Anna Maria Jönsson

Forests are important from many perspectives. Forestry delivers products such as timber, fiber and fuel; forests are also important for recreational activities as well as for the global carbon balance. Consequently, it is important to develop methods that enable efficient monitoring of disturbances, such as insect attacks, over vast forested areas. These methods can be based on remote sensing, since satellites provide images with frequent spatial coverage. Insect attacks can be detected in these satellite data if the resulting defoliation or discolouration is sufficiently severe. Satellite data also facilitates monitoring of migration patterns of invasive insects since some sensors provide time series that enables tracing of insect attacks back in time. In this study, SPOT and MODIS data were utilized to map damage in Norway spruce (Picea abies) caused by Physokermes inopinatus, and the associated black encrustation formed by sooty mold during an attack occurring 2010 in Scania, the southernmost province of Sweden. This attack is the first known presence of P. inopinatus in Sweden. The study shows that damage can be detected with high accuracy in satellite data. In SPOT-data, 78% of the damage was detected with an overestimation of 46%. The larger damaged areas could be detected with MODIS 16-days composite NDVI-product with 250 m resolution. In addition, the study indicates that there was damage already in 2009, the year before any damage was detected in field. Prior to 2009 no damage was detected, suggesting that this was the first year of the outbreak. This study documents the outbreak of P. inopinatus in S. Sweden and highlights the potential for remote sensing for monitoring and early detection of damage of this invasive insect. (C) 2012 Elsevier B.V. All rights reserved.