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.

<p><strong>Abstract.</strong> 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<sub>2</sub>), 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<sub>2</sub> < 0.5), ecosystem respiration (R<sub>2</sub> > 0.6), gross primary production (R<sub>2</sub> > 0.7), latent heat (R<sub>2</sub> > 0.7), sensible heat (R<sub>2</sub> > 0.7), net radiation (R<sub>2</sub> > 0.8). ML methods predicted very well the across sites variability and the seasonal cycle (R<sub>2</sub> > 0.7) of the observed fluxes, while the eight days deviations from the mean seasonal cycle were not well predicted (R<sub>2</sub> < 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.</p>

This paper analyses the temporal dynamics of soil water balance components in a representative recharge area of the Sierra de Gádor (Almeria, southeastern Spain) in two hydrological years. Two approaches are used to estimate daily potential recharge (PR): Approach 1 based on deriving PR from the water balance as the difference between measurements of rainfall (P) and actual evapotranspiration (E) obtained by eddy covariance; and Approach 2 with PR obtained from the dynamic pattern of the soil moisture (y) recorded at two depths in the site's thin soil (average 0.35 m thickess). For the hydrological year 2003/04, which was slightly drier than the 30-year average, E accounted for 64% of rainfall and occurred mainly in late spring and early summer. The PR estimated by Approach 1 was 181 AE 18 mm year -1 (36% of rainfall), suggesting an effective groundwater recharge in the study area. In the unusually dry hydrological year 2004/05, E was about 215 mm year -1 , close to the annual rainfall input, and allowing very little (8 AE 12 mm year -1 ) PR according to Approach 1. Estimation of PR based on Approach 2 resulted in PR rates lower than those found by Approach 1, because Approach 2 does not take into account the recharge that occurs through preferential flow pathways (cracks, joints and fissures) which were not monitored with the y probes. Moreover, using Approach 2, the PR estimates differed widely depending on the time scale considered: with daily mean y data, PR estimation was lower, especially in late spring, while y data at 30 min resolution yielded a more reliable prediction of the fraction of total PR resulting from the downward movement of soil water by gravity.Key words potential recharge; evapotranspiration; soil moisture; soil water balance; Mediterranean; Sierra de Gádor Dynamique temporelle des composantes du bilan hydrique du sol dans une chaine karstique du sud-est de l'Espagne: estimation de la recharge potentielle Résumé Ce travail analyse la dynamique temporelle des composantes du bilan hydrique du sol dans une zone de recharge représentative de la Sierra de Gador (Almeria, SE de l'Espagne) pendant deux années hydrologiques. Deux approches ont été utilisées pour estimer la recharge potentielle journalière (RP): dans l'approche 1, RP est calculée à partir de la différence entre les mesures de pluie (P) et d'évapotranspiration réelle (E) obtenues par la méthode de la covariance turbulente; dans l'approche 2, RP est obtenue à partir de l'humidité du sol (y) mesurée à deux profondeurs dans les sols peu profonds du site (épaisseur moyenne de 0.35 m). Pendant l'année hydrologique

CO₂ exchange between terrestrial ecosystems and the atmosphere is key to understanding the feedbacks between climate change and the land surface. In regions with carbonaceous parent material, CO₂ exchange patterns occur that cannot be explained by biological processes, such as disproportionate outgassing during daytime or nighttime CO₂ uptake during periods when all vegetation is senescent. Neither of these phenomena can be attributed to carbonate weathering reactions, since their CO₂ exchange rates are too small. Soil ventilation induced by high atmospheric turbulence is found to explain atypical CO₂ exchange between carbonaceous systems and the atmosphere. However, by strongly altering subsurface CO₂ concentrations, ventilation can be expected to influence carbonate weathering rates. By imposing ventilation-driven CO₂ outgassing in a carbonate weathering model, we show here that carbonate geochemistry is accelerated and does play a surprisingly large role in the observed CO₂ exchange patterns. We found that by rapidly depleting soil CO₂ during daytime, ventilation disturbs soil carbonate equilibria and therefore strongly magnifies daytime carbonate precipitation and associated CO₂ production. At night, ventilation ceases and the depleted CO₂ concentrations increase steadily. Dissolution of carbonate is now enhanced, which consumes CO₂ and largely compensates for the enhanced daytime carbonate precipitation. This is why only a relatively small effect on global carbonate weathering rates is to be expected. On the short term, however, ventilation has a drastic effect on synoptic carbonate weathering rates, resulting in a pronounced diel pattern that exacerbates the non-biological behavior of soil-atmosphere CO₂ exchanges in dry regions with carbonate soils

Knowledge of all the mechanisms and processes involved in soil CO₂ emissions is essential to close the global carbon cycle. Apart from molecular diffusion, the main physical component of such CO₂ exchange is soil ventilation. Advective CO₂ transport, through soil or snow, has been correlated with the wind speed, friction velocity or pressure (p). Here we examine variations in subterranean CO₂ molar fractions (χ_c) over two years within a vertical profile (1.5 m) in a semiarid ecosystem, as influenced by short-timescale p changes.

Analyses to determine the factors involved in the variations in subterranean χ_c were differentiated between the growing period and the dry period. In both periods it was found that variations in deep χ_c (0.5–1.5 m) were due predominantly to static p variations and not to wind or biological influences. Within a few hours, the deep χ_c can vary by fourfold showing a pattern with two cycles per day, due to p oscillations caused by atmospheric tides. By contrast, shallow χ_c (0.15 m) generally has one cycle per day as influenced by biological factors like soil water content and temperature in both periods, while the wind was an important factor in shallow χ_c variations only during the dry period. Evidence of emissions was registered in the atmospheric boundary layer by eddy covariance during synoptic pressure changes when subterranean CO₂ was released; days with rising barometric pressure – when air accumulated belowground, including soil-respired CO₂ – showed greater ecosystem uptake than days with falling pressure. Future assessments of the net ecosystem carbon balance should not rely exclusively on Fick's law to calculate soil CO₂ effluxes from profile data

Two years of continuous measurements of net ecosystem exchange (NEE) using the eddy covariance technique were made over a Mediterranean alpine shrubland. These ecosystems are little studied, since they have little CO 2 exchange potential. Nevertheless, their high susceptibility to environmental changes is far from being understood, 5 introducing some uncertainty in terrestrial CO 2 and water vapour assessments. High altitude sites might be undergoing a transition from sink to source of CO 2 , due to their high vulnerability to climate change. Indeed, this ecosystem was found to be a net source of CO 2 (+52 g C m −2 and +48 g C m −2 for 2007 and 2008) during the two-year study period. To understand the reasons underlying this net release of CO 2 into the 10 atmosphere, we analysed the drivers of seasonal variability in NEE across these two years. We observed that the soil water availability -driven by the precipitation pattern -and the photosynthetic photon flux density (PPFD) are the key factors for understanding both the carbon sequestration potential and the duration of the photosynthetic period during the growing season. Finally, the effects of the Burba correction for both 15 NEE and evapotranspiration (ET) are evaluated. This correction can sometimes be neglected on a daily basis, but becomes rather important in long-term assessments. For instance, the annual CO 2 budget in 2007 turned from sink (−136 g C m −2 ) to source (+52 g C m −2 ) when the Burba correction was taken into account.

Abstract.To study the effects of warming and declining species richness on the carbon balance of grassland communities, model ecosystems containing one, three or nine species were exposed to ambient and elevated (ambient +3 • C) air temperature. In this paper, we analyze measured ecosystem CO 2 fluxes to test whether ecosystem photosynthesis and respiration had acclimated to warming after 28 months of continuous heating, and whether the degree of acclimation depended on species richness. In order to test whether acclimation occurred, short term temperature response curves were established for all communities in both treatments. At similar temperatures, lower flux rates in the heated communities as compared to the unheated communities would indicate thermal acclimation. Because plant cover was significantly higher in the heated treatment, we normalized the data for plant cover. Subsequently, down-regulation of both photosynthesis and respiration was observed. Although CO 2 fluxes were larger in communities with higher species richness, species richness did not affect the degree of acclimation to warming. These results imply that models need to take thermal acclimation into account to simulate photosynthesis and respiration in a warmer world.

Solar radiation is a key driver of energy and carbon fluxes in natural ecosystems. Radiation measurements are essential for interpreting ecosystem scale greenhouse gases and energy fluxes as well as many other observations performed at ecosystem stations of the Integrated Carbon Observation System (ICOS). We describe and explain the relevance of the radiation variables that are monitored continuously at ICOS ecosystem stations and define recommendations to perform these measurements with consistent and comparable accuracy. The measurement methodology and instruments are described including detailed technical specifications. Guidelines for instrumental set up as well as for operation, maintenance and data collection are defined considering both ICOS scientific objectives and practical operational constraints. For measurements of short-wave (solar) and long wave (infrared) radiation components, requirements for the ICOS network are based on available well-defined state-of-the art standards (World Meteorological Organization, International Organization for Standardization). For photosynthetically active radiation measurements, some basic instrumental requirements are based on the performance of commercially available sensors. Since site specific conditions and practical constraints at individual ICOS ecosystem stations may hamper the applicability of standard requirements, we recommend that ICOS develops mid-term coordinated actions to assess the effective level of uncertainties in radiation measurements at the network scale.