Direct Measurements Of CO2 Fluxes Research Articles

Characteristics of atmospheric CO2 fluxes and the estimation of their potential sources around Boseong Standard Weather Observatory (BSWO)

We report the characteristics of the mixing ratios and fluxes of atmospheric carbon dioxide, observed in the period from July 2015 to December 2019. Measurements were noted for three different levels (60 m, 140 m, and 300 m above ground level (a.g.l.)) on a very tall (~300 m) tower in the Boseong Standard Weather Observatory (BSWO), located at the center of barley–rice double-crop fields in Boseong-gun, Jeolla-do, South Korea. The averaged CO2 mixing ratios show 420.7 ± 18.8 (60 m), 410.9 ± 16.7 (140 m) and 403.4 ± 16.7 ppm (300 m). The mixing ratios measured at the 140-m and 300-m heights display a typical seasonal pattern that is lower in the summer due to vegetation uptake and higher in the winter due to plant decaying combined with fossil-fuel emissions, but those at 60 m presented higher values in the spring and autumn, mainly owing to agricultural activities. During the study period, the averaged CO2 fluxes are −0.28 ± 8.81 (60 m), 0.34 ± 8.08 (140 m), and 0.79 ± 9.08 μmol m-2 s-1 (300 m). The monthly averaged fluxes show two dominant negative peaks, one in April and another in August, mainly caused by CO2 uptake during the barley and rice growing periods, respectively. The fluxes represent a strong correlation between daytime solar radiation and photosynthesis and between nighttime air temperatures and respiration. Directional analysis of the fluxes revealed dominant anthropogenic CO2 sources from not only the main industrial areas in the NE directions using geographic information system (GIS), but also from the aquaculture industry in the SE. We compare our measured CO2 fluxes (top-down) with a few bottom-up emissions datasets, which reveal an underestimation of the range from 31% to 69% within the footprint areas. The application of direct measurements of CO2 fluxes, combined with estimation of potential vicinity sources, can be a useful tool to detect local unknown or missing sources and supplement the gap between top-down and bottom-up approaches.

Groundwater mixing in a heterogeneous multilayer aquifer driven by geogenic CO2 fluxes: Evidence from chemical and isotopic composition of Ferrarelle waters (Riardo Plain, southern Italy)

The successful management of carbon in the Earth's crust is critical for mitigating the increase of anthropogenic CO2 in the atmosphere. Carbon Capture and Storage (CCS) requires an understanding of the behavior of carbon in the crust and the development of robust monitoring techniques to constrain the movement, mechanisms, and pathways for any potential CO2 leakage. Here, we examine an aquifer from the Riardo Plain (Campania Region, southern Italy), which serves as a suitable natural analogue for CO2 migration to the critical zone (i.e., shallow crust and aquifers) and as a case study to evaluate the geochemical processes that occur when CO2-saturated fluids mix with freshwater in shallow aquifers. We investigate the behavior of various geochemical constituents (major and trace elements, δ18O–H2O, δ13C-DIC, and Rn content). Water from this area has a high degree of mineralization (EC 2500–3000 μS/cm), high HCO3- (~2.5 g/L), is saturated with respect to CaCO3, and is enriched in alkali ions (e.g., Na+ + K+). The high degree of mineralization occurs in groundwater that discharges from the basal aquifer of the Roccamonfina volcanic edifice (~6 km NW), with vast CO2 inputs that promote host rock leaching. Superficial volcanic aquifers are recharged by fresh meteoric precipitation when groundwater flows from carbonates at the edge of the plain to aquifers hosted in the southeastern slope of the Roccamonfina volcano. The presence of normal faults in this area permits natural upwelling of CO2-rich groundwater, which locally mixes with shallow freshwater present within the upper volcanic succession. Significant (R > 0.8) linear correlations between conservative elements suggest that groundwater geochemistry is dominated by a mixture of two main endmembers: (i) deep/mineralized waters and (ii) shallow/diluted waters. The intrusion of freshwater to volcanic aquifers induces oxidation, leading to adsorption of select elements (e.g., As and Ba) onto Fe-oxyhydroxide precipitates within these aquifers. Geochemical modeling suggests that CO2 saturation approaches 3 g/L, which agrees with direct measurements of CO2 flux. We conclude that our conceptual geochemical model helps to constrain mixing of CO2 with freshwater and to diagnose the secondary geochemical processes that influence aqueous geochemistry within CO2-influenced groundwater.

Applicability and consequences of the integration of alternative models for CO<sub>2</sub> transfer velocity into a process-based lake model

Abstract. Freshwater lakes are important in carbon cycling, especially in the boreal zone where many lakes are supersaturated with the greenhouse gas carbon dioxide (CO2) and emit it to the atmosphere, thus ventilating carbon originally fixed by the terrestrial system. The exchange of CO2 between water and the atmosphere is commonly estimated using simple wind-based parameterizations or models of gas transfer velocity (k). More complex surface renewal models, however, have been shown to yield more correct estimates of k in comparison with direct CO2 flux measurements. We incorporated four gas exchange models with different complexity into a vertical process-based physico-biochemical lake model, MyLake C, and assessed the performance and applicability of the alternative lake model versions to simulate air–water CO2 fluxes over a small boreal lake. None of the incorporated gas exchange models significantly outperformed the other models in the simulations in comparison to the measured near-surface CO2 concentrations or respective air–water CO2 fluxes calculated directly with the gas exchange models using measurement data as input. The use of more complex gas exchange models in the simulation, on the contrary, led to difficulties in obtaining a sufficient gain of CO2 in the water column and thus resulted in lower CO2 fluxes and water column CO2 concentrations compared to the respective measurement-based values. The inclusion of sophisticated and more correct models for air–water CO2 exchange in process-based lake models is crucial in efforts to properly assess lacustrine carbon budgets through model simulations in both single lakes and on a larger scale. However, finding higher estimates for both the internal and external sources of inorganic carbon in boreal lakes is important if improved knowledge of the magnitude of CO2 evasion from lakes is included in future studies on lake carbon budgets.

Organic carbon transfers in the subtropical Red River system (Viet Nam): insights on CO2 sources and sinks

The Red River, draining a 169,000 km2 watershed, is the second largest river in Viet Nam and constitutes the main source of water for a large percentage of the population of North Viet Nam. Here we present the results of an investigation into the spatial distribution and temporal dynamics of particulate and dissolved organic carbon (POC and DOC, respectively) in the Red River Basin. POC concentrations ranged from 0.24 to 5.80 mg C L−1 and DOC concentrations ranged from 0.26 to 5.39 mg C L−1. The application of the Seneque/Riverstrahler model to monthly POC and DOC measurements showed that, in general, the model simulations of the temporal variations and spatial distribution of organic carbon (OC) concentration followed the observed trends. They also show the impact of high population densities (up to 994 inhab km−2 in the delta area) on OC inputs in surface runoff from the different land use classes and from urban point sources. A budget of the main fluxes of OC in the whole river network, including diffuse inputs from soil leaching and runoff and point sources from urban centers, as well as algal net primary production and heterotrophic respiration was established using the model results. It shows the predominantly heterotrophic character of the river system and provides an estimate of CO2 emissions from the river of 330 Gg C year−1. This value is in reasonable agreement with the few available direct measurements of CO2 fluxes in the downstream part of the river network.

Long-Term Release of Carbon Dioxide from Arctic Tundra Ecosystems in Alaska

Releases of the greenhouse gases carbon dioxide (CO2) and methane (CH4) from thawing permafrost are expected to be among the largest feedbacks to climate from arctic ecosystems. However, the current net carbon (C) balance of terrestrial arctic ecosystems is unknown. Recent studies suggest that these ecosystems are sources, sinks, or approximately in balance at present. This uncertainty arises because there are few long-term continuous measurements of arctic tundra CO2 fluxes over the full annual cycle. Here, we describe a pattern of CO2 loss based on the longest continuous record of direct measurements of CO2 fluxes in the Alaskan Arctic, from two representative tundra ecosystems, wet sedge and heath tundra. We also report on a shorter time series of continuous measurements from a third ecosystem, tussock tundra. The amount of CO2 loss from both heath and wet sedge ecosystems was related to the timing of freeze-up of the soil active layer in the fall. Wet sedge tundra lost the most CO2 during the anomalously warm autumn periods of September–December 2013–2015, with CH4 emissions contributing little to the overall C budget. Losses of C translated to approximately 4.1 and 1.4% of the total soil C stocks in active layer of the wet sedge and heath tundra, respectively, from 2008 to 2015. Increases in air temperature and soil temperatures at all depths may trigger a new trajectory of CO2 release, which will be a significant feedback to further warming if it is representative of larger areas of the Arctic.

Inferring <sup>222</sup>Rn soil fluxes from ambient <sup>222</sup>Rn activity and eddy covariance measurements of CO<sub>2</sub>

Abstract. We present a new methodology, which we call Single Pair of Observations Technique with Eddy Covariance (SPOT-EC), to estimate regional-scale surface fluxes of 222Rn from tower-based observations of 222Rn activity concentration, CO2 mole fractions and direct CO2 flux measurements from eddy covariance. For specific events, the regional (222Rn) surface flux is calculated from short-term changes in ambient (222Rn) activity concentration scaled by the ratio of the mean CO2 surface flux for the specific event to the change in its observed mole fraction. The resulting 222Rn surface emissions are integrated in time (between the moment of observation and the last prior background levels) and space (i.e. over the footprint of the observations). The measurement uncertainty obtained is about ±15 % for diurnal events and about ±10 % for longer-term (e.g. seasonal or annual) means. The method does not provide continuous observations, but reliable daily averages can be obtained. We applied our method to in situ observations from two sites in the Netherlands: Cabauw station (CBW) and Lutjewad station (LUT). For LUT, which is an intensive agricultural site, we estimated a mean 222Rn surface flux of (0.29 ± 0.02) atoms cm−2 s−1 with values > 0.5 atoms cm−2 s−1 to the south and south-east. For CBW we estimated a mean 222Rn surface flux of (0.63 ± 0.04) atoms cm−2 s−1. The highest values were observed to the south-west, where the soil type is mainly river clay. For both stations good agreement was found between our results and those from measurements with soil chambers and two recently published 222Rn soil flux maps for Europe. At both sites, large spatial and temporal variability of 222Rn surface fluxes were observed which would be impractical to measure with a soil chamber. SPOT-EC, therefore, offers an important new tool for estimating regional-scale 222Rn surface fluxes. Practical applications furthermore include calibration of process-based 222Rn soil flux models, validation of atmospheric transport models and performing regional-scale inversions, e.g. of greenhouse gases via the SPOT 222Rn-tracer method.

Energy and mass exchange and the productivity of main Siberian ecosystems (from Eddy covariance measurements). 2. carbon exchange and productivity

Direct measurements of CO2 fluxes by the eddy covariance method have demonstrated that the examined middle-taiga pine forest, raised bog, true steppe, and southern tundra along the Yenisei meridian (~90° E) are carbon sinks of different capacities according to annual output. The tundra acts as a carbon sink starting from June; forest and bog, from May; and steppe, from the end of April. In transitional seasons and winter, the ecosystems are a weak source of carbon; this commences from September in the tundra, from October in the forest and bog, and from November in the steppe. The photosynthetic productivity of forest and steppe ecosystems, amounting to 480–530 g C/(m2 year), exceeds by 2–2.5 times that of bogs and tundras, 200–220 g C/(m2 year). The relationships between the heat balance structure and CO2 exchange are shown. Possible feedback of carbon exchange between the ecosystems and atmosphere as a result of climate warming in the region are assessed.

Technical Note: Cost-efficient approaches to measure carbon dioxide (CO<sub>2</sub>) fluxes and concentrations in terrestrial and aquatic environments using mini loggers

Abstract. Fluxes of CO2 are important for our understanding of the global carbon cycle and greenhouse gas balances. Several significant CO2 fluxes in nature may still be unknown as illustrated by recent findings of high CO2 emissions from aquatic environments, previously not recognized in global carbon balances. Therefore, it is important to develop convenient and affordable ways to measure CO2 in many types of environments. At present, direct measurements of CO2 fluxes from soil or water, or CO2 concentrations in surface water, are typically labor intensive or require costly equipment. We here present an approach with measurement units based on small inexpensive CO2 loggers, originally made for indoor air quality monitoring, that were tested and adapted for field use. Measurements of soil–atmosphere and lake–atmosphere fluxes, as well as of spatiotemporal dynamics of water CO2 concentrations (expressed as the equivalent partial pressure, pCO2aq) in lakes and a stream network are provided as examples. Results from all these examples indicate that this approach can provide a cost- and labor-efficient alternative for direct measurements and monitoring of CO2 flux and pCO2aq in terrestrial and aquatic environments.

Sources and sinks of carbon dioxide in a neighborhood of Mexico City

Cities are the main contributors to the CO2 rise in the atmosphere. The CO2 released from the various emission sources is typically quantified by a bottom-up aggregation process that accounts for emission factors and fossil fuel consumption data. This approach does not consider the heterogeneity and variability of the urban emission sources, and error propagation can result in large uncertainties. These uncertainties might lead to unsound mitigation policies. Monitoring systems of greenhouse gases (GHG) based on independent methods are needed to validate the accuracy of the estimated emissions. In this context, direct measurements of CO2 fluxes that include all major and minor anthropogenic and natural sources and sinks from a specific district can be used to evaluate emission inventories. This study reports and compares CO2 fluxes measured directly using the eddy covariance (EC) method with emissions taken from the gridded local emissions inventory for the footprint covered by the EC flux system for a residential/commercial neighborhood of Mexico City. The flux measurements were conducted over 15-month period. No seasonal variability was found, but a clear diurnal pattern with morning and evening peaks in phase with the rush-hour traffic was observed. After adding contributions from human and soil respiration obtained by bottom-up approaches, and subtracting the CO2 sequestrated by vegetation calculated by applying biomass allometric equations and a growth predictive model to trees inventoried within the studied domain, results show that the current emissions inventory over-predicts 2.8 times the average daily flux measured on weekdays. Using traffic emissions data from a 2-year older inventory the difference decreased to 30%, suggesting that the traffic load for this part of the city is probably highly overestimated in the current emissions inventory. This study is expected to contribute to the verification capabilities of the GHG mitigation management of Mexico City, and to serve as a reference for other subtropical cities with similar urbanization patterns.

Spatial and temporal variability of pCO2 and CO2 efflux in seven Amazonian Rivers

Current estimates of CO2 outgassing from Amazonian rivers and streams have considerable uncertainty since they are based on limited-time surveys of pCO2 measurements along the Amazon mainstem and mouths of major tributaries, using conservative estimates of gas exchange velocities. In order to refine basin-scale CO2 efflux estimates from Amazonian rivers, we present a long time (5-year) dataset of direct measurements of CO2 fluxes, gas transfer velocities and pCO2 measurements in seven representative rivers of the lowland Amazon basin fluvial network, six non-tidal (Negro, Solimoes, Teles Pires, Cristalino, Araguaia and Javaes) and one tidal river (Caxiuana), with sizes ranging from 4th to 9th order. Surveys were conducted from January 2006 to December 2010, in a total of 389 campaigns covering all stages of their hydrographs. CO2 fluxes and gas transfer velocities (k) were measured using floating chambers and pCO2 was measured simultaneously by headspace extraction followed by gas chromatography analysis. Results show high CO2 flux rate variability among rivers and hydrograph stages, ranging from −0.8 to 15.3 μmol CO2 m−2 s−1, with unexpected negative fluxes in clear-water rivers during low waters. Non-tidal rivers showed marked seasonal CO2 flux patterns, with significantly higher exchange during high waters. Seasonality was modulated by pCO2, which was positive and strongly correlated with discharge. In these rivers k was well correlated with wind speed, which allowed the use of wind data to model k. We estimate a release of 360 ± 60 Tg C year−1 from Amazonian rivers and streams within a 1.47 million km2 quadrant in the central lowland Amazon. Extrapolating these values to the basin upstream of Obidos, results in an outgassing of 0.8 Pg C to the atmosphere each year. Our results are a step forward in achieving more accurate gas emission values for Amazonian rivers and their role in the annual carbon budget of the Amazon basin.

Inverse carbon dioxide flux estimates for the Netherlands

CO2 fluxes for the Netherlands and surroundings are estimated for the year 2008, from concentration measurements at four towers, using an inverse model. The results are compared to direct CO2flux measurements by aircraft, for 6 flight tracks over the Netherlands, flown multiple times in each season. We applied the Regional Atmospheric Mesoscale Modeling system (RAMS) coupled to a simple carbon flux scheme (including fossil fuel), which was run at 10 km resolution, and inverted with an Ensemble Kalman Filter. The domain had 6 eco‐regions, and inversions were performed for the four seasons separately. Inversion methods with pixel‐dependent and ‐independent parameters for each eco‐region were compared. The two inversion methods, in general, yield comparable flux averages for each eco‐region and season, whereas the difference from the prior flux may be large. Posterior fluxes co‐sampled along the aircraft flight tracks are usually much closer to the observations than the priors, with a comparable performance for both inversion methods, and with best performance for summer and autumn. The inversions showed more negative CO2 fluxes than the priors, though the latter are obtained from a biosphere model optimized using the Fluxnet database, containing observations from more than 200 locations worldwide. The two different crop ecotypes showed very different CO2uptakes, which was unknown from the priors. The annual‐average uptake is practically zero for the grassland class and for one of the cropland classes, whereas the other cropland class had a large net uptake, possibly because of the abundance of maize there.

Direct measurements of CO2flux in the Greenland Sea

[1] During summer 2006 eddy correlation CO2 fluxes were measured in the Greenland Sea using a novel system set‐up with two shrouded LICOR‐7500 detectors. One detector was used exclusively to determine, and allow the removal of, the bias on CO2 fluxes due to sensor motion. A recently published correction method for the CO2‐H2O cross‐correlation was applied to the data set. We show that even with shrouded sensors the data require significant correction due to this cross‐correlation. This correction adjusts the average CO2 flux by an order of magnitude from −6.7 × 10 −2 mol m −2 day −1 to −0.61 × 10 −2 mol m −2 day −1 , making the corrected fluxes comparable to those calculated using established parameterizations for transfer velocity. Citation: Lauvset, S. K., W. R. McGillis, L. Bariteau, C. W. Fairall, T. Johannessen, A. Olsen, and C. J. Zappa (2011), Direct measurements of CO2 flux in the Greenland Sea, Geophys. Res. Lett., 38, L12603, doi:10.1029/2011GL047722.

Cropland carbon fluxes in the United States: increasing geospatial resolution of inventory‐based carbon accounting

Net annual soil carbon change, fossil fuel emissions from cropland production, and cropland net primary production were estimated and spatially distributed using land cover defined by NASA's moderate resolution imaging spectroradiometer (MODIS) and by the USDA National Agricultural Statistics Service (NASS) cropland data layer (CDL). Spatially resolved estimates of net ecosystem exchange (NEE) and net ecosystem carbon balance (NECB) were developed. The purpose of generating spatial estimates of carbon fluxes, and the primary objective of this research, was to develop a method of carbon accounting that is consistent from field to national scales. NEE represents net on-site vertical fluxes of carbon. NECB represents all on-site and off-site carbon fluxes associated with crop production. Estimates of cropland NEE using moderate resolution (approximately 1 km2) land cover data were generated for the conterminous United States and compared with higher resolution (30-m) estimates of NEE and with direct measurements of CO2 flux from croplands in Illinois and Nebraska, USA. Estimates of NEE using the CDL (30-m resolution) had a higher correlation with eddy covariance flux tower estimates compared with estimates of NEE using MODIS. Estimates of NECB are primarily driven by net soil carbon change, fossil fuel emissions associated with crop production, and CO2 emissions from the application of agricultural lime. NEE and NECB for U.S. croplands were -274 and 7 Tg C/yr for 2004, respectively. Use of moderate- to high-resolution satellite-based land cover data enables improved estimates of cropland carbon dynamics.

Evaluation of Webb Correction on CO2 Flux by Eddy Covariance Technique Using Open-Path Gas Analyzer over Asphalt Surface

Webb correction is indispensable when evaluating trace gas flux by the eddy covariance technique using an open-path gas analyzer. In the present study, direct measurements of CO2 flux by the eddy covariance technique using an open-path CO2/H2O gas analyzer were carried out over an asphalt surface where CO2 and latent heat fluxes were considered almost zero. The power spectrum of CO2 density fluctuation had similar frequency structure to those of vertical wind velocity, sound virtual temperature, and water vapor density fluctuations. The cospectrum of raw CO2 flux was negatively correlated with that of sensible heat flux. The cospectrum of latent heat flux was almost zero and poorly correlated with that of raw CO2 or sensible heat fluxes. Raw CO2 flux was apparently downward from the atmosphere to the asphalt surface during this observation period. The Webb CO2 correction term for sensible heat flux was upward and larger than the raw CO2 flux, although the Webb CO2 correction term for latent heat flux was almost zero. Total latent heat and CO2 fluxes were evaluated as significant upward fluxes. This result shows that Webb correction may cause an overestimated correction to CO2 flux by the eddy covariance technique using an open-path CO2/H2O gas analyzer over surfaces, especially where small CO2 flux by soil respiration or air-sea exchange process is observed.

Primary productivity in the central equatorial Pacific (3°S 130°W) during GasEx‐2001

Measurements of chlorophyll concentration, phytoplankton productivity, and nutrient dynamics were made during the GasEx‐2001 cruise to the equatorial Pacific in February 2001. During the core measurement period of the experiment, a parcel of water was tracked over a 16‐day period in order to close the mixed layer carbon budget. Chlorophyll concentration averaged 0.16 mg m−3 and integrated mixed layer primary productivity increased from 10 to 50 mmolC m−2 d−1, concomitant with a shoaling of the thermocline. The mean f ratio (ratio of new to primary production) decreased from 0.17 at the surface to 0.04 at 60 m, and the ratio of silicate to nitrate uptake by phytoplankton was 0.8. These results are close to or slightly lower than climatological values, and are consistent with moderate productivity in an iron‐silicate co‐limited environment. Fast repetition rate (FRR) fluorometry indicated that the effects of iron limitation increased with distance from the equator, the origin of upwelled iron. Measured chlorophyll concentrations were used to calculate the attenuation of solar irradiance, which facilitated a more accurate simulation of mixed layer temperature in a separate physical study. The productivity data presented here were incorporated into a carbon budget of the mixed layer, which derived a gas transfer velocity in excellent agreement with direct CO2 flux measurements.

Direct covariance air‐sea CO2 fluxes

Direct covariance air‐sea CO2 flux measurements over the open ocean are reported. These measurements were performed during June 1998 in the North Atlantic within a significant CO2 sink. These direct estimates are in general agreement with the traditional geochemical isotope constraints. The covariance, or eddy correlation, technique directly measures the air‐sea CO2 flux over hour timescales by correlating the fluctuations of CO2 with the turbulent vertical velocity fluctuations in the atmospheric surface layer. These measurements quantify the transfer of CO2 between the atmosphere and ocean over a range of wind speeds and improve the understanding of the environmental factors controlling the flux. The relatively large flux of CO2 in the study region, together with improved analytical techniques, facilitated the measurements. The half‐hour mean wind speeds varied from 0.9 to 16.3 m s−1 over the month‐long experiment. The mean pCO2 during the study period was −85.8±16.0 μatm, and the mean covariance CO2 flux was estimated at 4.6 mol m−2 yr−1. The average observed wind speed was 7.7 m s−1. This is in close agreement with 3.9 mol m−2 yr−1, the approximate CO2 flux based on 14C parameterizations at this wind speed. At high winds, where the relationship between gas physical properties, surface processes, and air‐sea gas exchange is still elusive, direct CO2 flux measurements are crucial. The measurements for winds in excess of 11 m s−1 show a general enhancement of gas transfer velocity over previous indirect measurements, and it is believed that this enhancement can be explained by the fact that the indirect methods cannot discriminate surface process variability such as atmospheric stability, upper ocean mixing, wave age, wave breaking, or surface films.

Challenges in Linking Atmospheric CO2 Concentrations to Fluxes at Local and Regional Scales

We describe relationships between atmospheric CO2 concentration variations and CO2 source-sink distributions, at two important scales between the single plant and the whole earth: the vegetation canopy and the atmospheric planetary boundary layer. For both these scales, it is shown how knowledge of turbulence and scalar dispersion can be applied to infer CO2 source-sink distributions or fluxes from concentration measurements. At the canopy scale, the turbulent transfer of CO2 and other scalars is non-diffusive close to any point source or sink in the canopy, but diffusive at greater distances. This distinction leads to a physically tenable description of turbulent transfer, and thence to an 'inverse method' for finding the vertical profiles of sources and sinks in the canopy from measured concentration profiles. The method is tested with data from a wheat crop. At the scale of the planetary boundary layer, we consider the daily CO2 concentration drawdown (the depression of the near-surface CO2 concentration below the free-atmosphere value) of typically 20-40 ppm. This is determined by both the regionally averaged CO2 uptake at the surface and the growth of the daytime convective boundary layer (CBL). It is shown that, for a column of air which fills the CBL and is moved across the landscape by the mean wind, the net cumulative surface CO2 flux (in mol m-2) is given to a good approximation by h(t)[Cm(t) - C+]/V, where h(t) is CBL depth, Cm(t) the CO2 concentration in the CBL column in mol mol-1, C+ the concentration above the CBL, V the molar volume and time t is measured from the time at which Cm = C+ in the morning, typically about 0800 hours local time. The resulting CO2 flux estimates are regionally averaged over the trajectory followed by the column. This 'CBL budget method' for inferring surface fluxes is compared with direct measurements of CO2 fluxes, with satisfactory results. The technique has application to scalars other than CO2.