At lower elevations in Eastern Austria, many coniferous forests are dominated by Austrian pine (Pinus nigra).
The utilization of this species – which has its most northern natural distribution in the neighbouring Northeastern
Alps – in plantations, was facilitated from the 16th century until the early 20th century for resin production. At
present, many of these managed forest stands are threatened by the mycosis Diplodia sapinea. In its beginning
stages, this mycosis causes needles to be discolored brown. First infection signs appear on shoots and twigs, later
spreading to whole branches and parts of the crown, causing discoloration and necrosis. The mycosis mainly
spreads in wet spring seasons, and is exacerbated by following hot dry summers, which increase the stress level of
the trees. To evaluate the stages of infection and monitor affected forests, we mapped a black pine stand near the
city Wiener Neustadt (AT) with unmanned aerial system photogrammetry (UAS-P) in autumn 2016. Two different
sensors were used: i) a Sony NEX5 to generate high-resolution RGB orthophotos (OP) and digital surface models
(DSM) [0.05 m and 0.2 m ground sampling distance (GSD), respectively]; ii) a multispectral camera (MicaSense
RedEdge) to calculate vegetation indices (0.1 m GSD). OP, DSM and indices were calculated using standard
structure-from-motion photogrammetry software (AgiSoft PhotoScan Pro and Pix4D). The crowns of the black
pines were manually delineated and assigned to three categories: healthy, infected and dead. An NDVI-threshold
was determined for the each category. The comparison between the visual classification (RGB-OP) and the NDVI
classification shows very good accordance for dead trees, sufficient for healthy and inconclusive for infected trees.
In further investigations the automatic detection (NDVI treshhold) is attempted to be improved to identify spreading patterns of the mycosis and support mitigation measures. Very detailed field surveys were undertaken to control the infection stage; in a next step these data will be compared with the UAS-P results.
