Population Survey

This report documents a national aerial survey of elephants in Zambia, conducted in Sep. 2015. A total of 84 859 km² was sampled at an average intensity of 7.99%. Four ecosystems were covered: Kafue (45 030 km²), Luangwa (32 819 km²), Sioma Ngwezi (4 482 km²) and Lower Zambezi Valley (2 528 km²). The survey was conducted using a Cessna 182 and 206. A total elephant population of 21 760 (95% c.i. 17 237 - 26 283) was estimated across 11 strata covering the four ecosystems. Populations within each ecosystem were estimated as follows: Kafue (4 strata) 6 688, Luangwa (4 strata) 13 898, Sioma Ngwezi (2 strata) 48, Lower Zambezi Valley (1 stratum) 1 125. The report notes that the country total of 21 760 is similar to the estimate for 2009. The only evidence for change was seen in Sioma Ngwezi National Park, where there has been a steep population decline since 2004. The survey area was stratified on the basis of previously reported elephant densities. Other datasets that were considered to be comparable for the purposes of population trend analysis were sourced for Lower Zambezi NP (2003, 2005, 2013), Luangwa System (2008, 2009, 2011), Kafue System (2006, 2008, 2011) and Sioma Ngwezi (2004, 2005, 2013). Elephant carcass ratios were 7.02% (Kafue), 1.2% (Luangwa), 6.3% (Lower Zambezi) and 85.5% (Sioma Ngwezi). The overall carcass ratio for the country was 4.52%. The report states that the number of elephant carcasses is higher than would be expected from natural mortality alone in all areas except the Luangwa valley, indicating recent or continuing illegal hunting. In the case of Sioma Ngwezi the carcass estimate provides strong supporting evidence of a poaching-driven decline. In Kafue and the Lower Zambezi valley the carcass numbers are still, just, within the limits where the mortality is sustainable While the report notes that there were some minor deviations from best practice (carcasses were missed; some elephant observations were not classified; 1 transect was missed; speeds were too high; heights were too variable; some marker rods were insufficiently rigid; no permanent voice recordings were made; no usable photographs (of elephants) were taken), it judges that none of these deviations materially affected the conclusions of the survey.