Interactive comment on “ Air-Sea Fluxes of CO 2 and CH 4 from the Penlee Point Atmospheric Observatory on the South West Coast of the UK ”

The manuscript focuses on the air-sea flux of CO2 and CH4 using EC data. In particularly marine EC data of CH4 are previously mainly non-existent making the work highly interesting and worth publishing. There are, however, some major problems needed to be addressed before publication. The manuscript is very long, includes many different components and would benefit from being significantly shortened. The new and unique aspect of the manuscript is the marine CH4 fluxes and the paper would benefit from a much narrower focus. The CO2 analysis gives some numbers of the CO2 exchange, but as the water-side measurements are very limited and their representativity for the EC data highly questionable, this aspect of the paper does not bring much additional information compared to existing literature on air-sea CO2 exchange. The authors con-

in shallow waters. Furthermore, k W derived from the open ocean may not always be applicable to shallow waters, where waves shoal and break more frequently, and tidal-flow and currents could become more important (e.g. Upstill-Goddard 2006).
Monitoring of CO 2 fluxes in such dynamic and variable environments necessitates a continuous, high temporal resolution methodology (Edson et al. 2008), such as the eddy covariance (EC) technique.
Based on seawater CH 4 concentrations and global modeling, CH 4 emission from the open ocean to the atmosphere has 15 been estimated to be 0.4-18 Tg yr -1 , an uncertain but probably small term in the global CH 4 budget (Bates et al. 1996;Bange et al. 1994;Lelieveld et al. 1998). In certain regions such as the Arctic, however, ice melt can expose underlying CH 4 -rich waters (e.g. Shakhova et al 2010; Kitidis et al. 2010). Enhanced mixing ratios of CH 4 were measured on low elevation flights over regions of fractional ice cover and open leads in the Arctic, suggesting a large surface source (Kort et al. 2012). On a per area basis, shelf seas, rivers, and estuaries tend to have much greater CH 4 emissions than the open ocean due to benthic 20 methanogenesis Upstill-Godard et al. 2000;Middelburg et al. 2002). Global CH 4 emissions from coastal regions are poorly quantified and may be influenced by processes such as riverine outflow and tidal circulations. In shallow waters, ebullition (bubbles rising from the sediment) represents an additional pathway for CH 4 transfer (Dimitrov 2002;Kitidis et al. 2007). Some bubbles are not fully dissolved in seawater before surfacing and this transfer to the atmosphere is not accounted for in bulk flux calculations based on aqueous CH 4 concentrations.

25
Direct air-sea flux measurements would help to constrain CH 4 cycling and could also improve our understanding of the physical processes that drive gas transfer. Thus far, estimates of k W from sparingly soluble gases such as CO 2 and 3 He/SF 6 (e.g. Sweeney et al. 2007;McGillis et al. 2001;Nightingale et al. 2010) increase more rapidly with wind speed than those derived from the more soluble dimethyl sulfide (e.g. Huebert et al. 2004;Yang et al. 2011;Bell et al. 2013). This divergence may be due to bubble-mediated gas exchange resulting from breaking waves (Blomquist et al. 2006). CH 4 is much less soluble than CO 2 in seawater and should thus be transferred even more efficiently by bubbles.
We measured air-sea CO 2 , CH 4 , momentum, and sensible heat fluxes by the EC method at the Penlee Point Atmospheric Observatory (PPAO) during three periods at three sampling heights: May-June 2014 (~15 m above mean sea level, AMSL), June-July 2014 (~27 m), and April-June 2015 (~18 m). The influences of sampling height and wind direction on fluxes 5 are examined in Section 3.2. To evaluate how representative our measurements are of air-sea transfer, EC fluxes of momentum and sensible heat are compared to open-ocean bulk formulae based on mean wind speed and air/sea temperatures (Section 3.3).
We illustrate wind direction and diel variations in atmospheric CO 2 and CH 4 mixing ratios (Section 4.1). Marine CH 4 emission has not been quantified previously by EC and here we estimate the detection limit of this measurement (Section 4.2). Focusing on the open water wind sector, we elucidate the drivers for the variability in CO 2 and CH 4 fluxes (Sections 4.3 and 4.4).

Environmental Setting
The Penlee Point Atmospheric Observatory (50° 19.08' N, 4° 11.35' W; http://www.westernchannelobservatory.org.uk/penlee/) was established in May 2014 by the Plymouth Marine Laboratory (PML) on the South West coast of the United Kingdom for 15 long-term observations of air-sea exchange and atmospheric chemistry. PPAO is in close proximity to two nearby long-term marine stations that form the Western Channel Observatory (http://www.westernchannelobservatory.org.uk). Meteorological variables (wind, temperature, humidity, pressure), sea surface temperature (SST), salinity, chlorophyll, oxygen and dissolved organic matter are measured continuously from buoys stationed at L4 (50° 15.0' N, 4° 13.0' W) and E1 (50° 02.6' N, 4° 22.5' W), which are about 6 and 18 km south of PPAO. Seawater pCO 2 is measured on weekly cruises to the L4 station and biweekly 20 cruises to the E1 station (Kitidis et al. 2012).
PPAO is situated on an exposed headland on the western edge of the Plymouth Sound, which is primarily fed by the Tamar estuary from the northwest and is open to the Atlantic Ocean to the southwest (Figure 1). South/southwest of PPAO, the water depths increase steadily to ~8, 15, 22, and 24 m (relative to mean sea level) at horizontal distances of 100, 300, 1000, and 1300 m (www.channelcoast.org). Northeasterly wind comes over the Plymouth Sound to PPAO and is limited to a fetch of 25 about 5 km. Air from the southeast is affected by pollution from the European Continent as well as shipping emissions (Yang et al. 2016). In the southwest direction, the wind fetch is up to thousands of km and the wind speed sometimes exceeds 20 m s -1 .
This brings in air that has much less anthropogenic influence and is more representative of the background Atlantic atmosphere (see Section 4.1). 4 The stone PPAO building (length, width, height of 3.5, 3.5, 3.0 m) is approximately 11 m above mean sea level, mains powered, vehicle-accessible, and uses line-of-sight radioethernet to communicate with PML (6 km to the north/northeast). A small strip of land and a narrow, rocky intertidal zone separate the building from the sea. Southwest and northeast of PPAO, the horizontal distance to the water's edge is 30-60 m, depending on the tide. Southeast of PPAO, the distance to water is greater (about 70-90 m) due to an exposed rocky outcrop. The local tidal amplitudes (semi-diurnal) are ~5 m during spring tide and ~2 5 m during neap tide. The intertidal zone is only sparsely covered by macroalgae (<10% by area), likely due to frequent exposure to large waves.

Turbulent Flux Instrumentation
During

CO 2 and CH 4 Measurements
Atmospheric mixing ratios of CO 2 and CH 4 were measured by a Picarro cavity-ringdown analyzer (G2311-f) at a frequency of 10 Hz ("flux mode"). The inlet to this analyzer was mounted ~30 cm below the center volume of the Windmaster Pro anemometer.
An external dry vacuum pump drew sample air via a ~18 m long, 3/8'' OD Teflon perfluoroalkoxy (PFA) tubing at a flow rate of initially ~30 L min -1 . The pump performance deteriorated over time due to constant exposure to sea salt. A high performance particulate arrestance (HEPA) filter was installed immediately upstream of the pump in late 2014, which resulted in a ~15 L min -1 reduction of the main flow. The Picarro instrument subsampled from the main flow via a ~2 m long, ¼'' OD Teflon PFA tubing at a rate of ~5 L min -1 . Airflow was fully turbulent throughout the inlet.
The presence of water vapor (H 2 O) degrades the measurements of CO 2 and CH 4 via dilution, spectral interference and 5 line broadening (Rella, 2010). Miller et al. (2010) and Blomquist et al. (2014) found that ambient variability in H 2 O mixing ratio causes significant bias to the EC measurements of air-sea CO 2 flux. We followed the recommendation of Blomquist et al. (2014) and dried the sampled air using a high throughput dryer (Nafion PD-200T-24M). H 2 O efficiently permeates through the Nafion membrane while CO 2 and CH 4 essentially do not. Set up in counter-flow mode (reflux configuration), the dryer utilizes the lower pressure of the Picarro exhaust air to dry the sample air. The ambient H 2 O mixing ratio is typically on the order of 1% at 10 PPAO. With the dryer inline the measured H 2 O mixing ratio was reduced by 5 to 10-fold. The Picarro instrument reports mixing ratios of CO 2 and CH 4 in sample air based on precisely controlled cavity temperature and pressure. An internal, pointby-point correction by the instrument for residual humidity yields the "dry mixing ratios" (C CO2 and C CH4 ), which we use for flux computations. Air density fluctuations (i.e. Webb et al. 1980) should thus not affect our measurements. Tuned by the manufacturer prior to our first use, we checked the instrument calibration with CO 2 and CH 4 gas standards (BOC) and 15 occasionally determined the instrument backgrounds with nitrogen gas. CO 2 and CH 4 measurements were unavailable between August 2014 and March 2015 due to faults in the Picarro instrument.

20
In the eddy covariance method, flux is determined from the correlation between the vertical wind velocity (w) and the variable of interest (x): . Here the primes indicate fluctuations from the means while the overbar denotes temporal averaging. The coastal environment near PPAO is complex and heterogeneous in both air and water phases. Shifts in airmass and wind direction result in substantial changes in air temperature and gas mixing ratios. Thus we chose a relatively short averaging interval of 10 minutes (as used by e.g. Miller et al 2010) to more easily satisfy the homogeneity/stationarity requirements for eddy covariance

25
(see Appendix A for flux quality control).
For the computations of CO 2 and CH 4 fluxes ( , ), an hourly lag correlation analysis is performed to determine the time delay between the instantaneous vertical winds and gas mixing ratio measurements. Most of the atmospheric variability in CO 2 and CH 4 is caused by horizontal transport, rather than the air-sea flux. Detrending the gas mixing ratios to remove low frequency variability improves the accuracy of the lag time determination. Between May and July 2014, a delay of 30 6 1.9±0.1 s was found between w (Windmaster Pro anemometer) and C CO2 . After the installation of the HEPA filter, the delay increased to 3.3±0.1 s. Lag times derived from w and C CH4 are much noisier due to the smaller magnitude of the CH 4 flux. We apply the lag correction determined from the w:C CO2 analysis to the CH 4 flux calculation. Ten-minute segments of CO 2 and CH 4 fluxes that pass the quality control criteria (see Appendix A) are further averaged to hourly intervals, which reduces random noise by a factor of ~N 0.5 , where N is the number of valid flux segments. Only hours with at least three 10-minute flux intervals 5 are considered for further analysis.

Evaluation of Wind Sectors
A double rotation (Tanner and Thurtell, 1969;Hyson et al. 1977) streamline correction is applied to wind data in 10-minute blocks prior to flux computation. Tilt angles between the horizontal and vertical planes from this calculation for sampling 10 heights of 15, 18, and 27 m AMSL are shown in Figure 2. During onshore airflow, the mean tilt angle is positive as air is forced upwards. The magnitude of this tilt for southwesterly wind, which blows perpendicularly across the Penlee headland and makes contact with water again to the northeast, is comparable to shipboard measurements. The tilt angle is negative in the northwest sector due to the presence of a small hill behind the observatory building in that direction. A comparison of horizontal wind speed between Penlee and the L4 buoy when the wind is from the southwest does not show, within measurement uncertainties, a 15 significant acceleration in the Penlee measurement (e.g. as might be expected when air is forced over a superstructure). Thus the hill to the northwest of the site should not have a major influence on our measurements during southwesterly conditions. A peak in tilt angle near 120°, more apparent at low sampling heights, is likely caused by the exposed rocky outcrop in that direction.
The impact of this local topography is reduced with increasing sampling height.
From the friction velocity and wind speed (U true ), we compute the drag coefficient 20 . Bin-averaged C D at the three sampling heights as a function of wind direction is shown in Figure 3. At 15 and 18 m AMSL, measured C D from about 80 to 150° are clearly elevated compared to open ocean values (which typically range between 0.5×10 -3 and 2.5×10 -3 depending on the wind speed; Edson et al. 2013). This is likely because a part of the flux footprint overlapped with the rocky outcrop in that direction, which has a greater roughness length than the surface ocean.
Likewise, high C D values between 250° and 40° are caused by land. The impact from the rocky outcrop to the southeast is no 25 longer obvious at a sampling height of 27 m AMSL, when the flux footprint shifts further away from the observatory. For winds blowing from the northeast and southwest, measured C D is lower and much closer to values expected for the open ocean.
Northeasterly winds are relatively infrequent (~8% of the time) and limited in fetch; also the airmass from that direction is affected by terrestrial pollution and ship emissions. We thus focus on the more frequent (~20% of the time) southwest wind sector (180-240°) for most of this paper. In Appendix B, we compute the theoretical flux footprints at different sampling heights 7 and during various atmospheric conditions/tidal cycles. For southwesterly winds, land influence is predicted to be only a few percent when the mast height is ≥18 m AMSL.

Verification of Momentum and Sensible Heat Transfer
Here we compare the 10-m neutral drag coefficient ( ) and sensible heat fluxes to the fairly well established 5 open-ocean bulk formulae predictions. The 10-m neutral wind speed U 10N is determined using Businger-Dyer relationships (Businger, 1988) from the wind speed and air temperature at PPAO, tidal-dependent sampling height, and SST from L4. EC sensible heat flux is derived from the sonic temperature and further corrected for humidity using the bulk latent heat flux. To avoid sheltering by Rame Head to the west and near-shore processes, we limit our C D10N observations to a narrower wind sector of 180-220°. Figure 4 shows the relationship between C D10N and U 10N from the Windmaster Pro sonic anemometer. Also shown 10 are the predicted C D10N from the COARE model version 3.5 (Edson et al. 2013) and Smith (1980  values may partly be due to remaining uncertainties in the Windmaster Pro sonic anemometer even after applying the bias correction to the w axis. Our coastal measurements show that at a tilt angle of 5º, the recommended w correction increases u * from the Windmaster Pro by 6% (and increases scalar fluxes by 14%). Relative to the R3 sonic anemometer, this reduces the 20 low bias in the Windmaster Pro u * from 9-10% to 3-4%. The remaining 3-4% bias can account for an approximate 0.1×10 -3 underestimation of C D10N by the Windmaster Pro. Figure 5 shows a comparison between the EC sensible heat flux and the bulk sensible heat flux. The latter is computed from SST from the L4 buoy (1 m depth), potential air temperature and U 10N from PPAO, and the heat transfer rate from the COARE model (Fairall et al. 2003). Measurement and prediction are not far from the 1:1 line at a sampling height of 27 m 25 AMSL (slope = 0.82; r 2 = 0.72). A perfect agreement is not expected here, as any spatial heterogeneity in SST along the 6 km between L4 and PPAO (e.g. due to the Tamar estuary outflow) or near-surface vertical gradient in seawater temperature would contribute to the discrepancy between measured and predicted sensible heat flux. At the initial sampling height of 15 m AMSL, measured sensible heat flux is often very large and shows no correlation with the bulk flux estimate, most likely due to the terrestrial influence within the flux footprint. At 18 m AMSL, a better coherence is observed but significant scatter remains, probably because the largest horizontal variability in SST is close to shore (and occupies more of the footprint at 18 m than at 27 m). Overall, our comparison of measured and predicted momentum fluxes suggests that data collected at a sampling height ≥ 18 m during southwesterly winds are within 20% in the mean of the open ocean air-sea transfer rates.

Variability in CO 2 and CH 4 Mixing Ratios
Mixing ratios of CO 2 , and CH 4 (C CO2 and C CH4 ) varied at PPAO depending on wind direction ( Figure 6). On average between May and July 2014, C CO2 and C CH4 were generally higher for winds blowing from land than for winds blowing from the sea, likely due to the much greater terrestrial emissions of these gases and also different boundary layer dynamics. Mean C CO2 and C CH4 from the southwest sector (180-240°) are similar to "well mixed" atmospheric observations from sites such as Mauna Loa 10 and Mace Head, consistent with the long atmospheric lifetime of these gases. Mean diel cycles in C CO2 and C CH4 between May and July 2014 during onshore (110-240°) and offshore (300-60°) wind flows are shown in Figure 7. C CO2 and C CH4 for onshore winds show little diel variability, consistent with the relatively small air-sea CO 2 and CH 4 fluxes (on a per area basis). C CO2 and C CH4 for offshore winds increased at night and peaked in the early morning. Nighttime wind speeds tend to be low in offshore flow, with an average of ~3 m s -1 during these months. The resultant low atmospheric turbulence favors the formation of a 15 shallow nocturnal boundary layer, which traps surface emissions. Between about 11:00 and 20:00 UTC, C CO2 was lower for offshore winds than for onshore winds, probably due to terrestrial photosynthesis. Similar diel cycles in C CO2 and C CH4 are often observed at terrestrial sites (e.g. Winderlich et al. 2014). Clear day/night differences were also apparent in the mixing ratios of oxygenated volatile organic compounds measured from the rooftop of PML (Yang et al. 2013). While not the focus of this work, it is worth noting that the elevated atmospheric CO 2 and CH 4 in the early morning will influence their air-sea fluxes in coastal 20 regions during offshore conditions.

Detection Limit of CH 4 Flux Measurement
In this section, we examine the eddy covariance flux detection limit of CH 4 and its dependence on instrumental noise as well as ambient variability. Blomquist et al. (2014) estimated an hourly CO 2 flux detection limit of ~1 mmole m -2 d -1 for a prototype 25 version of the Picarro analyzer (G-1301-f) with a Nafion dryer at a wind speed of 8 m s -1 and in a neutral atmosphere. This represents an order of magnitude improvement over previous CO 2 sensors (e.g. Licor) and is lower in magnitude than the typical air-sea CO 2 flux. Based on terrestrial eddy covariance measurements, Peltola et al. (2014) estimated the CH 4 flux detection limit using the Picarro analyzers G-1301-f and G-2311-f to be ~170 µmole m -2 d -1 for an averaging interval (T) of 30 minutes (~120 µmole m -2 d -1 at T = 60 minutes). In comparison, the expected emission of CH 4 (F CH4 ) based on dissolved CH 4 in the open ocean is generally less than 10 µmole m -2 d -1 (e.g. Forster et al. 2009).
We estimate the air-sea CH 4 flux detection limit using an empirical and a theoretical approach. First, following Spirig et al. (2005), we compute the variability in the C CH4 :w covariance at a time lag far away from the true lag (i.e. +300 s). During periods of consistent southwesterly winds, the 1 σ of this "null" CH 4 flux is 15 µmole m -2 d -1 at T = 10 minutes. The flux 5 detection limit (defined as 3 σ) should thus be 18 µmole m -2 d -1 (=3•15/6 0.5 ) for an hourly average and 4 µmole m -2 d -1 for a daily average.

10
Here τ WC and τ Cn are the integral time scales for ambient variance and noise variance. The noise term in Eq. 1 relates to φ Cn , the band-limited noise. According to the manufacturer the precision of the Picarro G2311-f is ≤ 3 ppb for CH 4 at a sampling rate of 10 Hz. The variance spectra of CH 4 during two periods of southwesterly winds are shown in Fig. 8. Variance below ~0.025 Hz largely follows the expected -5/3 slope for atmospheric transport. At frequencies above ~0.025 Hz, the Picarro shows a "pink" background noise that approximately scales to a -1/5 slope. The integrated variance from 0.025 to 5 Hz is ~1.1 ppb 2 , while the 15 average φ Cn between 1 and 5 Hz is ~0.23 ppb 2 Hz -1 . Considering noise alone (i.e. σ Ca 2 = 0), for a neutral atmosphere at a wind speed of 10 m s -1 and a sampling height of 20 m AMSL, Eq. 1 predicts an uncertainty in hourly CH 4 flux of 11 µmole m -2 d -1 (Figure 9). From the expected air-sea CH 4 flux, using similarity theory we can estimate the variability in C CH4 caused by air-sea exchange in a neutral atmosphere as 3| F CH4 |/u * (e.g. Fairall et al. 2000;Blomquist et al. 2010). For F CH4 = 2-20 µmole m -2 d -1 and u * = 0.3 m s -1 , this corresponds to a predicted variability of 0.006-0.057 ppb. Figure 9 shows that if the ambient variability 20 in C CH4 were in this range, the hourly flux uncertainty would be dominated by sensor noise.
The observed ambient variability in C CH4 tends to be more than an order of magnitude greater than is predicted from similarity theory, which is likely related to processes other than air-sea flux (e.g. spatial heterogeneity and horizontal atmospheric transport). We estimate σ Ca 2 as the second point of the autocovariance of C CH4 (the difference between the first and second points of the autocovariance equates to σ Cn 2 of ~1 ppb 2 ). At PPAO, the minimum CH 4 ambient variability during onshore 25 flow is 0.2 ppb (σ Ca 2 = 0.04 ppb 2 ), which corresponds to a predicted hourly flux uncertainty of 20 µmole m -2 d -1 (Figure 9). This is close to our empirical estimate of the CH 4 flux detection limit above. With increasing σ Ca (i.e. more variable C CH4 ), the flux uncertainty increases substantially and becomes much greater than F CH4 , while the relative importance of σ Cn 2 decreases. Thus, we expect the 10-fold greater CH 4 flux detection limit estimated by Peltola et al. (2014) to be due to the higher variability in C CH4 over land than at our marine site (for onshore winds only). Over the open ocean where σ Ca in CH 4 is likely even lower than at PPAO, the flux detection limit for CH 4 should slightly decrease.
From the analysis above, it seems that an improvement in the precision of the CH 4 instrument will only marginally reduce the uncertainty in CH 4 flux. Blomquist et al. (2010) arrived at a similar conclusion in an analysis of air-sea carbon monoxide flux. At present, the relative CH 4 flux uncertainty is best minimized by measuring in regions of large flux (i.e. high 5 seawater supersaturation and strong winds) and minimal ambient variability (i.e. spatially homogenous environment). Blomquist et al. (2010) and Yang et al. (2011) estimated the high frequency loss in dimethylsulfide flux of typically less than 5% from the same type of Nafion dryer as used in this study. Flux attenuation by the tubing itself should be negligible given the turbulent flow. Considering the other larger random uncertainties in our CO 2 and CH 4 fluxes, we present the measured fluxes without any attenuation correction in this paper.

CO 2 Flux
Air-sea CO 2 fluxes measured at sampling height of 27 m AMSL between June and July 2014 were generally small ( Figure 10).
Diurnal land-sea breezes were common and durations of onshore winds tended to be short during this period. CO 2 fluxes from the southwest (negative = into the ocean) ranged between 3 and -9 mmole m -2 d -1 (mean of -3 mmole m -2 d -1 ) during the 15 relatively windy periods on 27 June and 4 July. Seawater pCO 2 at the L4 station ranged between 326 and 345 µatm (mean of 337 µatm) from 9 June to 7 July 2014. The atmospheric CO 2 mixing ratio at L4 agrees well with Picarro measurements at PPAO during onshore flow ( Figure 10). Using the air-sea difference in partial pressure of CO 2 (ΔpCO 2 ), SST and salinity at L4, as well as wind speed at PPAO, we compute the expected air-sea CO 2 flux as k W .α.ΔpCO 2 , where α is the solubility of CO 2 and k W is the gas transfer velocity from Nightingale et al. (2000) adjusted for Schmidt number. The expected air-sea CO 2 flux of -1 to -5 20 mmole m -2 d -1 (mean of -3 mmole m -2 d -1 ) on 27 June and 4 July are of the same magnitude as our EC measurements. The mean EC CO 2 flux could not be distinguished from zero in the second half of July, consistent with the increase in seawater pCO 2 at L4.
The spring algal bloom ended abruptly in early July 2014, with chlorophyll a concentration dropping from ~3 to less than 1 mg m -3 (http://www.westernchannelobservatory.org.uk/buoys.php). The rapid warming of seawater from ~13 °C in June to ~18 °C in July aided a rapid approach towards air/sea CO 2 equilibrium by the middle of July 2014.

25
Air-to-sea CO 2 fluxes as substantial as -90 mmole m -2 d -1 were observed between April and June 2015 (sampling height of 18 m AMSL, Figure 11). For the southwest sector, the mean fluxes (standard errors) computed from the Windmaster Pro and the R3 sonic anemometers were -19.3 (±1.4) and -23.7 (±1.4) mmole m -2 d -1 during this period, respectively. The reduced mean flux from the Windmaster Pro was primarily caused by signal dropouts in this anemometer during moderate-to-heavy precipitation, which tended to coincide with high wind speeds (and greater air-sea transfer). When both sonic anemometers were functional, CO 2 fluxes computed from the Windmaster Pro and the R3 demonstrate excellent agreement (slope = 0.98, r 2 = 0.95).
Example CO 2 cospectra over about half a day from 24 April (wind speed of 8 m s -1 ) and 10 May 2015 (wind speed of 6 m s -1 ) are shown in Figure 12. The observed cospectra are fairly well described by theoretical fits for a neutral atmosphere (Kaimal 1972).
Minimal (< 10%) flux loss at high frequencies is evident, as expected. Hourly CO 2 flux (reversed in sign for clarity) during this period clearly increased with wind speed (Figure 13). Unfortunately seawater pCO 2 was not measured during this period for 5 comparison. For reference, pCO 2 measurements from L4 in May 2014 had a mean (1 σ) of 306 (26) µatm, implying a ΔpCO 2 close to -100 µatm. We compute the predicted CO 2 fluxes at SST of 12.5 °C (mean from the E1 station) and ΔpCO 2 of -50 and -100 µatm. During most of this period, EC CO 2 flux is fairly close to prediction using ΔpCO 2 = -100 µatm. Towards late May/beginning of June, the magnitude of CO 2 flux appeared to be smaller at high wind speeds. A reduction in ΔpCO 2 as occurred in 2014 could explain the declining CO 2 fluxes in 2015. large magnitude of the CO 2 flux suggest that these fluxes were likely affected by photosynthesis and respiration from land upwind of the observatory building and/or organisms living on the foreshore. As atmosphere-land exchange of CO 2 can be more 15 than an order of magnitude greater than air-sea CO 2 flux on a per area basis (e.g. Goulden et al. 1996), a relatively small terrestrial contribution to the flux footprint (>5% spatially) could significantly bias the EC measurement. At sampling heights ≥ 18 m AMSL, CO 2 fluxes show much less diel variation, as would be largely expected for air-sea transfer ( Figure 14). However, the possibility of minor influence from land/foreshore on measurements at 18 m AMSL cannot be entirely ruled out. Such local effects might explain some of the scatter in CO 2 fluxes at wind speeds below ~5 m s -1 , i.e. when the flux footprint was probably 20 closer to land.
Overall, except at the lowest sampling height, air-sea CO 2 fluxes by EC show the expected magnitude and direction in the mean. High resolution CO 2 fluxes demonstrate significant temporal variability, which is often not well captured by the weekly seawater sampling at L4. We plan to make more regular measurements of seawater pCO 2 , SST and salinity within the flux footprint in the future (e.g. as discrete water samples or using a semi-automated dissolved CH 4 measurement system on

25
Plymouth Marine Laboratory's research vessel Quest), which will enable a direct estimate of the CO 2 gas transfer velocity in a coastal environment.

CH 4 Flux
We use historical observations to assess the validity of the EC CH 4 fluxes since dissolved CH 4 was not measured during concentrations found in fresher waters. According to the compilation by Bange et al. (2006), typical seawater saturations of CH 4 range from 110-340% in the shelf waters of the North Sea, resulting in fluxes on the order of 10 µmole m -2 d -1 .
Over the three measurement periods presented here, mean EC CH 4 fluxes ranged between 16 and 30 µmole m -2 d -1 in the southwest wind sector, with peak emissions above ~50 µmole m -2 d -1 (Figures 10 and 11). As with CO 2 , during April-June 10 2015 the smaller mean CH 4 flux computed from the Windmaster Pro anemometer than from the R3 is primarily due to signal dropouts in the former during rainy, windy conditions (Table 1). The cospectra of CH 4 are noisier than those of CO 2 ( Figure 12 variable seawater CH 4 concentrations. CH 4 emissions do not obviously vary with time of day but they tend to be higher during incoming (rising) tide than during outgoing (falling) tide. In Figure 16, CH 4 fluxes from the southwest direction (April-June 2015) are plotted against hours after low water (low tide occurs at hour zero; high tide occurs near hour six). The median, 25%, and 75% percentiles within each hour bin are also shown. The largest average CH 4 emissions are observed in the first ~4 hours after low tide, while CH 4 fluxes during the falling tide are lower and less variable. Mean CH 4 fluxes were also ~50% higher 25 during spring tide (here limited to daily tidal amplitude > 4 m) than during neap tide (daily tidal amplitude < 3 m). These patterns are consistent with an incoming tidal current that pushes the CH 4 -rich surface outflow from the Tamar estuary around the Rame peninsula (Uncles et al. 2015).
To further examine the influence of the Tamar estuarine plume, a 3-dimensional hydrodynamic Finite Volume Community Ocean Model (FVCOM, Chen et al. 2003) was run for April-June 2011 with tidal forcing at the boundaries (TPXO, Egbert et al. 2010), surface wind (Met Office Unified Model, Davies et al., 2005), surface heating (NCEP Reanalysis-2, Kanamitsu et al. 2002), and river input (E-HYPE, Donnelly et al. 2012)  our CH 4 flux observations. Natural processes other than direct air-sea gas transfer (e.g. ebullition) could also contribute to the variability in CH 4 fluxes. Quantifications of the temporal/spatial seawater CH 4 distribution within the PPAO flux footprint and measurements of the pelagic/benthic cycling of CH 4 is essential to address this uncertainty. measured CH 4 emission from a Swedish lake using the EC technique. Lake CH 4 emissions range from near zero during the day to over 20 mmole m -2 d -1 at night (three orders of magnitude higher than observations at PPAO). Aircraft mixing ratio 15 measurements suggest that CH 4 emission from the partially ice-covered Arctic is 4-5 times larger than mean emission at PPAO (Kort et al (2012). Our observations and estimates of the CH 4 flux uncertainty suggest that an EC system such as the one employed here should be able to quantify emissions from those CH 4 hot spots.

20
Air-sea fluxes of CO 2 , CH 4 , momentum, and sensible heat were measured by the EC technique in 2014 and 2015 from the Penlee Point Atmospheric Observatory (PPAO) on the southwest coast of the UK. Observed momentum and sensible heat transfer from the southwest wind sector are in the mean within ±20% of the bulk transfer estimates at a sampling height of ≥ 18 m AMSL, which makes PPAO a suitable site for long-term, high temporal resolution measurements of air-sea exchange in shelf regions.
Air-sea CO 2 fluxes demonstrate a positive dependence on wind speed and a rapid decline in magnitude from late spring 25 to early summer in both 2014 and 2015, coinciding with reduced air-sea ΔpCO 2 driven by the demise of the spring algal bloom and the seasonal warming of the sea. We report the first successful EC flux measurements of CH 4 from the marine environment.
The CH 4 flux detection limit is estimated to be 20 µmole m -2 d -1 for an hourly average (4 µmole m -2 d -1 for a daily average), which is valuable information for planning future open ocean applications of this technique. Uncertainty in CH 4 fluxes is largely due to ambient variability in atmospheric CH 4 mixing ratio rather than due to instrumental noise. Observed CH 4 emissions are 30 on the order of tens of µmole m -2 d -1 , reasonable in magnitude for an estuarine influenced coastal region. CH 4 fluxes are generally higher when the wind is from the Plymouth Sound than when the wind is from the open water sector. Furthermore CH 4 emissions from the open water are greater during rising tide than during falling tide, implying a source of CH 4 in the estuarine outflow that is affected by the local tidal circulation.

Appendix A: Quality Control on Eddy Covariance Fluxes
Conservative quality control criteria computed from 10-minute flux averaging intervals are used to remove flux measurements during unfavorable conditions (Table A1). Periods of highly variable wind direction (σ >10°) and positive momentum flux are discarded on the basis of nonstationarity, which tends to occur during calm conditions or the passage of a weather front. We also reject fluxes that do not pass the statistical quality control tests for skewness and kurtosis of w and integral turbulence 10 characteristics of (Foken and Wichura, 1996;Vickers and Mahrt, 1997). Averaged valid momentum cospectra and normalized Ogives (Oncley, 1989) on 3, 5, and 10 May 2015 (R3 sonic anemometer) are shown in Figure A1. Mean wind speeds were 12, 17, and 6 m s -1 on these three days, respectively. The Ogives approached zero at 0.0017 Hz and approached one at 5 Hz, indicating that the 10-minute averaging interval captured the majority of the turbulent flux.
To minimize the impact of horizontal transport on CO 2 and CH 4 fluxes, we set thresholds defined by the ranges and 15 trends in mixing ratios (C CO2 and C CH4 ) as well as the horizontal fluxes of these gases. Following Blomquist et al. (2012Blomquist et al. ( , 2014, we compute the horizontal fluxes as and . Here u and v represent the along-stream and cross-stream wind velocities after double rotation. Large horizontal fluxes suggest excessive spatial heterogeneity/nonstationarity. For CH 4 only, we also eliminate periods when the total variance (=σ Cn 2 +σ Ca 2 ) exceeds 2 ppb 2 . Since σ Cn 2 is ~1 ppb 2 (see Section 4.2), this equates to a σ Ca threshold of (2 ppb 2 -1 ppb 2 ) 0.5 = 1 ppb and an hourly flux uncertainty of ~80 µmole m -2 d -1 (Figure 9). We note that this 20 σ Ca threshold is almost two orders of magnitude greater than the expected ambient variability in C CH4 due to air-sea flux.
Both sonic anemometers show elevated noise at frequencies above 1 Hz when the relative humidity is near 100%, likely because of rain and sea spray. For computations of momentum and heat transfer, we remove moisture related artifacts by simply discarding fluxes when the relative humidity exceeds 95%. Noise in the sonic anemometer above 1 Hz shows little correlation with C CO2 and C CH4 , such that high humidity does not noticeably affect CO 2 and CH 4 fluxes.

Appendix B: Theoretical Flux Footprint
We use a theoretical flux footprint model (Kljun et al. 2004) to evaluate the suitability of PPAO for air-sea flux measurements.
Typical values for southwesterly conditions (i.e. clean marine air) are used in the flux footprint calculations: roughness length (z 0 ) = 0.0001 m, friction velocity (u * ) = 0.20 m s -1 , and standard deviations in w (σ W ) = 0.35, 0.26, 0.18 m s -1 (to represent unstable, neutral, stable atmospheres). At a sampling height of 27 m AMSL (fully raised mast), the predicted upwind distance of maximum flux contribution (X max ) is 600-1000 m and the distance of 90% cumulative flux contribution (X 90 ) is 1500-2600 m (the greater distances correspond to increased stability). For this set up, land/foreshore southwest of the observatory contributes to only 2-3% (stable) or 3-4% (neutral/unstable) of the cumulative flux, with the greater contributions corresponding to lower tide. The majority of the flux footprint is over waters ~20 m deep. Waves are considered to be in deep water if water depth is 5 greater than half of the wavelength. They start to deviate significantly from deep-water behavior when the depth is less than about a quarter of the wavelength. At a wind speed of 10 m s -1 , fully developed wind waves have a wavelength of ~80 m. For wind speeds more than 10 m s -1 , wind waves near Penlee could be affected by depth, while swell (which tends to be longer) almost always would be. Thus PPAO should be considered a coastal, rather than a deepwater site.
At moderate-to-high wind speeds, the marine atmosphere is usually near neutral, and the flux footprint tends to be  Nightingale P.D., Malin G, Law C.S., Watson A.J., Liss P.S., Liddicoat M.I., Boutin J, Upstill-Goddard R.C.: In situ evaluation Table 1. Summary of sampling periods, mast height above observatory rooftop and above mean sea level (AMSL), and hourly eddy covariance CH 4 fluxes (µmole m -2 d -1 ) for the southwest wind sector (180-240°). CH 4 fluxes when the sampling height was 15 m AMSL are likely underestimates of air-sea transfer because a significant portion of the flux footprint was over land (Section 3). For the last period (2015), fluxes are computed from both the Windmaster Pro and R3 sonic anemometer (shown in that order). SE indicates standard error.    . Atmospheric mixing ratios of CO 2 and CH 4 as a function of wind direction. Error bars indicate two standard errors within each wind direction bin. CO 2 and CH 4 mixing ratios were generally lower for southwesterly winds (180-240°) than for northerly wind sectors. Figure 7. Mean diel cycles in the mixing ratios of CO 2 and CH 4 . Error bars indicate two standard errors within each hour bin. Diel variability for both gases is small during onshore flow (marine winds, 110-240°). Mixing ratios of CO 2 and CH 4 during offshore flow (wind from land, 300-60°) increase at night and peak in the early morning. Figure 8. Variance spectra of CH 4 on two days of southwesterly winds. Variance at frequencies above ~0.025 Hz is dominated by noise, while ambient variability accounts for most of the low frequency variance.  . Cyan shading indicates onshore winds. Fluxes are limited to the southwest wind sector only. Also shown are pCO 2 and atmospheric CO 2 mixing ratio from the L4 station. Negative CO 2 fluxes on the order of a few mmole m -2 d -1 were observed during the windy periods on 27 June and 4 July. By late July, observed CO 2 fluxes 5 were indistinguishable from zero, consistent with near saturation of seawater pCO 2 at the L4 station. CH 4 flux has a positive mean, suggesting sea-to-air emission. Figure 11. As Fig. 10, but during April-June 2015 (sampling height of 18 m AMSL). Fluxes were computed from both the Windmaster Pro and the R3 sonic anemometers. Large air-to-sea flux of CO 2 is observed during high wind speed events, while CH 4 flux is almost always positive.     Table A1. Filtering criteria (within 10-minute averaging intervals) for quality control of eddy covariance fluxes. These criteria are shown for the southwest air sector only (180°<Wind direction<240°). The right column indicates the percentage of valid flux data that data that satisfy the filtering criteria by each stage of the quality control sequence. 10 Figure A1. Mean momentum cospectra and normalized Ogives on 3, 5, and 10 May 2015 (R3 sonic anemometer). Mean wind speeds were 12, 17, and 6 m s -1 on these three days, respectively.