Landsat-based detection of mast events in white spruce (Picea glauca) forests

Mast seeding in conifers is characterized by the spatially synchronous and temporally variable production of seed cone crops. Large mast events (known as “mast years”) can be a visually stunning ecologically important phenomenon, supporting trophic interactions survival predators well forest regeneration. Documenting patterns generally labor-intensive, requiring repeated visual counts at consistent widespread locations over long periods to quantify spatiotemporal variability production. Our goal this work was evaluate correspondence multispectral vegetation indexes (VIs) from Landsat with ground-based observations white spruce (Picea glauca) forests Kluane region, Yukon, Canada. Given characteristics spruce, we tested: 1) whether photosynthesis- color-oriented VIs identify senescence cones late summer autumn during years, 2) if moisture-oriented distinguish significant drying surrounding canopy that after seeds are released. We hypothesized slope season decline would related masting (i.e., greater VI years). Using generalized linear mixed-effects modeling (GLMM), compared more than 100 site-year combinations mast/non-mast develop VI-based regressions. found some success identifying years VIs, while models using were not supported, given data. However, containing multiple both categories successful any single-VI model, accurately predicting four sixteen site observations. provide compelling evidence mast-seeding may detectable large coniferous areas. Additional warranted signal for confounding disturbance-related effects differentiate variation signals attributable productivity contrast climatological on reflectance.