Bioavailability of Heavy Metals During Humification of Organic Matter in Pig Manure Compost

By G. F. Huang¹, J. W. C. Wong², B. B. Nagar³, Q. T. Wu¹ and F. B. Li¹
September 2005

  1. College of Natural Resources and Environment, South China Agricultural University, Guangdong, China
  2. Department of Biology, Hong Kong Baptist University, Kowloon Tong, Hong Kong
  3. Corresponding Author, House No 9, Ward No 3, Bishnah, Jammu-Tawi - 181 132, India
The aim of this study was to evaluate the bioavailability of heavy metals during co-composting of pig manure with sawdust. Addition of sawdust as co-compost to pig manure caused a significant reduction of total heavy metals in the compost due to dilution effect. However there was a slight increase in total Cu and Zn concentration in compost because of loss of mass due to biodegradation of organic matter. Composting process decreased the content of DTPA-extractable Cu and Zn, and affected the speciation of heavy metals in compost. It resulted in a significant reduction in percentage of water soluble, extractable, carbonate and Fe-Mn oxide fraction of Cu and Zn, and an increase in organic and residual fraction of Cu and Zn in compost. It could be suggested that co-composting of pig manure with sawdust proved beneficial in decreasing the potential mobility and bioavailability of heavy metals in compost product. The distribution of Cu in humic acid (HA) and Zn in fulvic acid (FA) increased significantly, while no great change of Zn and Cu was observed in HA and FA fractions respectively after composting. The organo-metallic complex was also evaluated using gel chromatography to separate soluble organic acid fraction of HA and FA fractions. HA fraction showed a selective preference for Cu rather than Zn, while opposite was listed for FA fraction. However, it showed that the peaks of humate-Cu content in elution fractionations overlapped the UV-absorbance peaks of the elution fractionation of HA smoothly, but did not appear in humate-Zn, fulvate-Cu and fulvate-Zn. The study showed that the bioavailability of Cu and Zn was dependent upon the HA/FA fraction ratio of organic compost, as both metals showed different preferential binding for HA & FA fractions. The presence of HA fraction reduced the Cu bioavailability, while FA fraction resulted in decreased bioavailability for Zn.

Key words: Manure compost, humification, heavy metal, bioavailability, Gel chromatography, C/P ratio


The large quantities of animal waste generated as by-products of intensive animal production find their utilisation in application to agricultural land. Continued disposal of large amounts of raw manure on limited areas of nearby land in close to the production facility may result in soil and water contamination via leaching of toxic elements, nutrient imbalances and phytotoxicity in crops. Increasingly, composting is becoming an environmentally and economically alternative method for treating solid wastes. Through composting, active organic matters in fresh waste are converted into source of nutrients for plant growth and soil conditioner for improving soil physical properties (Hoitink and Fahy, 1986; Chen et al., 1989). The high content of stabilized organic matter and the presence of nutrients of good quality composts are a guarantee of agronomic advantages. Increasing of organic matter and nutrient availability in soil are evident after compost stabilization (Bevacqua et al., 1993; McConnell et al., 1993). The presence of organic and inorganic contaminants in compost, pose a grave danger to the environment of which heavy metal is the main obstacle factor leading to restricted agricultural use of compost. Literatures on the effect of compost use on heavy metal levels in the soil environment show that it varies according to soil types, plant species and compost quality. Increased Zn, Cu and Pb concentrations have been often been observed, both in the soil and plants, while other heavy metals such as Cd, Ni, and Cr increase less consistently (Bevacqua et al., 1993). In the long term, the use of sewage sludge can also cause a significant accumulation of Zn, Cu, Pb, Ni and Cd in soil and pants (Mulchi et al., 1991; Williams et al., 1980). Bioavailability of heavy metals has been considered more important than the total content for a better understanding of a possible transfer of metal into the food chain (Petruzzelli et al., 1989).

In order to determine heavy metal availability from organic waste compost two chemical i.e., single extraction and sequential extraction with various chemical reagents were used in the present study. (Petruzzelli, 1989; He et al., 1995; Tisdell and Breslin, 1995; Qiao and Ho, 1997). Utilization of DTPA in single extraction for evaluating the availability of metals from compost by plants has been well documented (Garcia et al., 1990; Ciavetta et al., 1993). It provides additional information of the origin, biological and physicochemical availability, mobilization, and transport or heavy metals (Lake, 1987; (Clevenger and Mullins, 1982; Soon and Bates, 1982; Lake et al., 1984). Assuming that bioavailability is related to solubility, then metal bioavailability decreases in the order: water-soluble > exchangeable > carbonate > Fe-Mn oxide > organic > residual (Xian, 1989). On the other hand, the binding of metals to humic substances is also an important factor that can potentially control the mobility of heavy metals in soils. Classical chemical sequential extraction procedures are not sufficient to study the influence of composting process. There are a number of previous studies present on metal distribution in humic substances but the information provided in these studies are not appropriate with particular reference to its application to pig manure compost, due to difference in compost sources, composting technology and reagent choice for humic substance extraction. Therefore, the DTPA-extraction, sequential extraction, distribution and interaction of Cu and Zn with humic substances by gel chromatography were performed during co-composting of pig manure with sawdust in this study.

The progress of the maturation process of the compost significantly influences the humification of the organic and relative percentages of organic carbon present in the humic and fulvic acids. A correctly carried out maturation may in fact increase the content of HA fraction significantly with respect to the FA fraction,. Furthermore, these variations in the quantity and quality of humic substances influence the speciation of heavy metals. In fact, if the maturation process has been well carried out, a large amount of heavy metals are complexed by the humic substances and reach the soil in a less mobile chemical form (Canarutto, et al., 1991). Consequently it is much more difficult for them to enter the plants. Moreover, many studies demonstrated that organic compost as a source of humified organic matter for contaminated soils can also decrease the availability of heavy metals in amended soil (Ciavatta et al., 1993). Therefore, the main objectives of this study were to investigate different chemical forms of Cu and Zn; its distribution in HA and FA fractions; and the complexing ability of HA and FA fractions with Cu and Zn during the co-composting of pig manure with sawdust with ultimate goal to evaluate the bioavailability of Cu and Zn in matured compost product.

Materials and Methods

Composting Pile Construction

The experiment was conducted at the Kadoorie Botanic and Farm Corporation in Hong Kong. Two composting piles were prepared by mixing pig manure with sawdust at about 4:1 (w/w fresh weight basis) to obtain a C/N ratio of 30. Twigs were added to the pile at 10% (v/v) as a bulking agent, and the moisture content of the composting mass was kept at 60%-70% (w/w). The temperature was measured daily at a depth of 0.30 m within the composting piles. The 8-m³ heap was turned and mixed at every 3 days to provide aeration. Triplicated samples were collected from each pile at day 0, 3, 7, 14, 21, 35, 49 and 63. Sub samples were air-dried, ground to pass through a 0.25 mm sieve and stored in a desiccator for further analyses. The analysed physicochemical properties of the experimental materials are shown in Table 1.

Chemical Analysis

The aqueous extracts of compost were obtained by mechanically shaking the samples with double distilled water (DDW) at a solid: DDW of 1:10 (w/v) for 1 h. The suspensions were centrifuged at 12,000 rpm and filtered through 0.45-um membrane filters. The filtrates were used for soluble organic carbon determination by a SHIMADZU TOC-5000A Total Organic Carbon Analyzer, total soluble N by indophenol-blue method after Kjeldahl digestion. Total organic C of bulk compost was measured by using the Walkley-Black Method; total N was determined by indophenol-blue method after the Kjeldahl digestion (Page et al., 1982). C, H, N and S analysis were determined on a PEKIN-ELMER Series II CHNS/O analyzer, and oxygen were calculated by subtracting the above from total elemental composition.100%. Total acidity and acidic functional groups were measured according to the methods described by Schnitzer & Gupta (1965).

Preparation of Humic Substances

Fresh samples were extracted with 0.1 M Na2P4O7(at pH 7) for 24 h at a solid: extractant ratio of 1:10 (w/v, dry weight basis). Supernatant solution containing soluble humic substances were separated by centrifugation at 12,000 rpm for 20 min and the same procedure was performed in the residues for three more times. The combined extracts were filtered through 0.45-um membrane filters and acidified to pH 1 with 3 M H2SO4, and were allowed to stand at room temperature for 24 h,. The precipitated humic acids (HA) fraction was separated from the fulvic acid fraction (FA) by centrifugation at 12,000 rpm for 20 min. The humic acids fraction was were centrifuged, washed several times with distilled water and 0.1 M HCl. It was again re-dissolved in 0.1 M neutral Na2P4O7. The HA and FA fractions were dialyzed against distilled water until a negative Na+, K+, and Cl-test was obtained and then freeze-dried (Gonzalea-Vila et al., 1985; Filip et al., 1985; Garcia et al., 1992).

Sequential Extraction of Heavy Metals

One gram of each sample was weighed into a 30-ml polycarbonate centrifuge tube and the following fractions obtained (Lena and Gade, 1997).

Water Soluble
Composting samples were extracted with 15 ml of deionised water for 2 h.
The residues from water-soluble fractions were extracted with 15 ml of 1 M MgCl2(pH 7.0) for 1 h.
The residues from exchangeable fractions were extracted with 8 ml of 1 M NaOAc (adjust to pH 5.0 with HOAc) for 5 h.
Fe-Mn Oxides-bound
The residues from carbonate fractions were extracted with 15 ml 0.04 M NH2OH·HCl in 25% (v/v) HOAc at 96°C with occasional agitation for 6 h.
The residues from Fe-Mn oxide fraction were extracted with 3 ml of 0.02 M HNO3and 5ml of 30% H2O2(adjusted to pH 2 with HNO3). The mixture was heated to 85°C for 2 h, with occasional agitation. A second 3 ml aliquot of 30% H2O2(pH 2 with HNO3) was added and the mixture heated again to 85°C for 3 h with intermittent agitation. After cooling, 5 ml of 3.2 M NH4OAc in 20% (v/v) HNO3was added and the samples diluted to 20 ml and agitated continuously for 30 min.
The residues from organic fraction were digested using a mixed acid of 1:1 HNO3-HClO4(v/v).

After each successive extraction, separations were performed by centrifugation (Beckman Model J2-21) at 25000 rpm for 20 min. The supernatants were removed with a pipette, filtered with 0.2-um Nucleopore polycarbonate membrane filters, and analysed for metals. The residue was washed with 15 ml of deionised water followed by vigorous hand shaking, and then followed by 20 min of centrifugation before the next extraction. The metal concentrations of the supernatants from each step were analysed by atomic adsorption spectrophotometer (Varian-Techtron AA975).

DTPA-Extractable Metals

DTPA-extractable metal contents were obtained by mechanically shaking the compost samples at a 1:5 solid: extractant ratio (w/v) for 2 h each with 0.005M DTPA (diethylene triamine pentaacetic acid) plus 0.1M TEA (triethanolamine) and 0.01 M CaCl2buffer at pH 7.3 with diluted HCL (Page et al., 1982). Compost extracts were stored in plastic bottles at 4°C. The heavy metal analysis was performed using atomic absorption spectroscope (Varian-Techtron AA975).

Distribution of Heavy Metals in HA and FA

50 mg each sample was dissolved in 5 ml Milli-Q water, soniccated for 15 min and allowed to stand overnight at room temperature. The content of Cu and Zn were measured using atomic absorption spectrophotometer (Varian-Techtron AA975).

100 mg of freeze-dried HA and FA fraction containing samples were dissolved in 10 ml Milli-Q water. 1.0 ml of these samples was introduced onto the bed of a glass column carefully packed with 66 cm of Sephadex G-25 gel (nominal upper exclusion limit of 5000 Da). The total volume of the gel (Vt) was 298 ml., The gel void volume (V0), determined by passing Dextran Blue-2000 through the column, was 120 ml. Milli-Q water at pH 6.8 was employed as the elutant. Elution pressure was maintained by a peristaltic pump set up at a flow rate of 1 ml/min. The eluted fractionated samples were received in increments of 5 ml using an automatic rotary fraction collector. Each 5 ml fractions were monitored for absorbance at 280 nm on a SHIMADZU UV-1601 UV/VIS spectrophotometer. The metal contents of each fraction were analysed using an atomic absorption spectrophotometer (PERKIN-ELMER 2380) equipped with a graphite furnace atomizer.

Humic substances content, soluble organic C, soluble N, total carbon, total nitrogen values were calculated as average value for triplicate set of samples. DTPA-extraction and chemical fractionation were conducted in triplicate in acid-bathed (5% HNO3) polycarbonate lab ware. At least one duplicate and one-spike sample was run for every 10 samples to verify precision of the method during the determination of heavy metals. The spike recovery and precision were found to be within 100 ± 10%. Gel chromatographic diagrams are obtained from the mean absorbance and heavy metal content in elution fractionation at the same elution volume of duplicate samples.

Results and Discussion

Figure 1: Change in temperature during the co-
composting of pig manure with sawdust
Figure 1

Composting of Pig Manure with Sawdust

Three distinct phases were observed during the process (Figure 1). Temperature rose from 25°C to 68°C during the first 4 days and then remained at about 60°C for about 5 weeks for the thermophilic phase. Excessive rise in temperature was avoided by turning and mixing of compost pile. The temperature started dropping after 43 days, and reached the same as that of the ambient temperature after 57 days of composting.

Figure 2: Change of soluble carbon and total carbon
during the co-composting of pig manure with sawdust
Figure 2

Water-soluble organic C content gradually decreased from 6220 mg/kg to 5784 mg/kg after composting for a week, and then continued to 3991 mg/kg at 63 days (Figure 2). Total organic C content of bulk of the compost showed a similar trend same as that of soluble organic C content, which decreased rapidly from 45% at the beginning to 35% after 49 days. This indicated the decrease in easily biodegradable substances and attainment of compost maturity (Garcia et al., 1991; Chen and Inbar, 1993; Saviozzi et al., 1992; Inbar et al., 1993). In mature compost most of the soluble organic C content is present as humic substances, which are resistant to further decomposition.

Figure 3: Change of C/Nsolid ratio and C/Norganic ratio
during the co-composting of pig manure with sawdust
Figure 3

The C/Nsolid ratio decreased rapidly from an initial value of 28.8 to, to 16.4 after 63 days (Figure 3). This C/Nsolid ratio could be considered satisfactory for maturity when the initial value is 25 to 30 for composting materials. The organic C/Nsoluble ratio also followed a similar trend as C/Nsolid, but it decreased more sharply, from 11 at the beginning to 5.3 after 63 days. Chanyask and Kubota (1981) suggested that C/Nsoluble in aqueous phase could be used as an indicator for compost maturity as composting reaction is a biochemical decomposition of organic matter occurring mainly in the aqueous phase. Compost with a C/Nsoluble of 5 to 6 is considered to attain maturity. Therefore, it can be concluded that pig manure compost in the present experiment reached maturation after 63 days of composting.

Figure 4: Total Cu and Zn of the compost at
0 day, 35 day and 63 day
Figure 4

A slight increase in Cu and Zn ions was observed during the composting (Figure 4), which may be caused by the reduction in volume of pile and an increase in the concentration of metal content after composting. The content of total Cu and Zn ions was 78 mg/kg of Cu and 313 mg/kg of Zn in matured compost, which were much lower than o their respective initial concentration of 556mg/kg of Cu and 1446 mg/kg of Zn in raw manure. This was simply due to the addition of large amount sawdust into the compost. This enlarged the volume and diluted the metal content in compost. Moreover, the reduction in volume of pile after composting also resulted in an increased concentration of heavy metals.

Sequential Extraction of Metals

Figure 5: Chemical fractionation of Cu and Zn
at 0 day, 35 day and 63 day
Figure 5

Figure 5 shows the chemical speciation of Cu and Zn in pig manure compost at day 0, day 35 and day 63. The percentages of water extractable, exchangeable, carbonate and Fe-Mn oxide form of Cu and Zn decreased significantly after composting for 35 days, while a great increase in organic and residual form of Cu and Zn was observed in matured compost at 63 days. The water soluble fraction of Cu decreased from 13% to 2.8%, while in Zn, it decreased from 2.72% to 2.09%,. Exchangeable fraction of Cu decreased from 3.7% to 2.06%, while for Zn, it decreased from 7.87% to 2.98% after composting for 63 days,. The organic fraction of Cu increased from 32.7% to 41.6%, in Zn, it increased from 7.56% to 12.3%. Residual fraction of Cu and Zn increased from 29.9% to 40%, and 34% to 36%, respectively. The non-residual fraction is the sum of all fractions except the residual fraction, which is another useful factor for evaluating the bioavailability of heavy metals. After composting, the non-residual fraction of Cu decreased from 70.% at 0 day, to 52.% at the end of composting, while in case of Zn, the concentration decreased from an initial 67.% to 61.% at day 63.

The low percentage of metals in non-residual fraction after composting signify a low availability of the metals to crop plants, as the readily soluble form of a trace metal is only considered as the more bio-available. It can be suggested that the bioavailability of Cu and Zn in pig manure decreased significantly after co-composting with sawdust.

As shown in Figure 5, Cu in mature compost showed evident high percentages in organic and residual form of 41 % and 47% respectively, while 27% in Fe-Mn oxide form and 39% in residual form of Zn. It showed that Cu was mainly distributed in residual and organic fractionations, but Zn was found in residual and Fe-Mn oxide fractionations in mature compost. It can be suggested that the availability of Zn may be higher than that of Cu in mature compost products. The chemistry of metal ion distribution during the composting process suggests that during the initial phase of composting, the Cu-hydroxide may get precipitated in early composting period, showing its change from water-soluble portion to residual portion (Cotton & Wilkinson, 1980).

Figure 6: DTPA-extractable Cu and Zn of the compost at
0 day, 35 day and 63 day
Figure 6

DTPA-Extractable Metals

DTPA fraction is better related with trace metal availability to plants, as it is widely use in evaluating the bioavailability of heavy metals in soils and composts. As shown in Figure 6, the contents of DTPA-extractable Cu and Zn decreased along with the composting process. DTPA-Cu decreased from 15.7 mg/kg of the initial compost to 11mg/kg of the mature product, while DTPA-Zn decreased from 119.85 mg/kg to 94.9mg/kg. The reduction of DTPA-Cu was 27.0%, while DTPA-Zn was 20.8% after composting for 63 days. This indicated that composting of pig manure for 63 days was more efficient in reducing the bioavailability of Cu than Zn.

The Distribution of Heavy Metal in HA and FA

The levels of HA and FA fractions in pig manure compost increased after composting (Table 2). HA fraction increased from 2.05% to 3.79% after 63 days composting, but increase in FA fraction was very minor from 0.49% to 0.62%.

Distribution of Cu and Zn in HA and FA are presented in Table 2. Content of Cu in HA fraction increased significantly from 134 mg/kg to 433 mg/kg,. The increase of Zn in HA fraction was less obvious, which increased from 113.7 mg/kg to 170.4 mg/kg. In case of FA fraction, Zn increased from 601.7 mg/kg to 1703 mg/kg but no significant change of Cu concentration in FA fraction was observed after composting. It was suggested that HA fraction had more affinity towards Cu for stable complex formation as compared to Zn, which showed more affinity towards FA fraction for stable complex formation.

Change of Gel Chromatogram of HA and FA

Figure 7: Gel chromatographic diagram of HA and FA at 280 nm
Figure 8

The change of different molecular size fractions in HA and FA during composting were observed by gel chromatography. For HA chromatogram (Figure 7a), a great increase in relative absorbance for HA was observed at an elution volume of 120 ml for sample collected on 63 days as compared to that at day 0. It showed a dominant high peak and an evident shoulder of day 63, corresponding to the void elution volume (V0) being the eluted fraction with a nominal molecular weight higher than 5000 Da (excluded fraction). It showed that HA had a higher molecular weight than 5000 Da, it increased significantly after composting of 63 days. The FA chromatogram, on the other hand, showed two evident fractions (Figure 7b). The peak relative to the included fraction at an elution volume of 180 ml representing lower molecular size fraction was more than 60% of the total, similar to the results as reported by Giusquiani et al. (1998). However, there was no significant change of relative absorbance at elution volume of 120 ml and 180 ml in FA, as increasing the composting time from day 0 to day 63, indicating that there was no obvious increase in the lower molecular size fraction during the composting process.

The extraction of humic-like substances followed by gel chromatography associated with UV detection is a suitable method to follow the evolution of organic matter during composting. Absorbance at an elution volume of 120 ml indicated the existence of higher molecular weight compounds in these fractions, while lower molecular weight compounds in the fractions at elution volume of 180 ml. HA fraction had an apparent molecular weight higher than 5000 Da, as expect, while FA fraction had a lower molecular weight value. Moreover, the relative absorbance value increased in high molecular weight fraction at 120 ml of HA after a period of maturation, which related to the increase in Na2O4P7extractable humic acid, representing the humification of composting. And also, increase in UV-280nm indicated a relative increase in compounds with phenolic and benzene-carboxylic groups (Stevenson, 1994) and gave a measure of the aromaticity of HA. It can be concluded that more phenolic and benzen-carboxylic groups in HA after composting, but no great increase in FA. These results corresponded to the yield of HA and FA discussed above. The observations demonstrated the increase of numerous of compounds with high molecular weight and the increase in stable complexity of organic materials with increasing maturation level.

Behaviour of Cu and Zn in HA and FA

Figure 8: Diagram of distribution of Cu in HA-0, HA-63
FA-0, and FA-63 elution fractionation associated with UV
absorbance at 280 nm
Figure 8

Humic substances are some of the most powerful metal-binding agents among organic substances, because of the large amounts of functional groups, such as carboxyl, phenolic, alcoholic hydroxyl, carbonyl, and quinones (McCarthy et al., 1990). The elution diagrams (Figure 8) showed that large amount of Cu was associated with the major humic acid peak at wavelength 280 nm after composting for 63 days, i.e. the copper and humic acids were both eluted at 120 ml. It also shows that the peak of UV-absorbance and content of Cu was overlapped smoothly. It indicated that Cu was more favourably bound to humic acids, the high molecular weight fraction of humic acids seemed to have a greater Cu holding capacity than the lower molecular weight ones because of a relatively higher functional groups content. The same results obtained from metal distribution in HA and FA has been discussed above. He et al. (1995) found that more Cu was recovered in HA than in FA in both Na4P2O7and NaOH extracts of municipal solid waste compost, whereas Hofstede and Ho (1991) found Cu in FA was predominant over HA. Although the metal distribution in humic substances has been studied widely, the results were inconsistent. Many factors, such as compost source, regent choice, and contents of extracting regents would likely be the major reseaons responsible for, the difference of metal distribution.

Figure 9: Diagram of distribution of Zn in HA-0, HA-63
FA-0, and FA-63 elution fractionation associated with
UV absorbance at 280 nm
Figure 9

The Zn elution pattern showed that most of the zinc associated with fulvic but not humic acids (Figure 9). The peak of UV-absorbance was not overlapped very well with the peak of Zn at 63 days of composting period, but with many peaks before the elution volume of 180 ml, which is the evident peak of FA. It indicated that zinc was not complexed very well with fulvic acid fraction although most of the Zn distributed in fulvic acid fraction.

Composting process significantly increased the contents of Cu in HA and Zn in FA, which may explain that organic matter is oxidised and soluble organic acids containing functional groups are formed during its biochemical decomposition in composting (Tab 2). Lake (1987) reported that the stability constant of Cu is higher than the Zn between metals and humic matter, The stability of complex formed by heavy metals with organic matter or humus normally depends upon the ionic radii, radius: charge ratio as well as its electro negativity. Cu is more reactive than Zn in terms of ionic radii, radius: charge ratio as well as its electro negativity The softness factor, which renders Cu a more soft acid than Zn accounts for the preferential formation of inner sphere complexes in case of Cu with humic acids as compared to Zn. In case of FA fractions, the complex formation is preferred with Zn due to balanced radius: charge ratio as compared to Cu (Mac Bride, 1994).


The C/N ratio decreased rapidly during the first 35 days and attained a constant value thereafter. In addition, the amount of total organic carbon and soluble organic carbon content decreased gradually and reached a satisfactory level after 63 days of composting. This indicated attainment of maturity level for the compost of pig manure after composting for 63 days. DTPA-extractable and chemical fractionation of heavy metal showed a reduction of bioavailability after composting for 63 days. The soluble form of heavy metals decreased, but residual form of heavy metals increased significantly along with the composting process. Gel chromatography showed an increase in high molecular weight with humification process. Complex organic compound formation & aromaticity increased in HA fraction as compared to FA fraction. after composting. It indicated that humification was benefited to high molecular weight fraction during composting. Gel chromatograms associated with the contents of metal in elution fractions showed that large amount of Cu distributed in HA fraction and the Zn was distributed in FA fraction. This showed that Cu bioavailability was affected by HA fraction, while in case of Zn, it was affected by the presence of FA fractions in pig manure compost.


This work was financially supported by a grant from the RBF fund. The authors would like to thank Prof. Chen at Kadoorie Farm and Mr. K. K. Ma in Hong Kong Baptist University for their excellent technical assistance throughout this project.



Copyright © 2005, ECO Services International