The Method Estimating Daytime CO2 Flux Between Intertidal Sediment and the Atmosphere

The daytime variation in CO2 flux between the intertidal sediment and the atmosphere was great and impacted by environmental elements. This paper analyzed the daytime variation in CO2 flux between the intertidal sediment and the atmosphere and concluded its different rules during the ebb tide and flood tide. During the ebb tide, the CO2 flux rose as the tide ebbed and its rate of change was different when the redox potential changed (0.13 μmol m s per 100 cm height of the tide at Eh ≤ 300 mv, 0.15 μmol m s per 100 cm height of the tide at 300 mv < Eh < 500 mv and 0.07 μmol m s per 100 cm height of the tide at Eh ≥ 500 mv). During the flood tide, the CO2 flux maintained the largest of the day and almost unchanged with the increase of water level. The average CO2 flux in the flood tide increased with the increase of the redox potential at Eh ≤ 300 mv and 300 mv < Eh < 500 mv, but their linear regressions between the average CO2 flux and the redox potential were different. Compared with the average CO2 flux at Eh ≤ 300 mv and 300 mv < Eh < 500 mv, the average CO2 flux was no longer related to the redox potential but related to temperature at Eh ≥ 500 mv. According to these rules, the daytime CO2 flux can be calculated based on limited measurements well and truly. The fitting straight line equation between the estimated and measured CO2 flux was y= 0.9353x+0.0872 (R = 0.75, P < 0.01).


Introduction
Since global warming become the hot topic, a series of research projects have been trying to give a precise CO 2 flux between various ecosystems and the atmosphere.The intertidal zone is the junction of land and sea.It is also a special and important type of wetland.The biogeochemistry environment is complex and unconstant.These properties lead to distinct CO 2 exchange mode between intertidal zone and the atmosphere.According to statistics, there are 2.17×104 km 2 of the intertidal zone in China (Lei & Zhang, 2005).Although we need a relatively independent CO 2 flux research on capacious intertidal zone, the observation on the intertidal zone with static chamber and eddy covariance method was limited (Magenheimer, Moore, Chmura, & Daoust, 1996;Yang et al., 2006;Wang et al., 2007).Mo et al. (2005) estimated the soil respiration based on soil temperature in forest ecosystems; the estimation results were in good agreement with the measured results.Compared with the quantitative estimate of soil respiration in other ecosystems (Davidson, Belk, & Boone, 1998), there has been no similar research on intertidal zone so far.How to give a precise calculation of CO 2 flux between intertidal zone and the atmosphere?This paper intends to give a method estimating CO 2 flux between intertidal zone and the atmosphere.The method was based on the daytime course of CO 2 flux between intertidal sediments and the atmosphere.

Study Site
This paper selected the Beach (36°05.327'N,120°27.675'E) of the Shilaoren Scenic Area, Qingdao, China (Figure 1) as the study site.The study site is on the west side of Shilaoren statue.It located between Gaodijiao and Fushanjiao.The length of the beach is about 2.5 kilometers and beach area is about 0.55 km 2 .The average slope of the beach is 1.5%.The observation site is ~0.7 m above the sea level.
Figure 1.Location of the sample site (★) (Yang et al., 2009;Zhang et al., 2009) The salinity of the offshore seawater ranged from 26 to 28, because of the impact of the rivers at sourthern Laoshan (about 20 km away).The average temperature of the surface water is 13.5 °C.The observation site is 5 km away from the Maidao Marine Environmental Monitoring Station, Qingdao (www.nmfc.gov.cn).We collected the tide data from this station.The annual mean wave height is 0.7 m.The tide is semi-diurnal.The average high tide is 3.85 m.The average low tide is 1.08 m.This area is north temperate continental monsoon climate.We got the precipitation data from the Coastal Zone Program of the Laoshan Mountain Scenic Area, Qingdao.
We collected sediment samples from the observation site and measured the particle sizes of the sediments by the Cilas940L laser particle size analyzer with the range of 0.3-2000 μm (made in French).The results were shown in Table 1.The particles less than 0.85 mm were 99.99% in the sediments.The largest sediment was less than 8 mm.The median value of the sediment particle sizes was 0.23 mm.And the averages of the gravels were 0.5 mm in diameter.

Methods
We used a LI-8100 Automated Soil CO 2 Flux System (Licor, Lincoln, Nebraska, USA) to measure CO 2 fluxes between the sediment and the atmosphere.The LI-8100 chamber collar is 83.7 cm 2 .LI-8100 determined the CO 2 flux based on the diversification of the CO 2 concentration in the chamber.LI-8100 measured CO 2 concentration with a non-dispersive infrared detector.The range of the non-dispersive infrared detector was 0 to 3000 ppm.The accuracy was ±1.5%.In half an hour before the beginning of the observation, the collar for the chamber of LI-8100 was inserted into the sediment at 2.4 cm.Each measurement lasted 7 min.The interval between two consecutive measurements was 1min for a full exchange of gases (reference to LI-8100 specification).We took CO 2 concentration in the atmosphere 377 ppm and calculated the CO 2 flux with the software provided by Licor Corporation.
The LI-8100 system had a temperature probe.The accuracy of the temperature probe was ± 0.5 °C.When observing, the temperature probe was inserted into the sediment at 1 cm.At the same time, the LI-8100 system could also measure the atmospheric CO 2 partial pressure at the detector height of 16 cm above the sediment surface.
We used a FJA-16 polarization analyzer (Institute of Soil Science, Nanjing, China) to measure the redox potentials (Eh) of the sediment.The accuracy of FJA-16 was ± 10 mv.When observing, the electrodes of the FJA-16 polarization analyzer together with the chamber collar were inserted into the sediment at 1 cm.Each measurement took 2.5 min.Bohai Sea Korea Japan Yellow Sea China SPSS 13.0 was used for the statistical analyses of the data.
We selected 19 days from April 24 to July 26, 2006.In these days, there was no rain, the value of wind speed was less than 10 m/s, and there was a tidal cycle during the daytime.We studied the daytime course of the CO 2 flux between the sediment and the atmosphere in situ as follows: when the tide ebbed and the site was clear of water, we started the observation; when the tide rose and the site would be submerged we ended the observation; we measured the CO 2 flux from the sediment to the atmosphere continuously during the daytime.

The Variation in CO 2 Flux During a Daytime Tidal Cycle
We compared the data from different days with minimum anomaly analysis to conclude the course of CO 2 flux in a rounded daytime tidal cycle.The increments of CO 2 flux (subtracting the initial CO 2 flux from the CO 2 flux at a time) were plotted against the height of the tide in Figure 2. Figure 2 shows that the course of the increments of the CO 2 flux in the ebb tide was different from the course in the flood tide.In the ebb tide, the increments of the CO 2 flux increased as the water level fell and there was a linear relationship between water level and the CO 2 flux; in the flood tide, the increments of the CO 2 flux reached the maximum and no longer changed as the height of the tide changed.The variation in the increments of CO 2 flux determined the variation in CO 2 flux.So when the tide was on the ebb, the CO 2 flux increased as the water level fell, this increase continued until the tide reached its lowest point (which was also the beginning of the flood tide) and the CO 2 flux reached the maximum at the same time.Compared with the variation of the CO 2 flux in the ebb tide, the CO 2 flux maintained the largest value of the same day with no fundamental changes in the flood tide except slight decline at the end of the flood tide; in addition, the CO 2 flux during the flood tide depended on the CO 2 flux at the lowest ebb.Thus, the main period of CO 2 release in a day was during the flood tide.Because the tide is normal semi-diurnal, CO 2 released in the flood tide was about 60% of the CO 2 released in the wounded tidal cycle based on the average CO 2 flux in the ebb tide and flood tide.

The Impact of the Redox Potential on the Variation in CO 2 Flux
The process that sediment releases CO 2 to the atmosphere should be similar to soil respiration, which is mainly composed of oxidation of organic matter by soil microbe and plants root respiration, meanwhile, a minute part of which should be attributed to the respiration of soil animals and chemical oxidation of organic matter (Li, Lv, & Yang, 2002).The oxidation of organic matter, the main process releasing CO 2 , releases less CO 2 when the redox potential reduces (Tian, 2005).At different redox potential (Eh), the increments of CO 2 flux in the ebb tide were plotted against the height of tide in Figure 3 and the average CO 2 flux in the flood tide were plotted against redox potential in Figure 4.
Figure 3 shows that there were different rates of change of the increment in CO 2 flux with changes in redox potential in the ebb tide.At Eh ≤ 300 mv, the rate was 0.13 μmol m -2 s -1 per 100 cm height of the tide, and it reached 0.15 μmol m -2 s -1 per 100 cm height of the tide at 300 mv < Eh < 500 mv, however, it decreased and was only 0.07 μmol m -2 s -1 per 100 cm height of the tide at 500 mv ≤ Eh.The variations in the increment of CO 2 flux at the tidal height of 0cm were similar to the rates of change of the increment in CO 2 flux.The increment of CO 2 flux at the tidal height of 0 cm was 0.2654 μmol m -2 s -1 as the redox potential was lower than 300 mv, and it increased to 0.328 μmol m -2 s -1 when the redox potential ranged from 300 to 500 mv, however, it decreased to 0.1899 μmol m -2 s -1 as the redox potential was greater than 500 mv.When the sediment converted from Eh ≤ 300 mv to 300 mv < Eh < 500 mv, the increase in the rate of change of the increment of CO 2 flux and the increment of CO 2 flux at the tidal height of 0cm should be attributed to the intensification in the oxidation of organic matter.Both the rate of change in the increment of CO 2 flux and the increment of CO 2 flux at the tidal height of 0cm decreased because of the limitation of organic matter in the sediment when the redox potential was greater than 500 mv.
Redox potential not only affected the variation of CO 2 flux in the ebb tide, but also affected average CO 2 flux in the flood tide.It is shown in Figure 4 that the average CO 2 flux in the flood tide increased as the redox potential rose and there was different relevance between average CO 2 flux in the flood tide and redox potential with changes in redox potential.Plotting the average CO 2 flux against redox potential gave R 2 (0.9167) at Eh ≤ 300 mv, a lower R 2 (0.8502) at 300 mv < Eh < 500 mv and there was no significant linear regression between the average CO 2 flux and redox potential at 500 mv ≤ Eh.The change in the linear regression between the average CO 2 flux and redox potential means that the main controlling factor of CO 2 flux in the flood tide had changed when the redox potential rose.In a word, the redox potential was the main controlling factor of CO 2 flux in the flood tide when the redox potential was low.

The Impact of the Temperature on the Variation of CO 2 Flux
Temperature is one of the main influential factors in the past research on soil respiration and CO 2 flux between sediment and atmosphere in the coastal salt marsh (Hirota et al., 2007;Blanke, 1996;Luo et al., 2001;Bekku et al., 2003;Smith, 2003;Cao et al., 2004;Wu et al., 2006;Fang, & Moncrieff, 2001).In the ebb tide, however, the correlation coefficient between the temperature and the rates of change of the increment of CO 2 flux was -0.093 (p=0.751,n=14), which was lower than the correlation coefficient between the redox potential and the rates of change of the increment of CO 2 flux (0.537, p=0.048, n=14).Lower correlation coefficient means that the temperature didn't impact the rates of change of the increment of CO 2 flux as significantly as the redox potential in the ebb tide.Although the impact of temperature on CO 2 flux is not significant in the ebb tide, the impact of temperature on CO 2 flux should be considered in the flood tide, especially 500 mv ≤ Eh.Table 2 shows the Pearson correlation coefficient between the average CO 2 flux and temperature in different redox potential in the flood tide.As the redox potential rose, the impact of temperature on the average CO 2 flux became more significant.The correlation coefficient between two variables was 0.366 (p=0.419,n=7) at Eh ≤ 300 mv, and it reached 0.666 (p=0.334,n=4), 0.843 (p=0.017,n=8) at 300 mv < Eh < 500 mv and 500 mv ≤ Eh, respectively.Although the correlation coefficient between temperature and the average CO 2 flux was great at 300 mv < Eh < 500 mv, the correlation coefficients (0.922, p=0.078, n=4) between the redox potential and the average CO 2 flux was greater.Compared with the redox potential, temperature was not the main influential factor on the average CO 2 flux when the redox potential ranged from 300 to 500 mv.Only when the redox potential was greater than 500 mv, temperature was the main controlling factor of the average CO 2 flux in the flood tide, and the relationship of two variables was shown in Figure 5.According to the rule of the variation in the CO 2 flux, we can estimate the CO 2 flux in the wounded tide cycle based on the CO 2 flux in the ebb or flood tide (Figure 6).R T =R ebb ×t ebb +R flood ×t flood (1) Where R T is the total CO 2 flux from the intertidal sediment to the atmosphere in the daytime, R ebb is the average CO 2 flux in the ebb tide, R flood is the average CO 2 flux in the flood tide, t ebb is the time span (h) of the ebb tide after the sample site was exposed to the atmosphere and t flood is the time span (h) of the flood tide before the sample was submerged in the water.
There was a linear correlation between CO 2 flux and the height of the tide in the ebb tide (Figure 4), so ) Where H 1 is the height of tide (cm) at a point of time in the ebb tide, R 1 is the CO 2 flux (μmol m -2 s -1 ) at H 1 , H max is the height of tide (cm) when the sample point was submerged in the water, H min is the least height of tide (cm) in the daytime and a is the rate of change of CO 2 flux against the height of tide, R max and R min are corresponding CO 2 flux at H min and H max , a is the rate of change of the increment in CO 2 flux; Combining (1), ( 2), ( 3), ( 4), ( 5) and ( 6) gives: R T =0.5a (H max +H min -2H 1 ) t ebb +R 1 t ebb +Ct flood (7) Where C is the average CO 2 flux in the flood tide.
(2) If we got some data of CO 2 flux in the flood tide, then Combining ( 2), ( 3), ( 5), ( 6) and ( 8  Above equation is not a simple linear fitting equation, but a composite equation based on the rule of the variation in the CO 2 flux.We estimated the CO 2 flux with the equation and plotted the estimations against the measured CO 2 fluxes in Figure 7 to test the reliability of the equation.Figure 7 shows that the estimated fluxes were all but in the 90% confidence interval and most of the estimated fluxes were in the 68.26% confidence interval.On the other hand, the linear regression between the estimated and measured fluxes was y = 0.9353x+0.0872(R 2 = 0.75).Thus it can be seen that the Equations ( 7) and ( 9) could well estimate the daytime average CO 2 flux.

Conclusions
Based on the analysis performed, the following conclusion can be shown: (1) The CO 2 flux increased as the water level fell during the ebb tide, while it maintained the largest of the same day with no fundamental changes during the flood tide.
(2) During the ebb tide, the CO 2 flux rose as the tide ebbed and its rate of change was different when the redox potential changed.The rate of change of the CO 2 flux was 0.13 μmol m -2 s -1 per 100 cm height of the tide at Eh ≤ 300 mv, while it was 0.15 μmol m -2 s -1 per 100 cm height of the tide at 300 mv < Eh < 500 mv and 0.07 μmol m -2 s -1 per 100 cm height of the tide at Eh ≥ 500 mv respectively.
(3) The average CO 2 flux in the flood tide increased with the increase of the redox potential at Eh ≤ 300 mv and 300 mv < Eh < 500 mv, while it was no longer related to the redox potential but related to temperature at Eh ≥ 500 mv.
(4) According to the rules on the variation in CO 2 flux, the daytime CO 2 flux can be calculated based on limited measurements well and truly.The fitting straight line equation between the estimated and measured CO 2 flux was y = 0.9353x+0.0872(R 2 = 0.75, P < 0.01).) at the beginning of the ebb tide in the middle of the ebb tide at the end of the ebb tide at the beginning of the flood tide in the middle of the flood tide at the end of the flood tide

Figure 2 .
Figure 2. △flux plotted against the height of the tide, (a) for the ebb tide, (b) for the flood tide

Figure 3 .
Figure 3. CO 2 flux plotted against the height of the tide for different redox potential (Eh) values

Figure 4 .
Figure 4. Average CO 2 flux at the floodtime plotted against Eh

Figure 5 .
Figure 5. Average CO 2 flux at the floodtime plotted against Eh at 500 mv ≤ E

Figure 6 .
Figure 6.The sketch map for the daytime course of CO 2 flux between the sediment and the atmospher, R is the CO 2 flux, H is the height of the tide, a and b are the constants If we got some data of CO 2 flux in the ebb tide, because R=aH+b in the ebb tide, it should be R

Figure 7 .
Figure 7. Relationship between measured and estimated average CO 2 flux

Table . 1
Particle size distribution