Seasonal variation of carbon dioxide fluxes over irrigated soybean (Glycine max L.)

In this study, variations in carbon dioxide (CO₂) fluxes resulting from gross primary production (GPP), net ecosystem exchange (NEE), and respiration (R _e) of soybean (Glycine max L.) were investigated by the Eddy Covariance method during the growing period from June to November 2005 on an irrigated sand field at the Arid Land Research Center, Tottori University in Tottori, Japan. Although climatic conditions were humid and temperate, the soybeans required frequent irrigation because of the low water holding capacity of the sandy soil at the field site. Finally, it has been found that the accumulated NEE, GPP, and R _e fluxes of soybean over 126 days amount to ?93, 319, and 226 gC m^?2, respectively. Furthermore, the average ratio of GPP to R _e was 1.4 and the average ratio of NEE to GPP was about ?0.29 for the growth period of soybean. Daily maximum NEE of ?3.8 gC m^?2 occurred when LAI was 1.1.     

1981—2019年全球陆地生态系统碳通量变化特征及其驱动因子

陆地生态系统碳汇显著降低大气CO₂浓度上升和全球变暖的速率,受人类活动和气候变化的影响,陆地生态系统碳通量具有强烈的时空变化,其估算结果仍存在较大的不确定性,不同因子的贡献尚不清晰。为此,利用遥感驱动的陆地生态系统过程模型BEPS模拟分析了1981—2019年全球陆地生态系统碳通量的时空变化特征,评价了大气CO₂浓度、叶面积指数（Leaf Area Index,LAI）、氮沉降、气候变化对全球陆地生态系统碳收支变化的贡献。1981—2019年全球陆地生态系统总初级生产力（Gross Primary Productivity,GPP）、净初级生产力（Net Primary Productivity,NPP）和净生态系统生产力（Net Ecosystem Productivity,NEP）的平均值分别为115.3、51.3和2.7 Pg·a^-1（以碳质量计,下同）,上升速率分别为0.47、0.21和0.06 Pg·a^-1。全球大部分区域GPP和NPP显著增加,NEP显著上升（p<0.05）的区域明显少于GPP和NPP。1981—2019年,全球NEP累积为105.2 Pg,森林、稀树草原及灌木、农田和草地的贡献分别为76.4、15.8、9.4和3.6 Pg。CO₂浓度、LAI、氮沉降和气候变化各自对NEP的累积贡献分别为58.4、20.6、0.7和-43.6 Pg,全部4个因子变化对NEP的累积贡献为39.8 Pg,其中CO₂浓度上升是近40 a全球陆地生态系统NEP上升的主要贡献因子,其次为LAI。     

Net CO2 exchange of a subarctic mountain birch ecosystem

Summary  Net ecosystem CO₂ exchange was measured over a mountain birch forest in northern Finland throughout the growing season. The maximal net CO₂ uptake rate of about − 0.5 mg(CO₂) m⁻² s⁻¹ was observed at the end of July. The highest nocturnal respiration rates in early August were 0.2 mg(CO₂) m⁻² s⁻¹. The daily CO₂ balances during the time of maximal photosynthesis were about −15 g(CO₂) m⁻² d⁻¹. The mountain birch forest acted as a net sink of CO₂ from 30 June to 28 August. During that period the net CO₂ balance was −448 g(CO₂)m⁻². The interannual representativeness of the observed balances was studied using a simplified daily balance model, with daily mean global radiation and air temperature as the input parameters. The year-to-year variation in the phenological development was parameterised as a function of the cumulative effective temperature sum. The daily balance model was used for estimating the variability in the seasonal CO₂ balances due to the timing of spring and meteorological factors. The sink term of CO₂ in 1996 was lower than the 15-year mean, mainly due to the relatively late emergence of the leaves. Received October 11, 1999 Revised April 25, 2000     

Continuous measurement of methane flux over a larch forest using a relaxed eddy accumulation method

We measured the methane flux of a forest canopy throughout a year using a relaxed eddy accumulation (REA) method. This sampling system was carefully validated against heat and CO₂ fluxes measured by the eddy covariance method. Although the sampling system was robust, there were large uncertainties in the measured methane fluxes because of the limited precision of the methane gas analyzer. Based on the spectral characteristics of signals from the methane analyzer and the diurnal variations in the standard deviation of the vertical wind velocity, we found the daytime and nighttime precision of half-hourly methane flux measurements to be approximately 1.2 and 0.7?μg?CH₄?m^?2?s^?1, respectively. Additional uncertainties caused by the dilution effect were estimated to affect the accuracy by as much as 0.21?μg?CH₄?m^?2?s^?1 on a half-hourly basis. Diurnal and seasonal variations were observed in the measured fluxes. The biological emission from plant leaves was not observed in our studies, and thus could be negligible at the canopy-scale exchange. The annual methane sink was 835?±?175?mg?CH₄?m^?2?year^?1 (8.35?kg?CH₄?ha^?1?year^?1), which was comparable to the flux range of 379–2,478?mg?CH₄?m^?2?year^?1 previously measured in other Japanese forest soils. This study indicated that the REA method could be a promising technique to measure canopy scale methane fluxes over forests, but further improvement of precision of the analyzer will be required.     

Detecting a Terrestrial Biosphere Sink for Carbon Dioxide: Interannual Ecosystem Modeling for the Mid-1980s

There is considerable uncertainty as to whether interannual variability in climate and terrestrial ecosystem production is sufficient to explain observed variation in atmospheric carbon content over the past 20–30 years. In this paper, we investigated the response of net CO₂ exchange in terrestrial ecosystems to interannual climate variability (1983 to 1988) using global satellite observations as drivers for the NASA-CASA (Carnegie-Ames-Stanford Approach) simulation model. This computer model of net ecosystem production (NEP) is calibrated for interannual simulations driven by monthly satellite vegetation index data (NDVI) from the NOAA Advanced Very High Resolution Radiometer (AVHRR) at 1 degree spatial resolution. Major results from NASA-CASA simulations suggest that from 1985 to 1988, the northern middle-latitude zone (between 30 and 60°N) was the principal region driving progressive annual increases in global net primary production (NPP; i.e., the terrestrial biosphere sink for carbon). The average annual increase in NPP over this predominantly northern forest zone was on the order of +0.4 Pg (10¹⁵ g) C per year. This increase resulted mainly from notable expansion of the growing season for plant carbon fixation toward the zonal latitude extremes, a pattern uniquely demonstrated in our regional visualization results. A net biosphere source flux of CO₂ in 1983–1984, coinciding with an El Niño event, was followed by a major recovery of global NEP in 1985 which lasted through 1987 as a net carbon sink of between 0.4 and 2.6 Pg C per year. Analysis of model controls on NPP and soil heterotrophic CO₂ fluxes (R_h) suggests that regional warming in northern forests can enhance ecosystem production significantly. In seasonally dry tropical zones, periodic drought and temperature drying effects may carry over with at least a two-year lag time to adversely impact ecosystem production. These yearly patterns in our model-predicted NEP are consistent in magnitude with the estimated exchange of CO₂ by the terrestrial biosphere with the atmosphere, as determined by previous isotopic (¹³C) deconvolution analysis. Ecosystem simulation results can help further target locations where net carbon sink fluxes have occurred in the past or may be verified in subsequent field studies.     

Temperature and snow-melt controls on interannual variability in carbon exchange in the high Arctic

Summary Net Ecosystem CO₂ Exchange (NEE) was studied during the summer season (June–August) at a high Arctic heath ecosystem for 5 years in Zackenberg, NE Greenland. Integrated over the 80 day summer season, the heath is presently a sink ranging from −1.4 g C m⁻² in 1997 to −23.3 g C m⁻² in 2003. The results indicate that photosynthesis might be more variable than ecosystem respiration on the seasonal timescale. The years focused on in this paper differ climatically, which is reflected in the measured fluxes. The environmental conditions during the five years strongly indicated that time of snow-melt and air temperature during the growing season are closely related to the interannual variation in the measured fluxes of CO₂ at the heath. Our estimates suggest that net ecosystem CO₂ uptake is enhanced by 0.16 g C m⁻² per increase in growing degree-days during the period of growth. This study emphasises that increased summer time air temperatures are favourable for this particular ecosystem in terms of carbon accumulation.     

Modelling Carbon Dynamics of Boreal Forest Ecosystems Using the Canadian Land Surface Scheme

The ecosystem carbon (C) and nitrogen (N) simulations recently implemented in the Canadian Land Surface Scheme (CLASS) are presented. The main calculations include plant photosynthesis, autotrophic respiration, root N uptake, litterfall, plant growth, and soil heterotrophic respiration. Model experiments are made on two boreal forest ecosystems, deciduous (aspen) and conifers (black spruce). Simulated plant, soil, and ecosystem CO₂ exchanges are analysed on half-hourly, daily, and annual time scales and compared with tower eddy correlation flux measurements and estimates from various authors. Modeled daily ecosystem CO₂ exchange explained86% and 54%, respectively, of the observed variance of eddy correlationflux at the aspen site and at the black spruce site. Annual results show that the aspen ecosystem was simulated as a C sink in both 1994 (+164 g Cm^–2) and 1996 (+142 g C m^–2), and the black spruce ecosystem wassimulated as a C sink in 1994 (+39 g C m^–2) and 1995 (+25 g Cm^–2), but as a C source in 1996 (–27 g C m^–2).     

典型温带针阔叶混交林碳收支对温度的响应特征

森林生态系统是一个庞大的碳储备系统,在当前气候变暖条件下,温度变化会对森林生态系统的碳收支过程产生重要影响。该文选择长白山温带针阔混交林森林生态系统（CBS）作为研究对象,利用多年通量及小气候观测资料分析该生态系统碳收支过程对温度的响应特征,结果显示该温带森林碳交换的季节变化特征十分明显。生态系统总初级生产力GPP、生态系统呼吸R_e和净生态系统碳交换NEE在2003—2008年的月平均变化显示,碳收支3个组分最大值均出现在夏季,GPP最大值出现在7月,R_e最大值主要出现在8月,NEE负方向的最大值主要出现在6月或7月,表现为碳吸收。在日尺度和月尺度对温度的响应上,GPP和R_e都是随温度（气温和5 cm土壤温度）呈显著的指数升高形式。在日尺度上和月尺度上, NEE对气温的响应皆是分段线性形式,先是随气温的上升而正向增大,表现为碳排放;当超过临界温度,随气温的继续上升而负值增大,表现为碳吸收。根据温度、GPP、R_e以及NEE的季节的变化,每年达到最大的GPP、R_e以及NEE的最适温度均不同,这表明了在气温变化的背景下,生态系统的最适温度也在随之改变,也表明了不考虑其它因素的影响,在气候变暖的背景下,长白山针阔混交林森林生态系统的GPP、R_e随气温的升高增大,而NEE随气温的升高而减小。     

Seasonal carbon dioxide balance and respiration of a high-arctic fen ecosystem in NE-Greenland

Summary  Turbulent fluxes of CO₂ were continuously measured by eddy correlation for three months in 1997 over a gramineous fen in a high-arctic environment at Zackenberg (74°28′12″N, 20°34′23″W) in NE-Greenland. The measurements started on 1 June, when there was still a 1–2 m cover of dry snow, and ended 26 August at a time that corresponds to late autumn at this high-arctic site. During the 20-day period with snow cover, fluxes of CO₂ to the atmosphere were small, typically 0.005 mg CO₂ m⁻² s⁻¹ (0.41 g CO₂ m⁻² d⁻¹), wheres during the thawed period, the fluxes displayed a clear diurnal variation. During the snow-free period, before the onset of vegetation growth, fluxes of CO₂ to the atmosphere were typically 0.1 mg CO₂ m⁻² s⁻¹ in the afternoon, and daily sums reached values up to almost 9 g CO₂ m⁻² d⁻¹. After 4 July, downward fluxes of CO₂ increased, and on sunny days in the middle of the growing season, the net ecosystem exchange rates attained typical values of about −0.23 mg m⁻² s⁻¹ at midday and max values of daily sums of −12 g CO₂ m⁻² d⁻¹. Throughout the measured period the fen ecosystem acted as a net-sink of 130 g CO₂ m⁻². Modelling the ecosystem respiration during the season corresponded well with eddy correlation and chamber measurements. On the basis of the eddy correlation data and the predicted respiration effluxes, an estimate of the annual CO₂ balance the calender year 1997 was calculated to be a net-sink of 20 g CO₂ m⁻² yr⁻¹. Received October 6, 1999 Revised May 2, 2000     

Future vegetation changes in thawing subarctic mires and implications for greenhouse gas exchange—a regional assessment

One of the major concerns regarding climate change in high latitudes is the potential feedback from greenhouse gases (GHG) being released from thawing peat soils. In this paper we show how vegetational patterns and associated GHG fluxes in subarctic palsa (peat mounds with a permanently frozen core) mires can be linked to climate, based on field observations from fifteen palsa sites distributed in northern Fennoscandia. Fine resolution (100?m) land cover data are combined with projections of future climate for the 21st century in order to model the potential future distribution of palsa vegetation in northern Fennoscandia. Site scale climate-vegetational relationships for two vegetation types are described by a climate suitability index computed from the field observations. Our results indicate drastic changes in the palsa vegetational patterns over the coming decades with a 97?% reduction in dry hummock areas by 2041?C2060 compared to the 1961?C1990 areal coverage. The impact of these changes on the carbon balance is a decrease in the efflux of CO₂ from 130 kilotonnes C y^?1 to a net uptake of 11 kilotonnes C y^?1 and a threefold increase in the efflux of CH₄ from 6 to 18 kilotonnes C y^?1 over the same period and over the 5,520?km² area of palsa mires. The combined effect is equivalent to a slight decrease in CO₂-C emissions, from 182 to 152 kilotonnes C y^?1. Main uncertainties involve the ability of the vegetation community to adapt to new conditions, and long-term changes in hydrology due to absence of ice and frost heaving.     

Seasonal variation and meteorological control of CO2 flux in a hilly plantation in the mountain areas of North China

The carbon cycle of terrestrial ecosystems is an important scientific issue in global climate change research.Plantation forest plays an important role in terrestrial carbon budget in China.In this study,eddy covariance flux data measured at Xiaolangdi forest ecosystem research station(XLD) in 2007 and 2008 are used to analyze the seasonal variation and meteorological control of CO2 flux in a 30-yr-old mixed plantation.The plantation forest mainly consists of Quercus variabilis,Platycladus orientalis,and Robinia pseudoacacia.The results show that the seasonal variations of net ecosystem exchange of CO2(NEE),gross primary production(GPP),and ecosystem respiration(Re) display single-peak curves.The maximum of carbon sequestration appears during May and June each year.The relative contribution of carbon release from ecosystem respiration to GPP varied slightly between 2007 and 2008.The relationship between NEE and photosynthetic active radiation(Qp) accords with the rectangular hyperbola model on diurnal scale,and shows a good linear correlation on monthly scale.The ecosystem photosynthetic parameters:the maximum photosynthetic rate(Pmax),the ecosystem photosynthetic photonyield(α),and the daytime ecosystem respiration(Rd) exhibit seasonal variations.Pmax reaches the maximum in August each year,with small interannual difference.The interannual differences of α and Rd are obvious,which is attributed to the changes of meteorological factors,such as solar radiation,vapor pressure deficit(D),precipitation,etc.Parameters Re,GPP,and NEP(net ecosystem production) have obvious exponential relations with temperature on monthly scale.There is a hysteresis in the response of GPP and NEP to temperature,i.e.,the carbon sequestration is not the maximum when the temperature reaches the peak value.The Q10 values were 1.37 and 1.45 in 2007 and 2008,respectively.On monthly scale,Re,GPP,and NEE increase as D increases,but rise slowly and even decrease when D is higher than 1.5 kPa.     

Effects of Irrigation on Nitrous Oxide, Methane and Carbon Dioxide Fluxes in an Inner Mongolian Steppe

Increased precipitation during the vegetation periods was observed in and further predicted for Inner Mongolia. The changes in the associated soil moisture may affect the biosphere-atmosphere exchange of greenhouse gases. Therefore, we set up an irrigation experiment with one watered （W） and one unwatered plot （UW） at a winter-grazed Leymus chinensis-steppe site in the Xilin River catchment, Inner Mongolia. UW only received the natural precipitation of 2005 （129 mm）, whereas W was additionally watered after the precipitation data of 1998 （in total 427 mm）. In the 3-hour resolution, we determined nitrous oxide （N20）, methane （CH4） and carbon dioxide （CO2） fluxes at both plots between May and September 2005, using a fully automated, chamber-based measuring system. N20 fluxes in the steppe were very low, with mean emissions （±s.e.） of 0.9-4-0.5 and 0.7-4-0.5 μg N m^-2 h^-1 at W and UW, respectively. The steppe soil always served as a CH4 sink, with mean fluxes of -24.1-4-3.9 and -31.1-4- 5.3 μg C m^-2 h^-1 at W and UW. Nighttime mean CO2 emissions were 82.6±8.7 and 26.3±1.7 mg C m^-2 h^-1 at W and UW, respectively, coinciding with an almost doubled aboveground plant biomass at W. Our results indicate that the ecosystem CO2 respiration responded sensitively to increased water input during the vegetation period, whereas the effects on CH4 and N2O fluxes were weak, most likely due to the high evapotranspiration and the lack of substrate for N2O producing processes. Based on our results, we hypothesize that with the gradual increase of summertime precipitation in Inner Mongolia, ecosystem CO2 respiration will be enhanced and CH4 uptake by the steppe soils will be lightly inhibited.     

European CO2 fluxes from atmospheric inversions using regional and global transport models

Approximately half of human-induced carbon dioxide (CO₂) emissions are taken up by the land and ocean, and the rest stays in the atmosphere, increasing the global concentration and acting as a major greenhouse-gas (GHG) climate-forcing element. Although GHG mitigation is now in the political arena, the exact spatial distribution of the land sink is not well known. In this paper, an estimation of mean European net ecosystem exchange (NEE) carbon fluxes for the period 1998–2001 is performed with three mesoscale and two global transport models, based on the integration of atmospheric CO₂ measurements into the same Bayesian synthesis inverse approach. A special focus is given to sub-continental regions of Europe making use of newly available CO₂ concentration measurements in this region. Inverse flux estimates from the five transport models are compared with independent flux estimates from four ecosystem models. All inversions detect a strong annual carbon sink in the southwestern part of Europe and a source in the northeastern part. Such a dipole, although robust with respect to the network of stations used, remains uncertain and still to be confirmed with independent estimates. Comparison of the seasonal variations of the inversion-based net land biosphere fluxes (NEP) with the NEP predicted by the ecosystem models indicates a shift of the maximum uptake period, from June in the ecosystem models to July in the inversions. This study thus improves on the understanding of the carbon cycle at sub-continental scales over Europe, demonstrating that the methodology for understanding regional carbon cycle is advancing, which increases its relevance in terms of issues related to regional mitigation policies.     

Sensitivity of carbon budget to historical climate variability and atmospheric CO2 concentration in temperate grassland ecosystems in China

Chinese temperate grasslands play an important role in the terrestrial carbon cycle. Based on the parameterization and validation of Terrestrial Ecosystem Model (TEM, Version 5.0), we analyzed the carbon budgets of Chinese temperate grasslands and their responses to historical atmospheric CO₂ concentration and climate variability during 1951–2007. The results indicated that Chinese temperate grassland acted as a slight carbon sink with annual mean value of 7.3 T?g C, ranging from -80.5 to 79.6 T?g C yr^-1. Our sensitivity experiments further revealed that precipitation variability was the primary factor for decreasing carbon storage. CO₂ fertilization may increase the carbon storage (1.4 %) but cannot offset the proportion caused by climate variability (-15.3 %). Impacts of CO₂ concentration, temperature and precipitation variability on Chinese temperate grassland cannot be simply explained by the sum of the individual effects. Interactions among them increased total carbon storage of 56.6 T?g C which 14.2 T?g C was stored in vegetation and 42.4 T?g C was stored in soil. Besides, different grassland types had different responses to climate change and CO₂ concentration. NPP and R_H of the desert and forest steppes were more sensitive to precipitation variability than temperature variability while the typical steppe responded to temperature variability more sensitively than the desert and forest steppes.     

Biogenic hydrogen sulphide emissions and non-sea sulfate aerosols over the Indian Sundarban mangrove forest

Temporal variations in atmospheric hydrogen sulphide concentrations and its biosphere-atmosphere exchanges were studied in the World’s largest mangrove ecosystem, Sundarbans, India. The results were used to understand the possible contribution of H₂S fluxes in the formation of atmospheric aerosol of different size classes (e.g. accumulation, nucleation and coarse mode). The mixing ratio of hydrogen sulphide (H₂S) over the Sundarban mangrove atmosphere was found maximum during the post-monsoon season (October to January) with a mean value of 0.59?±?0.02 ppb and the minimum during pre-monsoon (February to May) with a mean value of 0.26?±?0.01 ppb. This forest acted as a perennial source of H₂S and the sediment-air emission flux ranged between 1213?±?276 μg S m^?2 d^?1(December) and 457?±?114 μg S m^?2 d^?1 (August) with an annual mean of 768?±?240 μg S m^?2d^?1. The total annual emissions of H₂S from the Indian Sundarban were estimated to be 1.2?±?0.6 Tg S. The accumulation mode of aerosols was found to be more enriched with non-sea salt sulfate with an average loading of 5.74 μg m^?3 followed by the coarse mode (5.18 μg m^?3) and nucleation mode (1.18 μg m^?3). However, the relative contribution of Non-sea salt sulfate aerosol to total sulfate aerosol was highest in the nucleation mode (83%) followed by the accumulation (73%) and coarse mode (58%). Significant positive relations between H₂S flux and different modes of NSS indicated the likely link between H₂S, a dominant precursor for the non-sea salt sulfate, and non-sea sulfate aerosol particles. An increase in H₂S emissions from the mangrove could result in an increase in enhanced NSS in aerosol and associated cloud albedo, and a decrease in the amount of incoming solar radiation reaching the Sundarban mangrove forest.     

Measurement of methane flux over an evergreen coniferous forest canopy using a relaxed eddy accumulation system with tuneable diode laser spectroscopy detection

Very few studies have conducted long-term observations of methane (CH₄) flux over forest canopies. In this study, we continuously measured CH₄ fluxes over an evergreen coniferous (Japanese cypress) forest canopy throughout 1?year, using a micrometeorological relaxed eddy accumulation (REA) system with tuneable diode laser spectroscopy (TDLS) detection. The Japanese cypress forest, which is a common forest type in warm-temperate Asian monsoon regions with a wet summer, switched seasonally between a sink and source of CH₄ probably because of competition by methanogens and methanotrophs, which are both influenced by soil conditions (e.g., soil temperature and soil moisture). At hourly to daily timescales, the CH₄ fluxes were sensitive to rainfall, probably because CH₄ emission increased and/or absorption decreased during and after rainfall. The observed canopy-scale fluxes showed complex behaviours beyond those expected from previous plot-scale measurements and the CH₄ fluxes changed from sink to source and vice versa.     

长江三角洲地区净生态系统二氧化碳通量及浓度的数值模拟

净生态系统碳通量(NEE)的计算对于准确模拟区域碳通量和大气CO₂浓度的时空变化至关重要。本文利用中尺度大气-温室气体耦合模式WRF-GHG(Weather Research and Forecasting Model with Greenhouse Gases Module),对2010年7月28日至2010年8月2日期间影响长江三角洲地区大气CO₂浓度及时空分布的各种过程进行了详尽模拟。结果表明,植被光合呼吸模型(VPRM)能模拟不同植被下垫面NEE的日变化;WRF-GHG模拟的大气CO₂浓度日变化与观测相吻合,但低估了大气CO₂浓度5~15 ppm(ppm表示10-6),这可能与人为排放源的低估、VPRM参数的不确定性以及气象场模拟的不准确性有关。太湖和植被覆盖较好的地区如浙江北部山区是该地区的主要碳汇,而城市为CO₂的主要排放源。太湖和陆地生态系统对区域内碳循环起到一定的调节作用,减缓区域大气CO₂浓度的升高。此外,局地气象条件如湖陆风对太湖周边地区大气CO₂浓度有显著影响。     

Carbon Stores,Sinks, and Sources in Forests of Northwestern Russia: Can We Reconcile Forest Inventories with Remote Sensing Results?

Forest inventories and remote sensing are the two principal data sources used to estimate carbon (C) stocks and fluxes for large forest regions. National governments have historically relied on forest inventories for assessments but developments in remote sensing technology provide additional opportunities for operational C monitoring. The estimate of total C stock in live forest biomass modeled from Landsat imagery for the St. Petersburg region was consistent with estimates derived from forest inventory data for the early 1990s (272 and 269 TgC, respectively). The estimates of mean C sink in live forest biomass also agreed well (0.36 and 0.34 Mg C ha^–1 yr^–1). Virtually all forest lands were accumulating C in live biomass, however when the net change in total ecosystem C stock was considered, 19% of the forest area were a net source of C. The average net C sink in total ecosystem biomass is quite weak (0.08 MgC ha^–1 yr^–1 and could be reversed by minor increases in harvest rates or a small decline in biomass growth rates.     

Comparing CO2 Storage and Advection Conditions at Night at Different Carboeuroflux Sites

Anemometer and CO₂ concentration data from temporary campaigns performed at six CARBOEUROFLUX forest sites were used to estimate the importance of non-turbulent fluxes in nighttime conditions. While storage was observed to be significant only during periods of both low turbulence and low advection, the advective fluxes strongly influence the nocturnal CO₂ balance, with the exception of almost flat and highly homogeneous sites. On the basis of the main factors determining the onset of advective fluxes, the ‘advection velocity’, which takes net radiation and local topography into account, was introduced as a criterion to characterise the conditions of storage enrichment/depletion. Comparative analyses of the six sites showed several common features of the advective fluxes but also some substantial differences. In particular, all sites where advection occurs show the onset of a boundary layer characterised by a downslope flow, negative vertical velocities and negative vertical CO₂ concentration gradients during nighttime. As a consequence, vertical advection was observed to be positive at all sites, which corresponds to a removal of CO₂ from the ecosystem. The main differences between sites are the distance from the ridge, which influences the boundary-layer depth, and the sign of the mean horizontal CO₂ concentration gradients, which is probably determined by the source/sink distribution. As a consequence, both positive and negative horizontal advective fluxes (corresponding respectively to CO₂ removal from the ecosystem and to CO₂ supply to the ecosystem) were observed. Conclusive results on the importance of non-turbulent components in the mass balance require, however, further experimental investigations at sites with different topographies, slopes, different land covers, which would allow a more comprehensive analysis of the processes underlying the occurrence of advective fluxes. The quantification of these processes would help to better quantify nocturnal CO₂ exchange rates.