An important first step leading to the better management of forests is that of obtaining better information about their timber resources. This article considers how such information can be derived from aerial photography. Review of the various approaches to estimating stem diameters and volumes from air-photo measurements and the results of ground studies with Pinus radiata have illustrated the feasibility of forest-sampling with macro-scale photographs. However, their use in this role is by no means resolved, although the major problems have been defined. A number of these relate to image qualities, but one of the crucial problems for operational use was seen to be the determination of photographic scale. All of these factors were considered in trials in plantations of radiata pine at Longford in southern Victoria, where the aim was to obtain macro-scale photographs suitable for detailed measurement. Methods used in these trials and the photographic results which were obtained are described in this paper. The simplest and most practical way to obtain macro-scale photographs of sharp quality is to use cameras of small format and rapid shutter speed. For stereoscopic coverage with acceptable parallax, a further requirement is a rapid recycling rate for timed-interval photography or careful selection of flying height and air base for fixed-base photography. Specifications relating to image qualities other than sharpness also require careful consideration, as does the method for determining photographic scale. Sharp photographs were obtained from flying heights of 200-500 ft using Vinten 70-mm cameras at a shutter speed of 1/500 sec and aircrafts speed of 30 knots. However, in some cases, image sharpness decreased towards the edges; a factor attributed to a lens defect which was most apparent on colour photographs taken at large apertures. Fixed air-base photography was selected in preference to timed-interval photography because it gave a simple way to determine the height of flying without the need for ground control. In the initial trials, stereoscopic pairs of macro-scale photographs were taken simultaneously from heights of 200-500 ft using twin Vinten 70-mm cameras mounted on a 16-ft boom which was attached to a helicopter and oriented parallel to the direction of flight. With this approach, imprecise synchronisation of exposure with the two cameras caused inaccuracies in the first estimates of flying height from longitudinal photographs i.e., those on which the air base was parallel to the long axis of the helicopter. However, regression improved the accuracy of estimated values such that they were within ± 7% for a flying height of 200 ft and ±17% for a flying height of 500 ft (0.95 probability). Transverse orientation of the air base largely overcame the problems of camera synchronisation, but introduced a new problem, viz, bending of the boom. However, as this was a constant factor, estimates of flying height from regression were. © 1979.
