The aerial surveys are conducted using a small airplane with four crew members: a pilot, two observers, and one recorder. The observers note the number of elephants, group composition (family or males), GPS coordinates, elephant carcasses, other species, and human disturbances within the counting strips on either side of the plane. Groups of elephants are photographed with cameras mounted on the aircraft to be confirmed later.
Within each country, survey areas are divided into regions, called strata, and depending on terrain and vegetation, survey teams use one of three survey methods to obtain elephant counts in these strata:
- Transect sample counts are the preferred survey method, suitable in relatively flat terrain with sparse vegetation. The survey aircraft flies back and forth along parallel transect lines that are about 2.5 km apart and that begin and end at opposite sides of the stratum. Observers count elephants within 150-meter "counting strips" on either side of the plane (see figure). The counting strips are the areas that are being sampled within the stratum.
- Block sample counts are used in hilly terrain where it is difficult to fly straight-line transects across a large area or where vegetation is dense. Small areas, called blocks, are chosen randomly within a stratum. Within each block, the aircraft flies along transects that are sufficiently close to one another to ensure that the counting strips cover the entire block. The blocks are the areas that are being sampled within the stratum.
- Total counts are used where the sampling area and elephant population are small or elephant distribution is highly clumped. The aircraft flies along transects within the stratum that are close enough together to ensure that the counting strips cover the entire stratum. Because researchers survey the entire stratum, this is a total count rather than a sample count.
Strip Transect Sampling Method
In all survey methods, transects are oriented perpendicular to rivers or other natural features, to reduce the variance in elephant densities between transects.
The width of the counting strips must remain consistent, because the width is later used to calculate the proportion of the survey area that was sampled. To achieve consistency, the pilot flies the plane at a constant height of 91.4 meters using a laser altimeter. Visual guides on the wings mark the 150-meter-wide visual search area for the observers. The plane flies at 160 to 180 km/hr covering about 1 to 1.5 km2 per hour.
The map shows the survey area for the Zambezi Valley in Zimbabwe, with transect sample counts in strata with flat terrain and block sample counts in strata with hilly terrain. Turn on the flight path layer to see the actual path that the airplane flew during the survey. The total count method was not used in this survey area.