Based on a full-resolution Meteosat dataset, an extensive climatological study of the mesoscale convective systems (MCSs) observed by satellite over the Sahel leads to the definition of a subpopulation of MCSs-called organized convective systems (OCSs)-that represents only 12% of the total number of MCSs observed during 9 yr over the central Sahel while accounting for almost 80% of the total convective cloud cover defined at the 233-K threshold. Using a high-resolution rainfall dataset, it is shown that these OCSs are also the main source of rain in this region, accounting for about 90% of the seasonal rainfall, with a mean areal rainfall of 14.7 mm per system. All of the OCSs are associated with a rain event, and more than 90% of the major rain events are associated with an OCS. These figures are compared with those obtained for mesoscale convective complexes (MCCs). Each MCC produces more rainfall on average (19 mm per system) but there are only a few of them (1.2% of the total number of MCSs), and they consequently produce only 19% of the seasonal rainfall. The interannual rainfall variability is first determined by the year-to-year fluctuation of the number of events defined from satellite rather than by the fluctuations of their mean rain efficiency. In fact, the total rain yield of an OCS appears to be linked primarily to its duration (which itself is largely determined by its spatial extension) rather than to its average rain rate. The diurnal cycle over the region is also studied, and it is shown that it is largely conditioned by the propagative nature of the OCSs associated with orography-driven generations located a few hundred kilometers to the east of the validation area.

This paper provides an eight‐year high‐resolution climatology of Sahelian mesoscale convective systems (MCSs) during the summer. MCSs are defined as convective cloud clusters larger than 5000 km2. They are extensively tracked from METEOSAT full‐resolution infrared images (time resolution 0.5 h and spatial resolution about 5 km). The method enables every MCS to be tracked throughout its entire lifetime. For each time step, the MCS location and its morphological and radiative characteristics are computed for three different brightness temperature thresholds. The methodology is presented, evaluated and compared with previous studies using low‐resolution data. Statistical MCS distributions, diurnal cycle and spatial variability of MCS characteristics are analysed on the basis of this high‐resolution tracking. It is shown that a few large and long‐lived cloud clusters contribute most of the total cloud cover. Sahelian cloud clusters propagate westward at a greater speed when very deep convection is well developed. The diurnal organization of the convection has been analysed, and has proved that the merging of MCSs is partly explained by the actual merging of independent convective entities, whereas the splitting of MCSs is mostly associated with weakening of convection. The importance of mesoscale convective complexes for the total MCS coverage has also been studied.

SUMMARYThis paper provides an eight-year high-resolution climatology of Sahelian mesoscale convective systems (MCSs) during the summer, MCSs are defined as convective cloud clusters larger than 5000 km'. They are extensively tracked from METEOSAT full-resolution infrared images (time resolution 0.5 h and spatial resolution about 5 km). The method enables every MCS to be tracked throughout its entire lifetime. For each time step, the MCS location and its morphological and radiative Characteristics are computed for three different brightness temperature thresholds. The methodology is presented, evaluated and compared with previous studies using lowresolution data.Statistical MCS distributions, diurnal cycle and spatial variability of MCS characteristics are analysed on the basis of this high-resolution tracking. It is shown that a few large and long-lived cloud clusters contribute most of the total cloud cover. Sahelian cloud clusters propagate westward at a greater speed when very deep convection is well developed. The diurnal organization of the convection has been analysed, and has proved that the merging of MCSs is partly explained by the actual merging of independent convective entities, whereas the splitting of MCSs is mostly associated with weakening of convection. The importance of mesoscale convective complexes for the total MCS coverage has also been studied.

We consider nonlinear programs in 0–1 variables with nonlinear constraints and survey the main approaches to their solution: (i) linearization; (ii) algebraic methods; (iii) enumerative methods and (iv) cutting-plane methods. We also present an extensive computational comparison of algorithms of all four categories. Enumerative methods appear to be the most promising. INFORMS Journal on Computing, ISSN 1091-9856, was published as ORSA Journal on Computing from 1989 to 1995 under ISSN 0899-1499.

[1] Relationship between Sahelian mesoscale convective systems and easterly waves is studied at various time and space scales during 5 rainy seasons (1st July -15th September 1990 -1994. This study focuses on a sub-population of MCSs called organised convective systems (OCS) which account for most of the rainfall. Wave activity periods are detected by spectral analyses on the 700 hPa meridional wind component, information on the convective systems is derived from an automatic tracking algorithm and rainfall data of the EPSATNiger experiment. We observe a modulation of OCS cloud cover at synoptic-scale during easterly wave activity, with an increase of the cloud cover in and ahead of the trough but OCS rainfall efficiency is maximum behind of the wave trough. At seasonal scale, OCS number and cloud cover distributions are not significantly affected by easterly wave occurrences, which means that the latter are not directly associated to the interannual variability of rainfall.INDEX TERMS: 3314