The FLUXNET2015 dataset provides ecosystem-scale data on CO 2 , water, and energy exchange between the biosphere and the atmosphere, and other meteorological and biological measurements, from 212 sites around the globe (over 1500 site-years, up to and including year 2014). These sites, independently managed and operated, voluntarily contributed their data to create global datasets. Data were quality controlled and processed using uniform methods, to improve consistency and intercomparability across sites. The dataset is already being used in a number of applications, including ecophysiology studies, remote sensing studies, and development of ecosystem and Earth system models. FLUXNET2015 includes derived-data products, such as gap-filled time series, ecosystem respiration and photosynthetic uptake estimates, estimation of uncertainties, and metadata about the measurements, presented for the first time in this paper. In addition, 206 of these sites are for the first time distributed under a Creative Commons (CC-BY 4.0) license. This paper details this enhanced dataset and the processing methods, now made available as open-source codes, making the dataset more accessible, transparent, and reproducible.

There were systematic differences in performance among the fluxes, with the following ascending order: net ecosystem exchange (R 2 < 0.5), ecosystem respiration (R 2 > 0.6), gross primary production (R 2 > 0.7), latent heat (R 2 > 0.7), sensible heat (R 2 > 0.7), and net radiation (R 2 > 0.8). The ML methods predicted the across-site variability and the mean seasonal cycle of the observed fluxes very well (R 2 > 0.7), while the 8-day deviations from the mean seasonal cycle were not well predicted (R 2 < 0.5). Fluxes were better predicted at forested and temperate climate sites than at sites in extreme climates or less represented by training data (e.g., the tropics). The evaluated large ensemble of ML-based models will be the basis of new global flux products.Published by Copernicus Publications on behalf of the European Geosciences Union.

The global terrestrial carbon sink offsets one-third of the world's fossil fuel emissions, but the strength of this sink is highly sensitive to large-scale extreme events. In 2012, the contiguous United States experienced exceptionally warm temperatures and the most severe drought since the Dust Bowl era of the 1930s, resulting in substantial economic damage. It is crucial to understand the dynamics of such events because warmer temperatures and a higher prevalence of drought are projected in a changing climate. Here, we combine an extensive network of direct ecosystem flux measurements with satellite remote sensing and atmospheric inverse modeling to quantify the impact of the warmer spring and summer drought on biosphereatmosphere carbon and water exchange in 2012. We consistently find that earlier vegetation activity increased spring carbon uptake and compensated for the reduced uptake during the summer drought, which mitigated the impact on net annual carbon uptake. The early phenological development in the Eastern Temperate Forests played a major role for the continental-scale carbon balance in 2012. The warm spring also depleted soil water resources earlier, and thus exacerbated water limitations during summer. Our results show that the detrimental effects of severe summer drought on ecosystem carbon storage can be mitigated by warming-induced increases in spring carbon uptake. However, the results also suggest that the positive carbon cycle effect of warm spring enhances water limitations and can increase summer heating through biosphere-atmosphere feedbacks.seasonal climate anomalies | carbon uptake | ecosystem fluxes | biosphere-atmosphere feedbacks | eddy covariance A n increase in the intensity and duration of drought (1, 2), along with warmer temperatures, is projected for the 21st century (3). Warmer and drier summers can substantially reduce photosynthetic activity and net carbon uptake (4). In contrast, warmer temperatures during spring and autumn prolong the period of vegetation activity and increase net carbon uptake in temperate ecosystems (5), sometimes even during spring drought (6). Atmospheric CO 2 concentrations suggest that warm-springinduced increases in carbon uptake could be cancelled out by the effects of warmer and drier summers (7). However, the extent and variability of potential compensation on net annual uptake using direct observations of ecosystem carbon exchange have not yet been examined for specific climate anomalies.In addition to perturbations of the carbon cycle, warmer spring temperatures can have an impact on the water cycle by increasing evaporation from the soil and plant transpiration (8-10), which reduces soil moisture. Satellite observations suggest that warmer spring and longer nonfrozen periods enhance summer drying via hydrological shifts in soil moisture status (11). Climate model simulations also indicate a soil moisture-temperature feedback between early vegetation green-up in spring and extreme temperatures in summer (12, 13). Soil water deficits during drou...

&lt;p&gt;&lt;strong&gt;Abstract.&lt;/strong&gt; Spatial-temporal fields of land-atmosphere fluxes derived from data-driven models can complement simulations by process-based Land Surface Models. While a number of strategies for empirical models with eddy covariance flux data have been applied, a systematic intercomparison of these methods has been missing so far. In this study, we perform a cross-validation experiment for predicting carbon dioxide (CO&lt;sub&gt;2&lt;/sub&gt;), latent heat, sensible heat and net radiation fluxes, in different ecosystem types with eleven machine learning (ML) methods from four different classes (kernel methods, neural network, tree methods, and regression splines). We employ two complementary setups: (1) eight days average fluxes based on remotely sensed data, and (2) daily mean fluxes based on meteorological data and mean seasonal cycle of remotely sensed variables. The pattern of predictions from different ML and setups were very consistent. There were systematic differences in performance among the fluxes, with the following ascending order: net ecosystem exchange (R&lt;sub&gt;2&lt;/sub&gt; &lt; 0.5), ecosystem respiration (R&lt;sub&gt;2&lt;/sub&gt; &gt; 0.6), gross primary production (R&lt;sub&gt;2&lt;/sub&gt; &gt; 0.7), latent heat (R&lt;sub&gt;2&lt;/sub&gt; &gt; 0.7), sensible heat (R&lt;sub&gt;2&lt;/sub&gt; &gt; 0.7), net radiation (R&lt;sub&gt;2&lt;/sub&gt; &gt; 0.8). ML methods predicted very well the across sites variability and the seasonal cycle (R&lt;sub&gt;2&lt;/sub&gt; &gt; 0.7) of the observed fluxes, while the eight days deviations from the mean seasonal cycle were not well predicted (R&lt;sub&gt;2&lt;/sub&gt; &lt; 0.5). Fluxes were better predicted at forested and temperate climate sites than at ones growing in extreme climates or less representated in training data (e.g. the tropics). The large ensemble of ML based models evaluated will be the basis of new global flux products.&lt;/p&gt;