Скачать 0.56 Mb.
3.1. Hierarchical Cluster Analysis
All 92 samples were subjected to Hierarchical Cluster Analysis (using SPSS for Windows 15.0) based on winter salinity and sand concentration. The cluster analysis defined 9 groups (G1-G9) with a clear geographical control (illustrated in Fig. 2; Table 5). Surface sediment samples are from the 91 sampling sites (without site PE43) covering G1-G9. The POC samples are only from the western estuary where suspended sediment concentration was high enough to allow measurements to be made, with samples ranging from G2-G6 for winter POC and G2-G5 for summer POC.
The 9 groups defined by selected environment variables clearly highlight a trend from the full-freshwater environment (G1-G3) to the full-marine environment (G8-G9), via brackish areas (G4-G7) (Fig. 2; Table 6). The freshwater groups have low annual average water salinity of 3.4±3.6, which increases to 15.8±5.3 and 28.4±3.1 in the brackish-water areas and reaches 32.9±0.8 under the full marine environment (Table 6). Mean water depth is lower than 5 m in the freshwater area, getting deeper seawards, and becomes deeper than 20 m in the shallow marine area. Meanwhile, average particle size gets finer offshore with a reduction in sand component. All variables tend to be more stable in the marine area than in the freshwater area (Table 6).
In addition to the spatial diversity of the environmental conditions, seasonal variability is also observed. For G2-G5, mean winter salinity is 14.0±11.6, ranging between 0.1 and 33.9. This is significantly higher than that in summer of 3.1±4.8, ranging between 0.1 and 18.2 (Table 3). A similar situation is found in the concentration of total dissolved solids (TDS) of 15.5±12.1 g l-1 in winter compared to 2.0±2.7 g l-1 in summer. Areas under the freshwater- and fresh-brackish conditions are more sensitive to seasonal changes than those under marine- and marine-brackish conditions, as seasonal variation of these environmental parameters are larger for G2 than G5.
3.2. Terrestrial organic matter
Plant samples collected in this study are composed of two main groups: C3 plants (dominant vegetation in the Pearl River delta), and some C4 plants. Results show that C3 plants in general have lower δ13C with an average of –29.0±1.8‰ compared to C4 plants with an average of –13.1±0.5‰ (Fig. 3a). C3 plants generally have the lowest δ13C of –29.9±1.3‰, followed by the agricultural C3 plants (e.g. rice, lotus and banana) and mangroves, with δ13C values of –28.2±1.4‰ and –27.1±1.7‰ respectively (Fig. 3b). C4 plants (mostly grasses) have δ13C of –13.2±0.5‰, slightly lower than sugarcane which has δ13C of –12.7±0.2‰ (Fig. 3b).
C/N of C3 and C4 plants are widely variable (7.4-61.8 for C3 plants and 8.2-40.3 for C4 plants) and overlap significantly, with the average value of 22.7±11.6, and 24.6±9.4 respectively (Fig. 3a). It is impossible to distinguish between the two plant types based on C/N alone. Within the C3 plants, agricultural plants have lower C/N (13.7±2.7) than other C3 plants (21.7±10.7) and mangroves (31.0±12.5). Within the C4 plants, sugarcane has a slightly higher C/N of 30.6±8.4 compared to other C4 plants of 21.5±8.9 (Fig. 3b).
3.2.2. Terrestrial soil
Terrestrial soil samples have δ13C between –28.9‰ and –19.3‰, and C/N between 7.7 and 20.4, with a negative correlation between δ13C and C/N (Fig. 4a). As expected soil samples from different areas of vegetation cover, have different δ13C and C/N values. Soil samples from pine forest have the lowest δ13C of –28.3±0.8‰, but the highest C/N of 17.9±3.6. Soil samples from riverbanks and mangrove areas have similar δ13C and C/N, around –24.1‰ and 12.5 respectively, although both δ13C and C/N are relatively more variable in the mangrove soil than in the riverbank soil (Fig. 4b). Soil samples from farmlands have the highest δ13C of –21.7±0.7‰ and the lowest C/N of 8.9±1.1 (Fig. 4b).
3.3. Estuarine particulate organic carbon
3.3.1. POC collected on filter paper
Overall, δ13C of POC shows very little if any seasonal variations, the average δ13C of POC in summer (–24.5‰) is within error of winter (–24.2‰) (there is slightly more variation in winter (SD=1.9) than in summer (SD=1.0)). C/N of the winter POC ranges between 6.0 and 9.0, with a mean value of 7.1±0.9. It is slightly lower than the summer POC, ranging between 8.0 and 10.2, averaging 9.2±1.1 (Fig. 5).
δ13C and C/N of POC varies between groups from different environments. In winter, freshwater POC has the lower δ13C (–27.6±0.8‰, G2) than the marine-brackish POC (–22.4±0.5‰, G6), while samples from the fresh-brackish water area have δ13C between the two (–24.5±1.3‰ for G4; –23.6±1.2‰, G5) (Fig. 6a). Variation in C/N of winter POC between groups is not as significant as δ13C (Fig. 6a). In summer, the average δ13CPOC of each group is similar and does not show any trend between groups. Differences in both δ13C and C/N of summer POC is not significant (Fig. 6b).
δ13C and C/N of POC show good correlation between salinity and other environmental conditions (Fig. 6c, d), especially in the winter. In winter, δ13C increases from –27.6‰ in the freshwater area further inland to around –21.6‰ in the marine-brackish area near the estuary mouth, with increasing salinity (Fig. 6c). Correlation between C/N and environmental variables is not as significant as that of δ13C. C/N is found to be high in the middle estuary (7.6±1.0, G5) and decreases landwards and seawards (<7.0). In summer, both δ13C and C/N are more uniform than winter and not clearly correlated to the environmental variables (Fig. 6d).
3.3.2. POC collected by a 70 µm net
Estuarine POC is composed of organic carbon from fine-size plankton and relatively coarse TOM. As the size of plankton (e.g. diatoms) is usually smaller than 80 μm, organic matter collected on filter paper has a higher proportion of plankton/algae, while those collected in a 70 µm net presents higher proportion of TOM over plankton. Results show that the POC samples from the 70 µm net have a lower average δ13C of –25.6±1.0‰, compared to the filter paper sample from the same location (i.e. –24.3±1.0‰ in summer and –24.2±1.5‰ in winter, Fig. 7). The average C/N of the POC from the net samples is 10.7±4.5, which is higher and more variable than that of the POC collected on filter paper of 9.4±1.1 in summer and 7.8±1.0 in winter respectively (Fig. 7). POC values from the samples collected using the net fall into two groups. POC from locations further seaward, such as PE65, PE67, PE68, and PE71 (Fig. 1b), tend to have lower δ13C and C/N than POC on filter paper from both seasons, while those from the inner estuarine area have similar δ13C with the POC from the filter papers but higher C/N (Fig. 7).
3.4. Estuarine surface sediment
The estuarine surface sediment samples have δ13C ranging from –27.0‰ to –20.8‰ and C/N ranging between 20.1 and 6.5, with δ13C and C/N negatively correlated (Fig. 8a). With the environmental change from the full-freshwater environment to the full-marine environment from G1 to G9, a clear trend is observed in both δ13C and C/N values. δ13C becomes progressively higher from G1 to G9, increasing from –25.0±1.3‰ in G1 to –21.0±0.2‰ in G9 (Fig. 8), with increasing water salinity and decreasing sand concentration (Table 5). Meanwhile, C/N also shows a clear trend decreasing from 15.2±3.3 in G1 to 6.8±0.2 in G9. G3 does not fit in this trend with δ13C of –23.0±1.6‰ and C/N of 11.1±1.3, which is lower than G4 (Fig. 8b). Both δ13C and C/N are more stable in the marine environment compared to the freshwater (Table 5).
4.1. Terrestrial organic matter and its sources
Plants from the study area can be placed in two main groups, C3 and C4 plants, based on the way they fix carbon from atmospheric CO2 during the photosynthetic process. This process produces significant difference in the δ13C signature of the organic carbon from these two groups of plants. C3 plants use the Calvic-Benson cycle (C3 pathway) (Craig, 1953; Park and Epstein, 1960), through which they absorb and conserve 12C from atmospheric CO2 more effectively compared to the Hatch-Slack pathway used by C4 plants (Hatch and Slack, 1970). This results in greater discrimination against 13C in C3 plants than C4 plants, with C3 plants producing lower δ13C around –28.1‰ (Graig, 1953; Fry and Sherr, 1984; O’Leary, 1985) than C4 plants with values around –13.0‰ (Bender, 1971; Fry and Sherr, 1984; Emerson and Hedges, 1988). The C3 plants analyzed in this study have an average δ13C of –29.9‰ (Fig. 3), about 1.8‰ lower than values reported by other studies. This might be due to the high precipitation in the Pearl River delta area, with precipitation over 1500 mm/yr (Li et al., 1990). A similar case has been reported by Austin and Vitousek (1998) where δ13C of C3 plants shifts from –29.9‰ to –25.6‰, when the sampling site changed from an area of 5000 mm/yr precipitation to an area of only 500 mm/yr. The average δ13C of C4 plants from this study is –13.1‰ (Fig. 3), similar to other studies (e.g. Bender, 1971; Fry and Sherr, 1984; Emerson and Hedges, 1988).
Terrestrial soil is one of the direct sinks of the organic carbon from the surface vegetation, and its organic carbon signature depends greatly on the dominant vegetation type. This explains the variation in organic carbon isotope signatures found in soil samples from different areas of the Pearl River delta, e.g. forest, riverbank and tidal flats (Fig. 4). Soils from the forest mainly receive organic carbon from C3 plants (e.g. pine), and tend to have low δ13C of –28.3±0.8‰, but high C/N of 17.9±3.6 (Fig. 4b). Riverbank soil samples are from the flood plain along the North River, dominated by shrubs and grass. The increasing proportion of C4 grass on the flood plain produces relatively higher δ13C in the riverbank soil (–24.1±1.0‰) compared to the forest soil (Fig. 4b). Similar δ13C and C/N are found in the riverbank and mangrove soil (–24.0±1.9‰) and are due to the similar proportion of C4 plants growing on the flood plain and the tidal flat. Other studies also report the δ13C of soils from C4-dominated areas as being around –21.0‰, and around −24.0‰ from areas mixed by C3 and C4 plants (Driese et al; 2005; Saia et al. 2008). Many studies have used δ13C of terrestrial soil to reflect different vegetation types (Mackie, et al., 2005; Fan et al., 2007; Cao et al 2008; Driese et al., 2008; Saia et al., 2008). Usually, C/N of soil organic matter is lower than 20. Although the range of C/N is not as wide as the C/N of either C3 or C4 plants, it is still impossible to use C/N alone to distinguish different sources of the TOM.
4.2. Estuarine particulate organic carbon and its sources
In the Pearl River delta, terrestrial organic matter enters the river system in the form of plant fragments and soil organic matter, and mixes with in situ (freshwater, brackish-water, marine) plankton and algae within the estuary. Accordingly, within the estuarine area, the POC receives carbon from the TOM and in situ aquatic productivity (freshwater, brackish-water or marine plankton). The terrigenous freshwater POC, having δ13C around −28.7‰ (Fig. 5 and 6), is delivered into the estuarine system by the river runoff. It dilutes the signal produced by the brackish-water and marine POC which usually has δ13C around −21.2‰ and C/N ≤ 11.0 (Fig. 5 and 6). Higher δ13C in marine POC compared with riverine POC has also been reported from other coastal areas (Middelburg and Nieuwenhuize, 1998; Countway et al., 2006; Bianchi et al., 2007; Middleburg and Herman, 2007; Zhang et al., 2007; Bird et al., 2008). Wu et al. (2007) report δ13C of the POC varying from –24.4‰ in the Yangtze River to –21.0‰ on the East China Sea Shelf. Middleburg and Herman (2007) suggested that the correlation between δ13CPOC and salinity is more significant in river-dominated estuaries, such as the Rhine estuary and the Douro estuary, than in tidal-dominate estuaries such as the Gironde estuary and the Loire estuary, Western Europe.
In the Pearl River estuary, the balance between TOM and in situ plankton in the estuarine POC is strongly influenced by the strength of the freshwater runoff. It is, therefore, the strength of runoff that appears to control the spatial and temporal variation of the carbon isotopic signature of the estuarine POC. In winter, significant reduction in monsoonal precipitation results in a remarkable reduction in the TOM input into the estuary, and therefore the POC within the estuary is dominated by in situ phytoplankton (e.g. diatoms, dinoflagellates, green algae, euglenoides). δ13C of the winter POC shows a clear trend from freshwater areas (around –28.0‰) to the marine-brackish areas (around –22.0‰), reflecting the variation in δ13C and C/N of plankton under different environments (Fig. 6a). In the Pearl River estuary, δ13C of the freshwater POC typically ranges from –30‰ to –25‰, and the brackish POC typically ranges from –22‰ to –26‰ (Fig. 6c, d) which is also reported by other studies (Lamb et al., 2006; Middelburg and Herman, 2007; Tesi et al., 2007). The low C/N (7.1±0.9) of winter POC supports the assumption that winter POC is dominated by plankton, as the phytoplankton tends to have low C/N of 5-7 (Lamb et al., 2006). In summer the high freshwater flux brings larger amounts of TOM into the estuary (clearly seen in the amount of sediment collected by the filter in the summer compared to the winter in this study) resulting in a higher proportion of TOM in summer POC, suggested by the relatively higher C/N. At the same time the influence of the strong freshwater flux extends further seawards of the estuary, and produces a relatively uniform mix of POC within the estuary. This mixing and homogenisation of the POC throughout the estuary reduces the spatial variability in δ13C of the summer POC. This is indicated by the considerable overlap in δ13C and C/N of the summer POC from groups G2 to G5 across the estuary (Fig. 6b).
The POC collected with the 70 µm net is assumed to have a higher proportion of plant fragments than the POC collected on the filter paper as a higher proportion of the relatively smaller phytoplankton can pass through the net, but are trapped on the filter. The POC from the net would, therefore, be expected to record (on average) lower δ13C and have higher C/N than the POC from the filter paper. This is indeed the case (Fig. 7) with POC samples from the net having an average δ13C of –25.6‰ and C/N of 10.7 compared to values of –24.3‰ and 9.5 for samples from the filter paper. Previous studies suggest that the range of δ13CPOC is small in estuaries of high turbidity, between –24‰ and –26‰ (Fontugne and Jouanneau, 1987; Tan et al., 2004), which reflects the dominance of terrigenous POC input into the estuary over other sources. This further supports the suggestion that the proportion of TOM in the POC is one of the main factors influencing δ13C and C/N of POC, while the strength of the freshwater flux controls seasonal and spatial distribution of the TOM in the POC. However, during periods of reduced freshwater flux (during winter in the Pearl River estuary area for example) in situ productivity will tend to be the dominant source of POC and, hence, be the important control on spatial variability of δ13C of POC (illustrated by winter distribution of δ13C, Fig. 6a).
|Chem 6v39: Organic Electronics (Special Topics in Organic Chemistry)||I organic air pollutants I 1 Volatile Organic Compounds (vocs)|
|11/13/2007 -the following are letters written by organic consumers in response to his comments that buying organic and local is dumb. Collin Peterson, chairman||教科書為Advanced Organic Chemistry: Part B: Reaction and Synthesis (Advanced Organic Chemistry / Part B: Reactions and Synthesis)|
|Comparison of bulk Sea Surface and Mixed Layer Temperatures||Circulation of the Mediterranean Sea and its Variability|
|The climate dynamics of total solar variability||Asian Summer Monsoon Variability and Climate Change|
|Principles of inorganic and organic chemistry||Introduction to organic chemistry chem 106|