Journal of Tropical Oceanography >
Characterization of particle size distribution in the South China Sea basin during summer 2016
Received date: 2018-02-08
Request revised date: 2018-04-09
Online published: 2018-10-13
Supported by
National Natural Science Foundation of China (41576030, 41431176, 4176045, 4176044, 41376042)
Open Research Program of the State Key Laboratory of Tropical Oceanography (LTOZZ1602)
Science and Technology Program of Guangzhou, China (201504010034, 201707020023, 201607020041)
Science and Technology Planning Project of Guangdong Province of China (2016A020222008)
Strategic Priority Programme of the Chinese Academy of Sciences (XDA11040302)
Copyright
Particle size distribution (PSD) describes the relationship between particle concentration and particle size, influencing marine ecosystem environment, the optical properties of sea water, and so on. Based on the in-situ profiles of biological and optical properties during summer 2016 in the South China Sea basin, characterization of PSD was studied. The power-law model was fit to describe the PSD, and the results indicated that the mean coefficient of determination between in-situ PSD and simulated PSD could reach 0.95 in the logarithmic space. The PSD slope (ξ) ranged in [1.27, 7.65] with a mean of 3.93±0.56. The mean of ξ in the surface water of the South China Sea basin was similar to the mean of ξ in global ocean surface water, but higher than that in the surface water of other areas such as the bay. There was a strong negative relationship between ξ and the mean diameter (DA). Taking section T1 as an example, we analyzed the mean PSD profile of these stations. The features of PSD profiles are as follows. 1) At the surface, there were high values of ξ with low values of DA because the dominate particle was pico-phytoplankton. 2) The minimum value of ξ appeared in the subsurface chlorophyll maximum layer (SCML) with higher DA, which may result from the higher proportion of big phytoplankton. 3) In the twilight layer, the values of ξ as well as the values of DA were between those at the surface and in the SCML. This phenomenon may be related to the process of flocculation, decomposition, and settlement of phytoplankton. The traits of PSD would influence the inherent optical properties (IOP) of seawater. We found that both particulate beam scattering coefficient at 532 nm (bp(532)) and particulate beam backscattering coefficient at 532 nm (bbp(532)) would be higher in the SCML because of the increasing chlorophyll-a concentration. However, the lowest mean bp(532) and the lowest mean bbp(532) were found in the twilight layer. Furthermore, although the relationship between ξ and the particulate beam attenuation sepctral slope was weak, the model of
ZHENG Wendi , ZHOU Wen , CAO Wenxi , WANG Guifen , DENG Lin , XU Wenlong , XU Zhantang , LI Cai , CAI Jiannan . Characterization of particle size distribution in the South China Sea basin during summer 2016[J]. Journal of Tropical Oceanography, 2018 , 37(5) : 74 -85 . DOI: 10.11978/2018017
Fig. 1 The optical survey stations in the South China Sea basin during summer 2016图1 2016年南海海盆光学调查站位图 |
Fig. 2 The power-law relationship between the line height absorption at 676 nm (aLH(676)) and in-situ Chl-a. The sample number is 85. The black curve is the fitting of this study图2 吸收线性高度aLH(676)与实测Chl a的幂指数关系 |
Tab. 1 The standard of water stratification表1 分层标准 |
ζ值范围 | 分层 |
---|---|
小于0.5 | 表层 |
[0.5, 1.3] | SCML |
大于1.3 | 弱光层 |
Fig. 3 The modeling performance of PSD in the South China Sea basin using the power-law model (a) and mean absolute percent errors (MAPE,%) in the modeled PSD (b) for total dataset and three water layers图3 南海海盆区域总体及各水层PSD幂律模型拟合效果(a)和平均相对误差绝对值(MAPE,%)随粒径变化的关系图(b) |
Fig. 4 Distribution of total ξ (a), distributions of ξ in three water layers (b) and the relationship between total ξ and DA (c). (a) The black curve is the Gaussian fitting curve. (b) The black curve represents ξ in the surface layer (N=881). The blue curve represents ξ in the SCML (N=1422). The red curve represents ξ in the twilight layer (N=2573). (c) The black curve represents the power-law fitting curve of ξ and DA图4 南海海盆总体ξ值的分布情况(a)、三个水层的ξ值分布情况(b)和总体ξ值与DA的关系(c) |
Tab. 2 Ranges of ξ in different regions表2 不同海区的ξ值分布情况 |
海域 | ξ值分布范围 | ξ值平均值(±方差) | 参考文献 | |
---|---|---|---|---|
哈德逊湾 | 表层 | 2.84~4.46 | 3.63(±0.40) | Xi et al, 2014 |
黄、渤海 | 表层 | 2.72~4.46 | 3.64(±0.38) | Qiu et al, 2016 |
美国东西海岸及南大洋的大西洋区域 | 表层 | 2.7~4.7 | 3.63 | Buonassissi et al, 2010 |
全球大洋(遥感反演) | 表层 | 3.04~5.99 | 4.22(±0.59) | Kostadinov et al., 2009 |
南海中部 | 表层 | 2.65~6.88 | 4.54(±0.62) | 本文 |
SCML | 1.27~7.65 | 3.57(±0.55) | ||
弱光层 | 2.67~4.90 | 3.92(±0.30) | ||
总体 | 1.27~7.65 | 3.93(±0.56) |
Tab. 3 The statistical table of distributions of DA, CHLLH, $\bar{n}\text{p}$, and inherent optical parameters表3 DA、CHLLH、$\bar{n}\text{p}$及固有光学参数的分布范围统计表 |
参数 | 单位 | 个数 | 平均值 | 最大值 | 最小值 | 标准差 | 变异系数 | |
---|---|---|---|---|---|---|---|---|
DA | μm | 表层 | 881 | 7.78 | 123.56 | 3.23 | 9.56 | 122.89 |
SCML | 1422 | 25.97 | 166.07 | 3.23 | 24.34 | 93.72 | ||
弱光层 | 2573 | 14.36 | 111.10 | 7.22 | 9.03 | 62.91 | ||
剖面 | 4876 | 16.56 | 166.07 | 3.23 | 16.57 | 100.07 | ||
CHLLH | mg·m -3 | 表层 | 647 | 0.17 | 0.32 | 0.03 | 0.06 | 35.71 |
SCML | 1289 | 0.26 | 0.44 | 0.11 | 0.06 | 24.63 | ||
弱光层 | 730 | 0.15 | 0.33 | 0.02 | 0.05 | 36.38 | ||
剖面 | 2666 | 0.21 | 0.44 | 0.02 | 0.08 | 37.6 | ||
$\bar{n}_{p}$ | 无 | 表层 | 621 | 1.01 | 1.05 | 1.002 | 0.009 | 0.87 |
SCML | 1282 | 1.03 | 1.16 | 1.005 | 0.024 | 2.34 | ||
弱光层 | 639 | 1.06 | 1.22 | 1.004 | 0.044 | 4.13 | ||
剖面 | 2542 | 1.03 | 1.22 | 1.002 | 0.033 | 3.2 | ||
cp(530) | m-1 | 表层 | 647 | 0.24 | 0.41 | 0.059 | 0.078 | 32.02 |
SCML | 1289 | 0.20 | 0.45 | 3.46×10-3 | 0.084 | 41.92 | ||
弱光层 | 730 | 0.08 | 0.19 | 5.36×10-4 | 0.046 | 59.01 | ||
剖面 | 2666 | 0.18 | 0.45 | 5.36×10-4 | 0.098 | 54.88 | ||
γ | 无 | 剖面 | 585 | 0.86 | 1.13 | 0.39 | 0.12 | 14.14 |
bp(532) | m-1 | 表层 | 647 | 0.24 | 0.41 | 0.062 | 0.078 | 32.17 |
SCML | 1289 | 0.20 | 0.45 | 6.15×10-3 | 0.084 | 41.85 | ||
弱光层 | 730 | 0.08 | 0.19 | 2.87×10-4 | 0.047 | 59.53 | ||
剖面 | 2666 | 0.18 | 0.45 | 2.87×10-4 | 0.097 | 54.86 | ||
bbp(532) | m-1 | 表层 | 928 | 6.37×10-4 | 1.13×10-3 | 3.8×10-4 | 8.84×10-5 | 13.88 |
SCML | 1427 | 6.59×10-4 | 1.08×10-3 | 3.45×10-4 | 1.17×10-4 | 17.67 | ||
弱光层 | 2616 | 4.13×10-4 | 1.33×10-3 | 1.75×10-4 | 1.00×10-4 | 24.25 | ||
剖面 | 4971 | 5.25×10-4 | 1.33×10-3 | 1.75×10-4 | 1.57×10-4 | 29.95 | ||
$\widetilde{b}_{bp}$ | 无 | 表层 | 628 | 0.0031 | 0.01 | 0.0012 | 0.001 | 48.38 |
SCML | 1289 | 0.0042 | 0.11 | 0.0013 | 0.004 | 98.85 | ||
弱光层 | 729 | 0.0148 | 0.56 | 0.0022 | 0.038 | 260.28 | ||
剖面 | 2646 | 0.0069 | 0.56 | 0.0012 | 0.021 | 306.24 |
变异系数=100%×标准差/平均值 |
Fig. 5 The vertical profiles of bio-optical and hydrological parameters along section T1. a) CHLLH; b) potential density; c) ξ; d) DA; e) bbp(532); f) bp(532). The white solid line represents the depth of maximum CHLLH图5 T1断面生物光学参数及水文参数的剖面等值线分布图 |
Fig. 6 The mean vertical profiles of DA, ξ, and CHLLH in the South China Sea basin. The standard deviation of ξ is displayed for every 0.1 ζ图6 南海海盆DA、ξ与CHLLH平均剖面图 |
Fig. 7 Relationship between γ and ξ (a), and the distribution of in-situ ξ and modeled ξ (b). The selected data should match the following conditions: (1) PSD and cp(λ) power-law function fitting R2 are both greater than 0.9; (2) the range of ξ is [2.5, 5], and the range of $\bar{n}\text{p}$is [1.02, 1.2]. The model of ξ is ξ=γ+3-0.5exp(-6γ)图7 γ与ξ的关系(a)和实测与估算的ξ值分布范围(b) |
The authors have declared that no competing interests exist.
[1] |
|
[2] |
|
[3] |
|
[4] |
|
[5] |
|
[6] |
|
[7] |
|
[8] |
|
[9] |
|
[10] |
|
[11] |
|
[12] |
|
[13] |
|
[14] |
|
[15] |
|
[16] |
|
[17] |
|
[18] |
|
[19] |
|
[20] |
|
[21] |
|
[22] |
|
[23] |
|
[24] |
|
[25] |
|
[26] |
|
[27] |
|
[28] |
|
[29] |
|
[30] |
|
[31] |
|
[32] |
|
[33] |
|
[34] |
|
[35] |
|
[36] |
|
[37] |
|
[38] |
|
[39] |
|
[40] |
|
[41] |
|
/
〈 | 〉 |