Warmer or cooler spring in northern high latitudes will, for the most part, directly impact gross primary productivity (GPP) of ecosystems, but also carry consequences for the upcoming seasonal GPP. Spatiotemporal patterns of these legacy effects are still largely unknown but important for improving our understanding of how plant phenology is associated with vegetation dynamics. In this study, impacts of spring temperature anomalies on spring, summer and autumn GPP were investigated, and the dominant drivers of summer and autumn GPP including air temperature, vapor pressure deficit and soil moisture have been explored for northern ecosystems (>30°N). Three remote sensing products of seasonal GPP (GOSIF-GPP, NIRv-GPP and FluxSat-GPP) over 2001–2018, all based on a spatial resolution of 0.05°, were employed. Our results indicate that legacy effects from spring temperature are most pronounced in summer, where they have stimulating effects on the Arctic ecosystem productivity. Spring warming likely lessens the harsh climatic constraints that govern the Arctic tundra and extends the growing season length. Further south, legacy effects are mainly negative. This strengthens the hypothesis that enhanced vegetation growth in spring will increase plant water demand and stress in summer and autumn. Soil moisture is the dominant control of summer GPP in temperate regions. However, the dominant meteorological variables controlling vegetation growth may differ depending on the GPP products, highlighting the need to address uncertainties among different methods of estimating GPP.

Phenology, encompassing the timing of the start, end, and duration of the growing season, is influenced by climate warming in northern regions. Altered phenological patterns carry significant implications for the global carbon cycle by disrupting the seasonal balance between gross primary productivity (GPP) and ecosystem respiration and complicating vegetation reproductive cycles. However, many current Earth system models, including those used in the “Trends and drivers of the regional scale terrestrial sources and sinks of carbon dioxide” (TRENDY) project, may inadequately capture recent phenological trends in northern ecosystems (Sitch et al., 2024). In this study, we aim to present a comprehensive analysis of phenology patterns across northern latitudes (>45°N) over the past two decades, using outputs from twelve state-of-the-art vegetation models included in the TRENDY project. These outputs, along with the TRENDY model ensemble average, are intercompared with a remote sensing-based phenology dataset derived using the Plant Phenology Index (PPI) and MODIS data. Compared with the in-situ measurements, the PPI has demonstrated improved accuracy in capturing northern phenology, particularly for boreal evergreen forests, by reducing the confounding effects of snowmelt and soil background signals (Jin et al., 2017). Furthermore, the PPI has proven effective in estimating large-scale GPP across diverse northern ecosystems, providing a robust benchmark for evaluating the performance of vegetation models (Marsh et al., 2024). We further examine the primary climatic drivers of phenological shifts (air temperature, precipitation and radiation) and assess the extent to which TRENDY models capture these drivers and the associated phenological responses to climate warming. Our findings highlight the current gap between model projections and observed phenology, offering insights into the limitations of current carbon cycle models in representing northern ecosystem dynamics. Our study contributes to advancing our understanding of the roles of northern ecosystems in the global carbon cycle.ReferencesJin, H., Jönsson, A. M., Bolmgren, K., Langvall, O., & Eklundh, L. (2017). Disentangling remotely-sensed plant phenology and snow seasonality at northern Europe using MODIS and the plant phenology index. Remote Sensing of Environment, 198, 203-212.Marsh, H., Jin, H., Duan, Z., Holst, J., Eklundh, L., & Zhang, W. (2025). Plant Phenology Index leveraging over conventional vegetation indices to establish a new remote sensing benchmark of GPP for northern ecosystems. International Journal of Applied Earth Observation and Geoinformation, 136, 104289.Sitch, S., O’sullivan, M., Robertson, E., Friedlingstein, P., Albergel, C., Anthoni, P., ... & Zaehle, S. (2024). Trends and drivers of terrestrial sources and sinks of carbon dioxide: An overview of the TRENDY project. Global Biogeochemical Cycles, 38(7), e2024GB008102.