Mapping mountain pine beetle infestation with high spatial resolution satellite imagery

Joanne C White, Michael A Wulder, Darin Brooks, Richard Reich, Roger D Wheate
2004 Forestry Chronicle  
The on-going mountain pine beetle outbreak in British Columbia has reached historic proportions. From 2002 to 2003, the area infested with mountain pine beetle doubled, increasing from approximately 2.0 million hectares to 4.2 million hectares. The extent of the current mountain pine beetle outbreak, the rapid rate of the infestation's spread, and the associated economic impacts, have prompted research into new techniques and data sources for reconnaissance and mapping of the infestation.
more » ... infestation. Detection and mapping of attacked trees serves to delineate and document the current impact of the infestation, enabling planning and mitigation activities. In addition, knowledge of current impact serves as a driver in parameterizing models of beetle spread that are designed to reduce future risks and impacts. Recently, management efforts at the local level have shifted from exhaustive mapping of the infestation, to the detection and mitigation of sites with minimal levels of infestation. The motivation behind this shift in management focus has been to identify new pockets of infestation, and thereby reduce or contain the outbreak to a size and distribution that may be handled within the capacity of the existing forest industry infrastructure. Trees in the red-attack stage of infestation have a distinctive red colour, which facilitates their detection with remote sensing instruments (Fig. 1) . Although the timing of the different stages of infestation is somewhat variable amongst individual trees, the general progression of foliage discoloration is predictable. Under typical conditions, adult mountain pine beetles attack trees in mid-summer, and lay eggs that develop into mature adults approximately one year later (the beetles must attack in large numbers to overcome the defences of a healthy tree and this is referred to as mass-attack). Once killed, but still with green foliage, the host tree is in the green-attack stage. The foliage of the host tree changes gradually and twelve-months after being attacked, over 90% of the killed-trees will have red needles (red-attack). Three years after being attacked, most trees will have lost all needles (grey-attack). The on-going mountain pine beetle outbreak in British Columbia has reached historic proportions. Recently, management efforts at the local level shifted from exhaustive mapping of the infestation, to detection and mitigation of sites with minimal levels of infestation, creating an operational need for efficient and cost-effective methods to identify red-attack trees in these areas. High spatial resolution remotely sensed imagery has the potential to satisfy this information need. This paper presents the unsupervised classification of 4 metre IKONOS multispectral imagery, for the detection of mountain pine beetle red-attack, at sites with minimal infestation (< 20% of trees infested). A 4-metre buffer (analogous to a single IKONOS pixel) was applied to the red-attack trees identified on the IKONOS imagery in order to account for positional errors. When compared to the independent validation data collected from the aerial photography, it was found that 70.1% (lightly infested sites) and 92.5% (moderately infested sites) of the red-attack trees existing on the ground were correctly identified through the classification of the remotely sensed IKONOS imagery. These results demonstrate the operational potential of using an unsupervised classification of IKONOS imagery to detect and map mountain pine beetle red-attack at sites with minimal levels of infestation. Keywords: mountain pine beetle, remote sensing, accuracy assessment, IKONOS, red-attack L'épidémie de dendroctone du pin actuellement en cours en Colombie-Britannique a atteint des proportions historiques. Dernièrement, les efforts d'aménagement au niveau local se sont dirigés de la cartographie exhaustive de l'épidémie vers la détection et la mitigation sur les sites connaissant des niveaux minimes d'épidémie, créant ainsi un besoin au niveau des opérations de méthodes rentables d'identification des arbres très atteints dans ces sites. L'imagerie de télédétection à haute résolution spatiale pourrait répondre à ces besoins d'information. Le présent article fait état de la classification non supervisée de l'imagerie multispectrale de 4 mètres IKONOS, utilisée pour la détection d'attaque massive de dendroctone du pin sur des sites d'épidémie légère (< 20% des arbres infectés). Une zone tampon de 4 mètres (analogue à un pixel individuel d'IKONOS) a été appliquée aux arbres très atteints identifiés par l'imagerie IKONOS de façon à tenir compte des erreurs de position. Lorsque les données furent comparées aux données indépendantes de validation recueillies à partir de photographies aériennes, on a constaté que 70,1% (sur les sites d'épidémie légère) et 92,5% (sur les sites d'épidémie moyenne) des arbres très atteints retrouvés au sol avaient été correctement identifiés suite à la classification de l'imagerie de télédétection IKONOS. Ces résultats soulignent le potentiel opérationnel de l'utilisation de la classification non supervisée de l'imagerie IKONOS pour détecter et cartographier les pics d'infestation du dendroctone sur des sites connaissant des niveaux minimes d'épidémie. Mots-clés : dendroctone du pin, télédétection, évaluation de la précision, IKONOS, attaque maximale
doi:10.5558/tfc80743-6 fatcat:7mwxsq3xzva5ra5er6cypdodry