Dispositifs permanents de nouvelle génération pour le suivi de la dynamique forestière en Afrique centrale : bilan en République du Congo

Managing tropical production forests sustainably requires knowledge of their diversity and dynamics, and of logging impacts on their characteristics. The necessary knowledge can be obtained by establishing large permanent sampling plots within which trees can be regularly measured. Such a monitoring experimental site exists in central Africa. It was settled in 1982 at M'BaIki, Central African Republic, and has provided reference standards to develop forest management and planning rules for many logging concessions in the region. Several projects have since established new monitoring experimental sites, focusing mainly on areas with different geological substrates. Two of these sites are located in managed logging concessions in the north of the Republic of Congo, on Cenozoic alluvia (Loundoungou) and on Mesozoic Carnot sandstone (Mokabi). Each site comprises two 400 ha blocks, each containing a trail of about 4,000 trees belonging to twenty commercial species and two 9 ha plots in which all trees more than 10 cm in diameter are measured. One block is logged after a few years of monitoring, with the second remaining as a control. Establishing a monitoring experimental site and making the first inventory required, on average, 34 months of work with a team of 8 people, at an average total cost of 94,500 (sic). The characteristics of the stands in the Loundoungou and Mokabi plots were, respectively: 240 and 230 tree species; densities of 346.8 (+/- 5.8) and 426.8 (+/- 3.6) trees/ha; biomass 432.6 (t. 13.8) and 457.3 (+/- 9.3) Mg/ha. This paper analyses the difficulties encountered in establishing the sites (position, location of the plots, setting up the trails, applying the inventory protocol, logistics, etc.) and puts forward several recommendations to improve both the establishment of such monitoring experimental sites and the inventory protocol.