Methane and carbon dioxide emissions from closed landfill in Taiwan

(1)

Methane and carbon dioxide emissions from closed landﬁll in Taiwan

I-Chu Chen, Ullas Hegde, Cheng-Hsiung Chang, Shang-Shyng Yang

*

Institute of Microbiology and Biochemistry, and Department of Biochemical Science and Technology, National Taiwan University, Taipei 10617, Taiwan Received 29 March 2007; received in revised form 16 August 2007; accepted 16 August 2007

Available online 1 October 2007

Abstract

The atmospheric concentrations and emission rates of CH4and CO2were studied at three sites of the Fu-Der-Kan closed landﬁll and

after as the multi-use recreational park in northern Taiwan. Atmospheric CH4and CO2concentrations of closed landﬁll were 1.7–4.6 and

324–409 ppm, respectively. CH4and CO2emission rates ranged from 8.8 to 163 mg m2h1and from 495 to 1531 mg m2h1,

respec-tively. Diurnal variation was noted with higher values at night than those in daytime. After creation of the park, atmospheric CH4and

CO2concentrations were 1.8–3.1 and 332–441 ppm, respectively. CH4and CO2emission rates ranged from1.1 to 2.3 mg m2h1and

from135 to 301 mg m2_h1_{, respectively. There were no notable diurnal variations in either atmospheric concentrations or emission}

rates.

Keywords: Landﬁll; Greenhouse gas; FTIR; Flux; Diurnal variation

1. Introduction

Landﬁlls are accounting for about 10–19% of annual global CH4 emissions (Kumar et al., 2004; USEPA,

2006). The global warming potential of CH4 is 23 times higher than that of CO2(USEPA, 2002). The atmospheric CH4concentration has been increasing at a rate of 0.2–

1% y1(Simpson et al., 2006). On a global scale,

approxi-mately 653 Tg y1of waste is landﬁlled, and annual global CH4 emissions from landﬁlls range from 16 to 223 Tg

(Bogner and Matthews, 2003; Simpson et al., 2006).

Landfilling and incineration are the major methods of waste disposal in Taiwan. 89% of waste was disposed by landfills in 1989. This percentage dropped to 11.2% in 2006 as the result of increased incineration and the resource recycling policy (EPA/Taiwan, 2007). Landfills function as bioreactors due to the controlled and managed burial of biodegradable organic materials. These organic materials decompose via a complex series of microbial reactions

under anaerobic conditions. Eventually, they are converted into CO2, CH4, N2O and H2O. Landfill CH4and CO2 gen-erations are controlled by landfill management, cover soil, composition, temperature, pressure, moisture content, water table level, wind induced ebullition and pH of the refuse (Bogner and Matthews, 2003; Hegde et al., 2003). Methanogenesis in landfills occurs when the pH of organic waste is between 6.8 and 7.4 and it is stimulated by increas-ing moisture content (Jang and Yang, 2001).

The Fu-Der-Kan landﬁll site was closed in 1995, and a multi-use recreational park was constructed from 1999 to 2003. The soil surface is covered with grass and small trees. Although the CH4and CO2emission rates of active land-ﬁlls are well documented (Bogner et al., 1999; Czepiel

et al., 2003; Meraz et al., 2004), emissions from closed sites

and sites which have been converted to park have not been investigated in detail.

Atmospheric CH4and CO2concentrations and emission rates from an active landﬁll in northern Taiwan were described in a previous study (Hegde et al., 2003). Two main methods are used for such measurements. Gas-type open-path Fourier-transform infrared (FTIR) spectros-copy has the advantages of good precision, rapid and

*

Corresponding author. Tel.: +886 2 23621519; fax: +886 2 23679827. E-mail address:ssyang@ntu.edu.tw(S.-S. Yang).

www.elsevier.com/locate/chemosphere Chemosphere 70 (2008) 1484–1491

(2)

simultaneous measurement of diﬀerent volatile com-pounds, and the ability to cover long-distance. Chamber techniques can be used to measure CH4and CO2emission rates from small areas, typically less than 1 m2(Yang and

Chang, 1997). The advantages include simplicity, an

appro-priate scale for concurrent measurement of controlling variables, and the ability to determine the heterogeneity of surface emissions. Chamber techniques can also be easily compared to results from other studies because these meth-ods have been widely employed in non-landﬁll settings for a variety of greenhouse gases (Rolston, 1986; Yang and

Chang, 1998, 2001; Chang and Yang, 2003). In this study,

FTIR was used successfully, along with gas chromatogra-phy (GC), to monitor atmospheric greenhouse gases. The atmospheric concentrations and emission rates of CH4 and CO2 were measured at Fu-Der-Kan closed landﬁll and after the construction of a multi-use recreational park in northern Taiwan.

2. Materials and methods 2.1. Site description

The Fu-Der-Kan landfill (2501’53.9’’N, 1203’33.8’’E) is located near Taipei City in northern Taiwan. This landfill was opened in 1985 for municipal solid waste (MSW) dumping, closed in 1995 and converted to a multi-use recre-ational park during 1999 to 2003. It covers 98 ha and 37 ha were used for landfilling of MSW. It received 3200 t of MSW every day during the operation period, with an aver-age burial depth of 22.9 m. The landfill was covered by 1– 2 m of waste landfill soil as surface material. Upon closing, the landfill was covered with another 1–1.5 m of loam-clay loam soil and reconstructed as a multi-use recreational park in 2001 (Fig. 1). The test areas cover about 32 ha of landfill-ing area includlandfill-ing areas A, B, C, E, F, G and part of J around the testing areas. Site C is younger than sites A and B. Site C had received new MSW in 1999. During 1999-2000 testing period, site A (about 8 ha) included coun-try activity area, lake, boating area and part of natural resource landscape; site B (about 12 ha) included Sun plaza, creation garden and part of natural resource landscape; site C (about 12 ha) included culture of indigenous people’s area, grass skiing field and part of natural resource land-scape. The testing area covers about 90% of landfilling except management center and wastewater treatment fac-tory. The vegetations in the sites A, B, and C were grass and small trees during the testing periods 1999–2000. The major vegetations were Centipede grass (Eremochlora ophi-uroides) and Carpet grass (Axonopus affinis) in sites A, B, and C during testing period 2006. There is a setup for CH4gas extraction system and each year recovered 3442 t of CH4, which should equivalent to 79.2 Gg of CO2, while the recovered CH4 was burned and produced 10.2 Gg of CO2only. Soil samples were collected at 0–20 cm depth in different burial locations. Randomly three soil cores in each site were collected, sieved to 2 mm and stored at 4C before

analysis. Each soil core was analyzed separately for mois-ture content, pH and organic carbon.

2.2. Gas-type open-path FTIR spectroscopy measurement FTIR spectroscopy was used to measure the concentra-tions of atmospheric greenhouse gases at a height of 197 cm above ground with a scan number of 100 for 7 min. Absorbance was read at wavenumber of 2999 cm1 for CH4and peak area from 2239 to 2393 cm1for CO2. The distance between the light source and the reflecting mirror was 35 m (for four chambers) or 100 m (for nine chambers). The details of the experimental set-up adopted for the FTIR spectroscopy method have been described in a previous paper (Chang et al., 2000). Because the interferometer char-acteristics are dominated by a high spectral resolution up to 0.06 cm1 and a configuration suitable for field measure-ments was needed, the MB-104 (BOMEM, Hartmann and Braun, Canada) was used for radiation emission as well as absorption measurements. The spectrum was observed on a PC screen and the mirror was adjusted to optimize the signal. The absorption peak areas and atmospheric concen-trations of CH4and CO2have high correlation coefficients (0.998–0.999 and 0.996–0.997, respectively) with distance between 0 and 100 m. The accuracy of FTIR spectroscopy in CH4 and CO2 measurement is within 2% and 3%,

Fig. 1. Fu-Der-Kan closed landﬁll (37 ha) was reconstructed a multi-use recreational park: (A) country activity area (2.6 ha), (B) Sun plaza (1.4 ha), (C) culture of indigenous people’s area (5 ha), (D1) and (D2) slightly lake, (E) boating area (0.8 ha), (F) creation garden (8 ha), (G) grass skiing ﬁeld (2 ha), (H) management center, (I) wastewater treatment factory and (J) natural resource landscape (12 ha).

(3)

respectively. The atmospheric concentrations of greenhouse gases were determined from long-path measurements by dif-ferential absorption and a least squares ﬁt of measurement, and were simulated by modeling air transmittances for various absorbed concentrations.

2.3. CH4and CO2emission rates

Gas samples were collected using a homemade closed acrylic chamber (length 40 cm, width 40 cm, height 65 cm, volume about 96 L) equipped with a fan, a thermometer, and a sampling hole. Either four or nine acrylic chambers were installed on the various landfill soils along a linear path. The chambers were inserted into the soil to a depth of 5 cm and allowed to equilibrate for 10 min before each measure-ment. Gas samples (35 ml) were collected from the head-space of each chamber and transferred to a 12.6 ml serum bottle sealed with a butyl rubber stopper and previously flushed with oxygen-free nitrogen gas (Yang and Chang, 1997). Gas samples were collected from chamber at 0.0, 0.5, and 1.0 h. Samples were taken again at 3.0, 3.5, and 4.0 h for comparison. CH4and CO2emission rates were cal-culated by performing a linear regression on the differences in CH4and CO2concentrations and adjusting for chamber volume and area covered. The following equation was used for the calculation (Rolston, 1986; Hegde et al., 2003):

f ¼ V A DC Dt

where, f = CH4 or CO2 emission rate (mg m2h1), V = volume of chamber above the soil (m3), A = cross-section of chamber (m2), DC = concentration diﬀerence between time zero and time t (mg m3), and Dt = time duration be-tween two sampling periods (h).

The ratios of atmospheric CH4and CO2concentrations measured with GC method to those measured with FTIR spectroscopy were analyzed. Gas samples were collected from the center of the FTIR spectrometer line at a height of 197 cm. CH4was analyzed by a Shimadzu 14A GC (Shi-madzu Co., Japan) with a glass capillary column (0.26 mm· 2 m) packed with Porapak Q (80/100 mesh) and a ﬂame ionization detector. The column temperature was set at 100C and the injection and the detector temper-atures were set at 130C. CH4 concentration was calcu-lated with a standard curve from 0.1 to 100 ppm, with a detection limit of 0.01 ppm (Yang and Chang, 1997). For CO2 analysis, a thermal conductivity detector was used. The column temperature was set at 150C and the injec-tion and the detector temperatures were set at 200C. CO2 concentration was calculated using a standard curve from 0.1 to 1000 ppm, with a detection limit of 0.05 ppm

(Chang and Yang, 2003).

2.4. Analytical methods

Air and soil temperatures were determined on site either directly or under at 10 cm depth with a Hg-thermometer.

Wind speed was measured with a weathercock (Weather link 4.0, USA). Light intensity was detected with a Toshiba SPI-5 photometer. Soil pH was determined using a pH meter directly in the soil or in a 1:1 (w/v) soil to water sus-pension. Moisture content was measured by drying the samples at 105C for 24 h until a constant weight was achieved. Total organic carbon (TOC) was estimated using a TOC-5000A analyzer (Code HI 8424C, Shimadzu, Japan). Experiments were carried out in quadruplicate; while nine experiments were used in the measurement of CH4and CO2emission rates on June 30, 2000. Data were subjected to coeﬃcient of variance analysis and Duncan’s multiple range tests (p = 0.05) using the Statistical Analysis System (SAS Institute, 2002).

3. Results

3.1. Properties of cover soils

The properties of cover soil at different burial locations are shown inTable 1. The cover soils were acidic (pH 5.6– 6.0) in 1999 and 2000, and they were neutral (pH 7.4–7.7) after the construction of a multi-use recreational park in 2006. There were significant pH differences between 1999–2000 and 2006 (p < 0.05). The soil moisture, soil tem-perature and TOC in 2006 were lower than those in 1999– 2000.

3.2. CH4and CO2concentrations and emission rates at site A

Air temperature, atmospheric concentrations of CH4 and CO2at site A are presented inFig. 2. Atmospheric con-centrations of CH4and CO2on October 30, 1999 ranged from 1.7 to 2.7 ppm and from 324 to 390 ppm, respectively. Average values were 2.2 ± 0.3 and 366 ± 24 ppm, respec-tively. On November 24, 1999, atmospheric concentrations of CH4and CO2ranged from 1.9 to 4.6 ppm and from 373 to 409 ppm, respectively, with average values of 3.1 ± 0.9 and 386 ± 11 ppm, respectively. Atmospheric concentra-tions of CH4and CO2on October 30 were lower than those on November 24 and it might be due to the high wind velocity on October 30 (4.2 ± 0.7 m s1) (Table 1). Further, the total organic C content of cover soil on November 24, 1999 in site A was about 25% lower than that on October 30, 1999 might be due to the construction as a multi-use recreation park during this period, and some new cover materials with low organic C content were layered on the surface. Therefore, both the organic C content of cover soil and the CH4emission on November 24, 1999 were lower than those on October 30, 1999. CH4 and CO2 emission rates ranged from 8.8 ± 0.3 to 21.7 ± 4.3 mg m2h1and from 495 ± 31 to 1012 ± 43 mg m2h1, respectively

(Table 1). CH4 emission rate was higher on October 30,

1999 than those on November 24, 1999 because of high soil organic C and moisture content for active microbial reaction.

(4)

Site A was subsequently converted into a country activ-ity area during the period from 1999 to 2003. Atmospheric concentrations of CH4on January 4, 2006 were between 1.8 and 3.1 ppm, with an average of 2.3 ± 0.5 ppm. CO2 on that date ranged from 373 to 441 ppm, with an average of 396 ± 23 ppm. CH4 and CO2 emission rates were

0.4 ± 0.3 to 2.3 ± 0.6 mg m2h1 and 72 ± 17 to 301 ± 5 mg m2h1, respectively. Average CH4and CO2 emission rate was 0.7 ± 0.4 and 95 ± 6 mg m2h1, respec-tively (Fig. 3a–c). CH4and CO2emission rates in 2006 were lower than those in 1999 because of the more stable burial waste, low soil organic C and low moisture content.

Table 1

Properties of cover soils in Fu-Der-Kan landﬁll and after as the multi-use recreational park

Sampling pH Moisture Soil Atmos. Humidity Wind Organic CH4emission CO2emission

date content temp. pressure velocity carbon rate rate

(%) (C) (Pa) (%) (m s1) (%) (mg m2h1) (mg m2h1) Site A Oct. 30, 1999 5.5–6.1b _{31.1 ± 1.7}a _{31.9 ± 1.0}a _{1015 ± 1}a _{74.4 ± 6.1}a _{4.2 ± 0.7}b _{20.8 ± 1.4}a _{21.7 ± 2.4}b _{1012 ± 14}a Nov. 24, 1999 5.5–6.2b _{29.6 ± 1.8}a _{30.9 ± 1.0}a _{1014 ± 1}a _{79.3 ± 7.8}a _{2.3 ± 0.9}a _{15.4 ± 1.0}b _{8.8 ± 0.3}c _{495 ± 11}b Jan. 3, 2006 7.6–7.7a _{19.6 ± 2.2}c _{20.5 ± 0.8}c _{1018 ± 4}a _{73.7 ± 6.4}a _{3.0 ± 1.6}a _{2.0 ± 0.2}c _{0.7 ± 0.4}d _{95 ± 6}c Site B Oct. 30, 1999 5.1–6.2b 24.0 ± 0.3b 31.1 ± 1.0a 1015 ± 1a 74.4 ± 6.1a 4.2 ± 0.7b 14.7 ± 1.7b 12.8 ± 0.2c 901 ± 15a Nov. 24, 1999 5.3–6.3b 29.3 ± 2.2a 26.9 ± 0.6b 1014 ± 1a 79.3 ± 7.8a 2.3 ± 0.9a 19.3 ± 2.2a 35.0 ± 3.1b 813 ± 17a May 11-13, 2000 5.3–6.4b 26.5 ± 3.0b 29.5 ± 2.3a 1007 ± 1b 62.7 ± 6.5b 2.4 ± 1.2a 19.7 ± 3.0a 18.4 ± 2.2b 1531 ± 23a Jan. 4, 2006 7.2–7.7a 17.5 ± 1.0c 20.6 ± 0.7c 1018 ± 4a 73.7 ± 6.4a 3.0 ± 1.6a 2.3 ± 0.2c 0.1 ± 0.4d 20 ± 5e Site C June 30, 2000 5.6–6.3b _{33.6 ± 2.9}a _{33.6 ± 2.1}a _{1007 ± 1}b _{71.1 ± 7.4}a _{1.5 ± 0.8}a _{20.7 ± 4.0}a _{163.3 ± 7.7}*a _{1349 ± 16}*a Jan. 5, 2006 7.3–7.6a _{22.5 ± 2.0}c _{20.0 ± 0.9}c _{1018 ± 4}a _{73.7 ± 6.4}a _{3.0 ± 1.6}a _{2.6 ± 0.5}c _{0.2 ± 0.2}d 32 ± 7d

Means ± SD (n = 4, *_{n = 9). Emission rate was measured by chamber method. Values in the same column that do not share the same alphabetic}

superscript are signiﬁcantly diﬀerent at 5% level according to Duncan’s multiple range tests.

Temp. o ( C) 5 15 25 35 45 CO 2 Conc . (ppm) 300 350 400 450 CH 4 Conc . (ppm) 1 2 3 4 5 CO₂ CH₄

b

c

a

CO 2 Conc. (ppm) 300 350 400 450 CH 4 Conc . (ppm) 1 2 3 4 5 Sampling time Temp. o ( C) 5 15 25 35 12:00 12:30 13:00 13:30 14:00

d

Fig. 2. Air temperature and atmospheric concentrations of CO2and CH4measured with FTIR spectroscopy at site A of the Fu-Der-Kan closed landﬁll.

Data are presented for (a–b) October 30, 1999 and (c–d) November 24, 1999: (a) and (c) air temperature; (b) and (d) atmospheric concentrations of CO2

(5)

CO 2 emission (mg m -2 h -1 ) -500 -250 0 250 CO 2 Conc. (ppm ) 300 350 400 450 CO₂ CH₄ Te mp . ( o C) 5 15 25 35 45

b

c

a

Sampling time 20 :00 10 :00 12 :00 14 :00 16 :00 18 :00

e

f

d

20 :00 12 :00 14 :00 16 :00 18 :00 CH 4 Co n c . (ppm )

h

i

g

CH 4 emission (mg m -2 h -1 ) -2 -1 0 1 20 :00 12 :00 14 :00 16 :00 18 :00 1 2 3 4 5

Fig. 3. Air temperature, atmospheric concentrations and emission rates of CO2and CH4of the Fu-Der-Kan landﬁll after as the multi-use recreational

park. (a–c) country activity area on January 4, 2006. (d–f) Sun plaza on January 5, 2006. (g–i) culture of indigenous people’s area on January 12, 2006. (a), (d) and (g) air temperature; (b), (e) and (h) atmospheric concentrations of CO2and CH4; (c), (f) and (i) CO2and CH4emission rates.

Temp. o ( C) 5 15 25 35 45 CO 2 Conc. (ppm ) 300 350 400 450 CH 4 Conc . (ppm ) 1 2 3 4 5 CO₂ CH₄

b

c

a

Sampling time CO 2 Conc . (ppm) 300 350 400 450 CH 4 Conc . (ppm ) 1 2 3 4 5 Temp. o ( C) 5 15 25 35 15:30 16:00 16:30 17:00 17:30

d

Fig. 4. Air temperature, and atmospheric concentrations of CO2and CH4measured with FTIR spectroscopy at site B of the Fu-Der-Kan closed landﬁll.

Data are presented for (a–b) October 30, 1999 and (c–d) November 24, 1999: (a) and (c) air temperature; (b) and (d) atmospheric concentrations of CO2

(6)

3.3. CH4and CO2concentrations and emission rates at site B

Air temperature, atmospheric concentrations of CH4 and CO2at site B are shown in Fig. 4. Atmospheric CH4 concentrations on October 30, 1999 ranged from 1.8 to 3.0 ppm, with an average of 2.4 ± 0.3 ppm. CO2 concentra-tions were between 373 and 394 ppm, with an average of 388 ± 6 ppm. On November 24, 1999, atmospheric CH4 and CO2concentrations were ranged from 2.6 to 4.2 ppm, and from 347 to 397 ppm, respectively, with an average of 3.6 ± 0.5 and 378 ± 15 ppm, respectively. CH4 and CO2 emission rates were 12.8 ± 0.2 to 35.0 ± 3.1 mg m2h1 and 813 ± 17 to 901 ± 15 mg m2h1, respectively (Table 1). Both the atmospheric concentration and emission rate of CH4 were high on November 24, 1999, which may be partly attributed to high soil organic C and moisture con-tent. Diﬀerent situations, in contrast to site A, were found in site B on November 24, 1999. The new cover material used in site B on November 24, 1999 might contain high organic C. Therefore, both the organic C content of cover soil and the CH4emission in site B on November 24, 1999 were higher than those on October 30, 1999.

Sun plaza was constructed at site B during the years 1999 to 2003. Atmospheric CH4and CO2concentrations on January 5, 2006 ranged from 1.8 to 2.3 ppm and from 332 to 395 ppm, respectively. Average values were 1.9 ± 0.2 and 361 ± 12 ppm, respectively. CH4 emission rate was 1.1 ± 0.5 to 0.6 ± 0.5 mg m2h1 and CO2 emission rate was98 ± 11 to 224 ± 38 mg m2h1. Aver-age CH4 and CO2 emission rates were 0.1 ± 0.4 and 20 ± 3 mg m2h1, respectively (Fig. 3d–f). These values

were lower than those in 1999 because of the more stable buried refuse, low soil organic C and moisture content. 3.4. CH4and CO2concentrations and emission rates at site C

The CH4 emission rate (163.3 ± 7.7 mg m2h1) and CO2emission rate (1349 ± 16 mg m2h1) at site C were the highest of the three sites and the variation among the nine installed chambers was also large (Table 1). The ele-vated CH4 and CO2 emission rates were due to new MSW was buried in this site during 1999.

Site C was converted into a culture of indigenous peo-ple’s area during the years 2001 to 2003. Atmospheric CH4concentrations were 1.9 to 2.3 ppm, with an average of 2.1 ± 0.2 ppm on January 12, 2006. CO2 concentra-tions ranged from 343 to 422 ppm with an average of 397 ± 13 ppm. CH4 and CO2 emission rates were 0.8 ± 0.7 to 1.1 ± 0.5 mg m2h1 and 135 ± 8 to 113 ± 15 mg m2h1, respectively. Average CH4and CO2 emis-sion rates were 0.2 ± 0.2 and32 ± 7 mg m2h1, respec-tively (Figs. 3 g–i). CH4 emission rates were in the same level as those in sites A and B.

3.5. Diurnal variation of atmospheric CH4and CO2 concentrations and emission rates at site B

Fig. 5 shows the diurnal variations in air temperature,

atmospheric concentrations and emission rates of CH4 and CO2 from May 11 to 13, 2000 at site B. The atmo-spheric CH4 concentrations ranged from 1.3 to 3.8 ppm with an average of 2.7 ± 0.8 ppm. The concentration of

Temp . ( o C) 5 15 25 35 45 CO 2 Conc. (ppm ) 300 350 400 450 CH 4 Conc. (ppm) CO₂ CH₄

b

c

a

CO 2 emission (mg m -2 h -1 ) 0 1000 2000 CH 4 em ission (m g m -2 h -1 )

May,11 May,12 May,13

9 15 21 3 9 15 21 3 9 Sampling time 1 2 3 4 5 0 25 50 75

Fig. 5. Diurnal variation in the atmospheric concentrations and emission rates of CO2and CH4at site B in the Fu-Der-Kan closed landﬁll during May 11

(7)

CH4was high at night, which correlates with CH4emission rate. The atmospheric CO2 concentration averaged at 374 ± 15 ppm and ranged from 350 to 399 ppm. The air temperature was the highest (31.7C) at noon on May 11 and the lowest (19.5C) at night on May 12.

The average CH4 emission rate at site B was 18.4 ± 3.3 mg m2h1 (Table 1). The maximum rate was 54.1 mg m2h1 at 3 a.m. on May 13, and the minimum rate was 2.5 mg m2h1at midnight on May 12. The aver-age CO2 emission rate was 1531 ± 23 mg m2h1. The maximum rate was 2302 mg m2h1at 9 a.m. on May 11, and the minimum rate was 893 mg m2h 1 at midnight on May 12.

The diurnal variation in CO2emission was very consis-tent after completion of the recreational park for country activity area, Sun plaza and culture of indigenous people’s area in January 2006. The CO2emission rates were high at night because of daytime photosynthetic activity by the grass and small trees. The diurnal variation of CH4 emis-sion was insigniﬁcant due to the low organic matter content and more stable burial waste.

4. Discussion

Compared to other terrestrial ecosystems, landfills are characterized by a high rate of CH4production and large CH4 gradients from the deeper production zones to the soil–atmosphere interface. Bogner et al. (1999)found that CH4 concentration was low at the surface and increased by about two orders of magnitude at 100 m depth. In paddy fields, the CH4emission rate has been found to be high at 11 a.m. to 2 p.m. and low in the early morning due to the shallow of rice root and the high effect of soil temperature (Yang and Chang, 1999, 2001). In contrast, the CH4 emission rate at the closed Fu-Der-Kan landfill was the greatest at night for the collection of emission gases and CH4 at daytime. A similar phenomenon was also observed at the Shan-Chu-Ku landfill during years 1–2, 2–3 and 5 (Hegde et al., 2003). However, no significant diurnal variation of emission rate was observed in the park constructed at the Fu-Der-Kan landfill. There was little fluctuation of soil temperature in the deeper zone of landfill.

CH4and CO2emission rates depend on soil pH, mois-ture content, pressure, total organic carbon, kind and age of the burial waste, gas extraction system, kind and depth of soil cover and methane oxidation. CH4and CO2 emis-sion rates were the highest in the site C because of new MSW was buried during 1999. CH4 and CO2 emission rates were higher in site A on October 30, 1999 than those on November 24, 1999 due to high organic C and moisture content for active microbial reaction.

CH4and CO2emissions in the park varied widely and included negative values. Negative CH4 emission resulted from a net oxidation of atmospheric CH4 by indigenous CH4 oxidizing microbes. CO2 emission was negative due to net photosynthetic use of CO2by grass and small trees.

Atmospheric concentrations and emission rates of CH4 and CO2 were lower in 2006 than those in 1999 and 2000. This might be due to more stable buried refuse and low organic C content (2.0–2.6% in 2006 vs. 14.7–20.8% in 1999–2000). The ﬂuctuation of CH4 emission rates (1.1 to 2.3 mg m2h1) was minor in 2006. The CO2 emission rate was low in the daytime due to photosynthesis and it increased at night.

Hegde et al. (2003) measured the greenhouse gas

con-centrations in a homogenous grassland at the National Tai-wan University campus, and found no major differences in the ratios of the concentrations measured with the GC method to those measured with FTIR spectroscopy (ratio is 1.10 in CH4 and 1.02 in CO2). However, the ratios of atmospheric CH4 concentration measured with GC method to those measured with FTIR spectroscopy in the Shan-Chu-Ku landfill were 2.28–4.78, 2.82–8.05, and 1.72–7.57 at sites 1–2, 2–3, and 5 years old, respectively. The ratio decreased with the increasing of burial period. In this study, the ratios of atmospheric CH4concentration in the closed landfill during 1999–2000 were 1.08–1.55 at site A, 1.06–1.75 at site B, and 1.21–2.47 at site C. The ratios had decreased to 0.96–1.16, 0.97–1.15, and 0.99– 1.25 in 2006 for the country activity area, Sun plaza and culture of indigenous people’s area, respectively. These val-ues were lower than those in the Shan-Chu-Ku landfill because of the more stable organic matter of buried refuse. Furthermore, the ratios at the reconstructed areas in 2006 were close to those in the campus grassland, which had the same level of soil organic C content. Therefore, the ratio of atmospheric CH4 concentration measured with GC method to FTIR spectroscopy depends on the organic C content and could be used as a parameter to assess the sta-bility of the buried refuse.

The presence of water vapor interfered with the CO2 measurement by FTIR spectroscopy. CO2 concentration was suddenly dropped at 16:45 p.m. on November 24, 1999 for the increase of relative humidity from 81 to 91%. While CO2 concentration increased at afternoon on January 12, 2006 for the decrease of relative humidity from 88 to 67%. Therefore, the background of measurement and alignment are necessary to crosscheck the disturbances of relative humidity change on CO2concentration.

Due to low organic C content, both the atmospheric concentration and emission rate of CH4 at the Fu-Der-Kan closed landfill were less than 10% of those at the Shan-Chu-Ku landfill (Hegde et al., 2003). The CH4 emis-sion rate at the reconstructed areas in 2006 was less than 2 mg m2h1 which was less than 5% of that measured at the Fu-Der-Kan closed landfill site in 1999–2000. Again, this might be due to low organic C content and more stable organic matter of buried refuse.

In conclusion, both gas-type open-path FTIR spectros-copy and GC chamber method have been successfully used to measure atmospheric concentrations and emission rates of greenhouse gases in a closed and reconstructed landﬁll. The Fu-Der-Kan landﬁll closed in 1995 and was

(8)

reconstructed to form a multi-use recreational park during 1999 to 2003. Atmospheric concentrations and emission rates for CH4were higher at night than the daytime during 1999 to 2000. The CH4concentrations and emission rates in the closed landfill and reconstructed park were less than 10% and 1% of those in the active landfill, respectively. A systematic and continuous study of greenhouse gas emis-sions from closed landfills provides information about the field degradation of MSW, which in turn will help in the design of future landfills.

Acknowledgements

The authors thank Professor E. H. Chang and Dr. I. C. Lin for their technical assistances, and the National Science Council of Taiwan for ﬁnancial support (NSC 88-2811-Z-002-0001, NSC 89-EPA-Z002-003 and NSC EPA-90-FA17-03-90B040).

References

Bogner, J., Matthews, E., 2003. Global methane emissions from landﬁlls: New methodology and annual estimates 1980–1996. Global Biogeo-chem. Cy., 17. Article No. 1065.

Bogner, J.E., Spokas, K.A., Burton, E.A., 1999. Temporal variations in greenhouse gas emissions at a midlatitude landﬁll. J. Environ. Qual. 28, 278–288.

Chang, T.C., Leo, Y.C., Yang, S.S., 2000. Determination of greenhouse gases by open path gas-type FTIR spectroscopy. Food Sci. Agri. Chem. 2, 7–14.

Chang, T.C., Yang, S.S., 2003. Methane emission from wetlands in Taiwan. Atmos. Environ. 37, 4551–4558.

Czepiel, P.M., Shorter, J.H., Mosher, B., Allwine, E., McManus, J.B., Harriss, R.C., Kolb, C.E., Lamb, B.K., 2003. The inﬂuence of atmospheric pressure on landﬁll methane emissions. Waste Mange. 23, 593–598.

Environmental Protection Administration/Taiwan, 2007. Oﬃcial Statistic Report-2006. Environmental Protection Administration, Taipei, Taiwan.

Hegde, U., Chang, T.C., Yang, S.S., 2003. Methane and carbon dioxide emissions from Shan-Chu-Ku landﬁll site in northern Taiwan. Chemosphere 52, 1275–1285.

Jang, H.D., Yang, S.S., 2001. Greenhouse gases emission from municipal solid wastes in the column bioreactor. J. Biomass Energy Soc. China 20, 101–112.

Kumar, S., Gaikwad, S.A., Shekdar, A.V., Kshirsagar, P.S., Singh, R.N., 2004. Estimation method for national methane emission from solid waste landﬁlls. Atmos. Environ. 38, 3481–3487.

Meraz, R.L., Vidales, A.M., Dominguez, A., 2004. A fractal-like kinetics equation to calculate landﬁll methane production. Fuel 83, 73–80. Rolston, D.E., 1986. Gas ﬂux. In: Klute, A. (Ed.), Methods of Soil

Analysis, second ed. American Society Agronomy and Soil Science Society, American Monograph No. 9, Wisconsin, USA, pp. 1103–1119. SAS Institute., 2002. SAS/STAT user’s Guide, Release 6.03 SAS, Inst.

Cary., North Carolina, USA.

Simpson, I.J., Rowland, F.S., Meinardi, S., Blake, D.R., 2006. Inﬂuence of biomass burning during recent ﬂuctuations in the slow growth of global tropospheric methane. Geophys. Res. Lett., 33. Article No. L22808. US Environmental Protection Agency, 2002. Greenhouse Gases and

Global Warming Potential Values. Excerpt from the Inventory of US Greenhouse Emissions and Sinks: 1990–2000. pp. 16. (http://www. epa.gov/globalwarming/publications/emissions).

US Environmental Protection Agency, 2006. Global Mitigation of Non-CO2Greenhouse Gas. 3. Landﬁll Sector. US Environmental

Protec-tion Agency, Washington DC, USA.

Yang, S.S., Chang, E.H., 1997. Eﬀect of fertilizer application on methane production in paddy soils of Taiwan. Biol. Fert. Soils 25, 245–251. Yang, S.S., Chang, H.L., 1998. Eﬀect of environmental conditions on

methane production and emission from paddy soil. Agri. Ecosyst. Environ. 69, 69–80.

Yang, S.S., Chang, H.L., 1999. Diurnal variation of methane emission from paddy ﬁelds at diﬀerent growth stages of rice cultivation in Taiwan. Agri. Ecosyst. Environ. 76, 75–84.

Yang, S.S., Chang, H.L., 2001. Effect of green manure amendment and flooding on methane emission from paddy fields. Chemosphere: Global Change Sci. 3, 41–49.