Boreal tree seed production and quality (size, nutrient content, viability) are sensitive to environmental change, yet we lack understanding of how global change agents interact over broad spatio-temporal scales. To fill this gap, our larger project will leverage historical herbaria collections (1828-2023) from locations across North America, continental-scale contemporary data collected across the boreal forest, an ecosystem-scale manipulative experiment, and spatial modeling to forecast the future of four large boreal landscapes. Our central hypothesis is that global change agents interact to alter the magnitude of reproductive output, seed quality and viability, leading to spatial variation in their effects at landscape, regional, and continental scales. Our project targets five conifer species across their distribution in North America, including two lowland species (Picea mariana, Larix laricina), and three upland species (Abies balsamea, Picea glauca, Pinus banksiana). We are examining the chemistry of leaves and seeds of historical and contemporary samples of the five focal species across their North American distribution. We aim to leverage LTER and NEON sites where these species occur (including the Harvard Forest) to collect needles, cones, and soils in 2026 or 2027 to improve our broader understanding of how global change agents interact to affect reproduction.