Predicting Sphaeropsis sapinea Damage in Pinus radiata Canopies Using Spectral Indices and Spectral Mixture Analysis

Maintaining the health and condition of the forest plantation estate is critical to ensuring there are no adverse losses in productivity. Within Australian Pinus radiata plantations a diverse range of damaging agents are present. One significant agent is a fungal pathogen Sphaeropsis sapinea. In this research, we detail the development of relationships between a range of individual crown health attributes representing symptoms of Sphaeropsis sapinea infection and high spatial and spectral resolution remotely sensed imagery characteristics. To do this, two methods were used; the first utilized vegetation spectral indices including simple and normalized difference ratios, and the second, linear spectral mixture analysis. Results indicate that spectral indices that utilize either chlorophyll absorption wavelengths at 680 nm with a non-chlorophyll region of the spectrum (such as 710 or 750 nm) or the slope of the upper red-edge between 710 and 740 nm were most significantly related to individual crown damage attributes. Linear unmixing analysis consistently extracted four fraction endmember images (sunlit canopy, soil, shadow, and non-photosynthetic vegetation (NPV)) from the 12 channel imagery. Multiple linear stepwise regression models developed using mixed fractional abundances provided similar results to those derived using spectral indices. The NPV and shadow endmembers, in order, were consistently identified as the most significant in these developed models. Introduction Australian softwood, Pinus radiata (D. Don), plantations contain a number of abiotic and biotic damaging agents that affect the health of the forest (Will, 1985; Bollmann et al., 1986; Lewis and Ferguson, 1993). Sphaeropsis sapinea (formerly known as Diplodia pinea) is one of the most severe and ubiquitous. Under weather conditions favorable for infection (wet and warm weather), the fungus penetrates Predicting Sphaeropsis sapinea Damage in Pinus radiata Canopies Using Spectral Indices and Spectral Mixture Analysis Nicholas C. Coops, Nicholas Goodwin, and Christine Stone succulent stems through the intact epidermis and enters needles through stomata. The fungus may also infect stems and older branches through fresh wounds inflicted by insects, hail, or damaging agents. This capacity to infect a range of host tissue types can result in a number of different symptoms including needle discoloration, shoot blight or dieback, stem cankers, root diseases, and blue stain. Many of these symptoms become apparent after hail damage or when the tree is stressed due to environmental factors such as drought. While tree death rarely occurs, severe malformation resulting from repeated attacks has been shown to have a major effect on potential stand volume. Improvements in remote sensing technologies, particularly in the spatial and spectral resolution of optical sensors, have made the prospect of using digital remotely sensed imagery to detect and classify stressed vegetation a realistic and attractive option (Franklin, 2000). A number of different approaches have been used to derive forest health and condition from remote sensing imagery. These include the use of narrow band spectral indices which are sensitive to leaf pigment content (Datt, 1998; Zarco-Tejada et al., 2002) and crown biomass (Spanner et al., 1990a and 1990b; Coops et al., 2002) and the use of image fractions derived from endmembers which represent the spectral characteristics of cover types regarded as having uniform properties (GarciaHaro et al., 1999). The use of narrow-band spectral indices is the most widely applied, however, a potential limitation with the use of spectral indices is that they are often calculated from a small number of spectral bands, usually two, and thus do not utilize new and potentially important information in other channels (Peddle et al., 2001). In addition to using spectral image information to assess canopy condition, research has also demonstrated that structural change occurs at both an individual crown or forest canopy scale (e.g., Wulder et al., 2004). High spatial resolution imagery, with pixels smaller than the dimensions of individual tree crowns, allows the derivation of variance measures and spatial statistics which have been related to the physical structure of individual trees. For example, the relationship between shadow in a patchy canopy compared to a bright full and dense canopy (e.g., Gougeon, 1999; Lévesque and King, 1999; Olthof and King, 2000) has been modeled using high spatial resolution imagery. PHOTOGRAMMETRIC ENGINEER ING & REMOTE SENS ING Ap r i l 2006 405 N.C. Coops was formerly at CSIRO Forestry and Forest Products, Private Bag 10, Clayton South 3169, Victoria, Australia, and is now at the Department of Forest Resource Management, 2424 Main Mall, University of British Columbia, Vancouver, Canada ([email protected]). Nicholas Goodwin is at CSIRO Forestry and Forest Products, Private Bag 10, Clayton South 3169, Victoria, Australia, University of New South Wales. Christine Stone is with the Research and Development Division, State Forests on NSW, P.O. Box 100, Beecroft, NSW 2119. Photogrammetric Engineering & Remote Sensing Vol. 72, No. 4, April 2006, pp. 405–416. 0099-1112/06/7204–0405/$3.00/0 © 2006 American Society for Photogrammetry and Remote Sensing 04-067 3/14/06 8:57 PM Page 405