What it will take to monitor forest elephant populations

The decision by the Convention on International Trade in Endangered Species (CITES) to condition a limited sale of ivory on the status of elephant poa ching has brought into sharp focus the technical, organizational, and finan cial challenges inherent in setting up a continent-wide monitoring program. The task will be particularly arduous in central Africa, where forest elep hant populations are difficult to monitor and the infrastructure necessary for population monitoring is lacking. The magnitude of effort that will be necessary is illustrated by an elephant survey done in the Gamba Complex, a network of protected areas on the southwest coast of Gabon. Implementing t he survey required a large-scale program for training Gabonese nationals in field survey and analysis methods necessary for elephant monitoring. Field work carried out during the training program suggests that a combination o f reconnaissance and line transects can product statistically valid populat ion estimates for a substantially lower effort than line transects alone. W e illustrate statistical frameworks for comparing the efficiency of alterna tive sampling methods, and we analyze the sensitivity of current survey met hods, which can detect abundance changes on the order of 15% given fairly h igh effort levels. Although these training and methodological results are e ncouraging, they are just small steps in a complex and ongoing process. We argue that this process has many parallels with what needs to be done if th e international conservation community is to respond to the challenge set f or it by the 1997 CITES meeting. There must be a concerted effort focused b oth on improving survey methods and on developing the human resources and o n-the-ground infrastructure necessary to implement these methods.