Abstract: Coniferous trees in North America face many threats, with both native and invasive insect damage a key component of their decline and mortality. As insects, particularly bark beetles, attack the trees, they undergo a color shift from green to red to gray. Attacks from other insects can result in defoliation, or the thinning and loss of needles. These changes can be detected by remote sensing instruments such as satellites and drones. Mortality may however come from multiple variables such as fire or drought, which causes stress to a tree, making it more susceptible to insect infestation. Three areas of interest were chosen, one each in Washington, Idaho, and Montana, to study these effects.
This study aims to examine the spectral and spatial patterns displayed across multiple types of trees, with the damage caused by a range of insects. Multiple levels of spatial resolution are analyzed and classified, using field measurements, drone data and high- and moderate - resolution satellite data. Field studies consisted of measuring variables such as diameter, health, and needle color on both trees near each other in plots as well as isolated trees. For drone and satellite data, the pixels from these data sets are classified using several modeling techniques, resulting in a map of different tree health classes and other land classes such as bare ground and herbaceous vegetation.
Current results are preliminary but include a classification of different tree and land cover types along with measurements of their accuracy. By comparing classification results at different resolution levels, it is possible to see what information is retained or lost at each step down in resolution. Field measurements provide corroborating evidence for modeling input and results. This can assist forest managers and natural resource scientists in analyzing forests, giving them guidelines for when to invest more time and resources. This research will also allow for general trends for areas with insect-specific mortality, allowing for possible future comparisons with other causes of tree mortality.
