Estimation of primary production in the South China Sea and observation of particulate organic matter at SEATS station: application in the validation and improvement of the coupled physical-biogeochemical model of South China Sea
Date Issued
2004
Date
2004
Author(s)
Chen, Ying-Jie
DOI
zh-TW
Abstract
Though the existing coupled physical-biogeochemical model of the South China Sea (Liu et al., 2002) has achieved reasonable success in predicting the temporal variation and spatial distribution of chlorophyll and primary production, there are still plenty of room for improvement. In the first place, the magnitudes of modeled primary production are considerably smaller than those derived from the ocean color data. Secondly, some biogeochemical variables of the model, such as the particulate organic carbon (POC), have not been compared with observed data yet. The major objectives of this research are to better understand the performance of the model by comparing modeled primary production and POC with data from direct observations or remote sensing, and to improve the model by modifying the algorithm or the parameters.
In order to achieve the objectives above, this study has accomplished the following items: (1) to estimate primary production from the ocean color data and to compare the results with the observed data to acquire more precise estimates of primary production in the South China Sea; (2) to collect and analyze the POC samples from the SEATS (South-East Asia Time-series Study) station and to establish the vertical distribution and seasonal variation of POC at the SEATS station; (3) to validate and improve the modeled temporal and spatial distribution of primary production and POC according to the results above.
According to the comparison between the primary production results derived from the SeaWiFS ocean color data by a biological-optical model and the observed data, the former are in reasonable agreement with the observed data in the deep zone (depth > 50 m), but significantly overestimated in the coastal zone (depth < 50 m). Therefore, we reduce the primary production values in the coastal zone by 1/2 and obtained the estimated average annual PP of 412 mg C m-2 d-1 for the South China Sea.
The vertical distribution of POC at the SEATS station reveals the following pattern. The POC concentration fluctuates between 1 and 10 uM in the top 200 m and drops to around 1 uM in the water column below. The standing stock of POC (0-100 m) shows considerable seasonal variation with high values (250-350 mmol C m-2) occurring between October and March and low values (100-200 mmol C m-2) occurring between May and July. The standing stock of chlorophyll (0-30 m) at SEATS station also reaches a peak in winter.
Comparing the modeled results with the observations described above, we have found quite a few shortfalls of the model, which need to be improved. In the case of primary production, modeled and ocean color derived primary productions in the South China Sea both show a stronger peak in winter and a weaker peak in summer, but the average annual value of modeled primary production is much smaller. Further comparison has revealed that the modeled result seriously underestimates the primary production in the coastal zone (depth < 100 m). This underestimation results from the setup of the model that the detritus is assumed to be buried and, therefore, removed from the system as it reaches the seafloor, and, hence, the model fails to simulate the high nutrient conditions in the coastal zone. Therefore, the model is modified to include the process that the detritus is decomposed to release nutrients. This improves the performance of the model in the coastal zone considerably and makes the spatial distribution of modeled primary production more reasonable (coastal zone > deep zone).
Regarding the vertical distribution and seasonal variation of POC at the SEATS station, the modeled results exhibit the basic pattern of observed distributions, but the magnitude is a little too small. Because decreasing the sinking velocity of detritus can increase the residence time of detritus, we have adopted this method to improve the simulation of the POC distribution. As a compromise to optimize both the modeled POC and the modeled chlorophyll, we have chosen 7.5 m d-1 as the sinking speed for modeling POC. This decrease of sinking speed does not affect the modeled POC flux much.
By modifying both the model setup and the sinking speed at the same time, we have found that the former primarily affects the coastal zone and the latter primarily affects the deep zones with relatively little cross effects. Therefore, we suggest that the model may include both mechanisms for better performance. In the future we may further improve the model by modifying the model structure and importing more detailed mechanism of nutrient regeneration from sediments.
Subjects
初級生產力
耦合模式
南海時間序列研究
南海
顆粒態有機碳
coupled model
particulate organic carbon(POC)
South East Asia Times-series Station(SEATS)
South China Sea
primary production
Type
thesis
File(s)![Thumbnail Image]()
Loading...
Name
ntu-93-R91241408-1.pdf
Size
23.53 KB
Format
Adobe PDF
Checksum
(MD5):ace50fafb9b7bf9784e7dce2dd8fda40