In Situ Measurement of Dissolved Methane and Carbon Dioxide in Freshwater Ecosystems by Off-Axis Integrated Cavity Output Spectroscopy

Transcription

1 YNCenter Video Conference In Situ Measurement of Dissolved Methane and Carbon Dioxide in Freshwater Ecosystems by Off-Axis Integrated Cavity Output Spectroscopy Gonzalez-Valencia et al., 2014 Reporter: Chen Zheng

2 Outline Background Materials and Methods Results and Discussion Conclusion 2

3 Background CH 4 and CO 2 emissions from lakes and reservoirs depend on numerous processes involved in biogeochemical carbon cycling. C CH4 is important to understand the complexity of CH 4 cycling in freshwater ecosystems and allows the quantification of total diffusive CH 4 emissions to the atmosphere or can be used as a pollution indicator. C CO2, combined with other parameters, gives valuable information about bioprocesses occurring in an ecosystem. 3

4 Background Current measuring equipment and its shortcomings: Standard gas chromatography, mass spectrometry, laser based detectors. (Relatively expensive, huge operation facilities, long response time, not real-time test data.) Objective: Develop and deploy a low-cost, real-time measurement and portable in situ detector for the combined measurement of C CH4 and C CO2. 4

5 Materials and Methods Detector and Prototype Detector: UGGA (1Hz; Simultaneous detection of methane, carbon dioxide and water vapor.) Prototype: It is based on the transfer of the dissolved gases from water to a gas phase, followed by the analysis of the gas phase by the UGGA. Gas line Water line Figure 1. Prototype for dissolved CH 4 and CO 2 concentration measurements. 5

6 Materials and Methods 1 H-ICOS: (Henry's law) Cw Cg Vg Cg H ' Vl Vl (1) H' RT K H 1 1 exp T (2) 6

7 Materials and Methods 2 M-ICOS: (Fick's second law) dm dt 1000 K A M Cw Cg H' (3) CgQg 1000 K A M Cw Cg H' (4) CgQg Cw 1000 K A M Cg H' Qg Cg 1000 K A M 1 H' Cg (5) 7

8 Materials and Methods An important issue with the M-ICOS method is the delay time (td) and the response time (tr) of the system. A continuous flow stirred tank reactor (CSTR) model of hydraulic residence time describes well the hydraulic behavior of the system. (eq 6). Cwm Cw 1 exp t tr (6) Cw dcwm dt t r Cwm (7) Cw, t dcwm, t dt t d tr Cwm, t t d (8) 8

9 Materials and Methods Laboratory testing The writer testes the precision and linearity of the UGGA by injecting several CH 4 and CO 2 standards. Then he establishes the peak response of the UGGA to several volumes and CH 4 and CO 2 concentrations injected in the gas line. Prepare water samples of known concentration and establish the time necessary to reach equilibrium between the water sample and the headspace.test the H-ICOS and M-ICOS method by comparing the measured Cw to the theoretical concentrations. The td and tr of the M-ICOS method were established by switching water containing CH 4 and CO 2 to degassed water using a 3-way valve. 9

10 Materials and Methods Field-Testing Four test sites: (i)lake Guadalupe; (ii) Lake Llano; (iii) Lake Goldstream; (iv) Lake Otto. In all lakes, Cw profiles were determined by the M-ICOS method after determination of α with the H-ICOS method. The profile procedure that best worked was as follows: the probe was maintained a few centimeters below the water surface for about 30 s; then the probe was lowered slowly and steadily by hand to the bottom of the lake, maintained for an additional 30 s. The diving speed was about 0.6 m min 1. The Cw data were corrected according to eq 8 before being interpreted. 10

11 Results and Discussion Laboratory Testing Figure 2. (A) Example of peak response of the UGGA to triplicate injection of 5 ml nitrogen containing 2 ppm of CH 4 and the minimum injected CH 4 quantity that was distinguishable from the background (inner Figure); (B) Integrated area of the peak response to increasing CH 4 (white dots) and CO 2 (black dots) quantities. 11

12 Results and Discussion Figure 3. Normalized CH 4 (white dots) and CO 2 (black dots) headspace concentrations in the sampling syringe for several shaking times. 12

13 Results and Discussion The H-ICOS and M-ICOS method were tested in the laboratory, under simulated field-conditions; that is, using the prototype and sampling water prepared in the STR, which contained several values of Cw. Figure 4. Measured C CH4 by H-ICOS (white dots) and by M-ICOS (black dots) vs. theoretical C CH4 concentration prepared in a stirred tank reactor. 13

14 Results and Discussion Figure 5. Example of measured CH 4 concentrations by the M-ICOS after sudden changes in C CH4 ; decreasing concentration gradient (white dots) and increasing concentration gradient (black dots). 14

15 Results and Discussion Field-Testing Coefficient of variation in Field-Testing Coefficient of variation in Laboratory-Testing C CH4 3.40% 2.15% C CO2 2.59% 1.45% Table 1.Standard error of the mean measured by H-ICOS at the same site of the Lake Guadalupe. Variation range Coefficient of variation C CH % C CO % Table 2. α measured at several depths of each lake. 15

16 Results and Discussion Figure 6. Example of the field response of the M-ICOS method to sudden change in water concentrations in Lake Guadalupe (black dots) and C CH4 calculated from eq 8 (white dots). 16

17 Results and Discussion Figure 7. Example of triplicate C CH4 (A) and C CO2 (B) profiles measured in Lake Guadalupe and absolute relative error between triplicate measurements of C CH4 (C). 17

18 18

19 Conclusion The M-ICOS prototype and method, combined with the H-ICOS method for field calibration, allowed the determination of Cw with a frequency of 1 Hz and with a MDL of mol L 1 for C CH4 and of mol L 1 for C CO2. These MDL are significantly lower than the minimum concentrations in lakes, as reported in the literature. The method is easily operable by a single person from a small boat, and the small size of the suction probe allows the determination of dissolved gases with a minimized impact on shallow freshwater ecosystems. 19

20 YN Center Video Conference Thanks for your attention! 20