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.

Summary Based on review and original data, this synthesis investigates carbon pools and fluxes of Siberian and European forests (600 and 300 million ha, respectively). We examine the productivity of ecosystems, expressed as positive rate when the amount of carbon in the ecosystem increases, while (following micrometeorological convention) downward fluxes from the atmosphere to the vegetation (NEE = Net Ecosystem Exchange) are expressed as negative numbers. Productivity parameters are Net Primary Productivity (NPP=whole plant growth), Net Ecosystem Productivity (NEP = CO2 assimilation minus ecosystem respiration), and Net Biome Productivity (NBP = NEP minus carbon losses through disturbances bypassing respiration, e.g. by fire and logging). Based on chronosequence studies and national forestry statistics we estimate a low average NPP for boreal forests in Siberia: 123 gC m–2 y–1. This contrasts with a similar calculation for Europe which suggests a much higher average NPP of 460 gC m–2 y–1 for the forests there. Despite a smaller area, European forests have a higher total NPP than Siberia (1.2–1.6 vs. 0.6–0.9 × 1015 gC region–1 y–1). This arises as a consequence of differences in growing season length, climate and nutrition.  For a chronosequence of Pinus sylvestris stands studied in central Siberia during summer, NEE was most negative in a 67‐y old stand regenerating after fire (– 192 mmol m–2 d–1) which is close to NEE in a cultivated forest of Germany (– 210 mmol m–2 d–1). Considerable net ecosystem CO2‐uptake was also measured in Siberia in 200‐ and 215‐y old stands (NEE:174 and – 63 mmol m–2 d–1) while NEP of 7‐ and 13‐y old logging areas were close to the ecosystem compensation point. Two Siberian bogs and a bog in European Russia were also significant carbon sinks (– 102 to – 104 mmol m–2 d–1). Integrated over a growing season (June to September) we measured a total growing season NEE of – 14 mol m–2 summer–1 (– 168 gC m–2 summer–1) in a 200‐y Siberian pine stand and – 5 mol m–2 summer–1 (– 60 gC m–2 summer–1) in Siberian and European Russian bogs. By contrast, over the same period, a spruce forest in European Russia was a carbon source to the atmosphere of (NEE: + 7 mol m–2 summer–1 = + 84 gC m–2 summer–1). Two years after a windthrow in European Russia, with all trees being uplifted and few successional species, lost 16 mol C m–2 to the atmosphere over a 3‐month in summer, compared to the cumulative NEE over a growing season in a German forest of – 15.5 mol m–2 summer–1 (– 186 gC m–2 summer–1; European flux network annual averaged – 205 gC m–2 y–1).  Differences in CO2‐exchange rates coincided with differences in the Bowen ratio, with logging areas partitioning most incoming radiation into sensible heat whereas bogs partitioned most into evaporation (latent heat). Effects of these different surface energy exchanges on local climate (convective storms and fires) and comparisons with the Canadian BOREAS experiment are discussed.  Following a classification of disturbances and their effects on ecosystem carb...

The study presents a data set of above-ground biomass (AGB), structure, spacing and fire regime, for 24 stands of pristine Siberian Scots pine (Pinus sylvestris) forests with lichens (n = 20) or Vaccinium/mosses (n = 4) as ground cover, along four chronosequences. The stands of the "lichen" site type (LT) were stratified into three chronosequences according to stand density and fire history. Allometric equations were established from 90 sample trees for stem, coarse branch, fine branch, twig and needle biomass. The LT stands exhibited a low but sustained biomass accumulation until a stand age of 383 years. AGB reached only 6-10 kg m after 200 years depending on stand density and fire history compared to 20 kg m in the "Vaccinium" type (VT) stands. Leaf area index (LAI) in the LT stands remained at 0.5-1.5 and crown cover was 30-60%, whereas LAI reached 2.5 and crown cover was >100% in the VT stands. Although nearest-neighbour analyses suggested the existence of density-dependent mortality, fire impact turned out to have a much stronger effect on density dynamics. Fire scar dating and calculation of mean and initial fire return intervals revealed that within the LT stands differences in structure and biomass were related to the severity of fire regimes, which in turn was related to the degree of landscape fragmentation by wetlands. Self-thinning analysis was used to define the local carrying capacity for biomass. A series of undisturbed LT stands was used to characterise the upper self-thinning boundary. Stands that had experienced a moderate fire regime were positioned well below the self-thinning boundary in a distinct fire-thinning band of reduced major axis regression slope -0.26. We discuss how this downward shift resulted from alternating phases of density reduction by fire and subsequent regrowth. We conclude that biomass in Siberian Scots pine forests is strongly influenced by fire and that climate change will affect ecosystem functions predominantly via changes in fire regimes.

No abstract

Measurements of aboveground biomass and nitrogen (N) nutrition were made during July 1993 in 50-, 130-, and 380-year-old stands of Larixgmelinii (Rupr.) Rupr. in eastern Siberia. Constituting six forest types based on understorey plants, the stands were representative of vegetation throughout the Yakutsk region. Average tree height, diameter, and density ranged from 2 m, 23 mm, and 50 800 stems/ha in the 50-year-old stand to 11 m, 160 mm, and 600 stems/ha in the oldest stand. Aboveground biomass in the 50-year-old stand was 4.4 kg•m−2, and the aboveground N pool was 1.1 mol•m−2. This was slightly higher than the N pool in a 125-year-old stand with a Ledum understorey (1.0 mol•m−2), despite its higher biomass (7.2 kg•m−2). The highest observed aboveground biomass in a 125-year-old stand (characterized by the N2-fixing understorey plant Alnasterfruticosa) reached 12.0 kg•m−2, but the corresponding N pool was only 1.6 mol•m−2. In the oldest stand, aboveground biomass was 8.9 kg•m−2 and the N pool was 1.1 mol•m−2. There was thus a relatively constant quantity of N in the aboveground biomass of stands differing in age by almost 400 years. We postulate that N sets a limit on carbon accumulation in this boreal forest type. Trees were extremely slow growing, and there was essentially no aboveground biomass accumulation between the ages of 130 and 380 years because of a lack of available N. This conclusion was supported by graphical analysis indicating that the self-thinning process in our stands was not governed by the availability of radiation according to allometric theory. Much of the available N was used in the production of tree stems where 86% of the aboveground N (and 96% of aboveground biomass) was immobilized in the oldest stand. N in wood of the old stand exceeded the N pool in the litter layer and was 20% of the N pool in the Ah horizon. The processes of carbon and N partitioning were further explored by the estimation of carbon and N fluxes during three periods of forest development. We calculated a loss of ecosystem N during the period of self-thinning, while in the mature stands the N cycle appeared to be very tight. The immobilized N is returned from the wood into a plant-available form only by a recurrent fire cycle, which regenerates the N cycle. Thus fire is an essential component for the persistence of the L. gmelinii forest.