes aux bassins du Brahmapoutre et du Meghna.

No abstract

No abstract

Bangladesh is a very flat delta built up by the Ganges-Brahmaputra-Meghna/Barak river systems. Because of its geographical location, floods cause huge destruction of lives and properties almost every year. Water control programs have been undertaken to enhance development through mitigating the threat of disasters. This structural approach to flood hazard has severely affected floodplain fisheries that supply the major share of protein to rural Bangladesh, as exemplified by the Chandpur Irrigation Project. Although the regulated environment of the Chandpur project has become favorable for closed-water cultured fish farming, the natural open-water fishery loss has been substantial. Results from research show that fish yields were better under preproject conditions. Under project conditions per capita fish consumption has dropped significantly, and the price of fish has risen beyond the means of the poor people, so that fish protein in the diet of poor people is gradually declining. Bangladesh is planning to expand water control facilities to the remaining flood-prone areas in the next 15-20 years. This will cause further loss of floodplain fisheries. If prices for closed-water fish remain beyond the buying power of the poor, alternative sources of cheap protein will be required.

No abstract