南海北部表层颗粒有机碳的季节和年际变化遥感分析

海洋颗粒有机碳（POC）是海洋固碳的一个关键参数。为了研究南海北部陆架及海盆表层POC浓度的时空分布特征以及变化趋势，本文利用2009-2011年4个季节的实测数据，对NASA发布的MODIS/AQUA卫星月平均POC遥感产品，进行了验证和校正；并利用校正后的遥感数据分析了2003-2014年POC的时空分布特征和变化趋势。发现POC遥感产品与南海北部实测数据具有较好的线性关系（R2=0.72），但存在系统性偏高，需利用实测数据对遥感数据进行区域性校正。分析校正后的遥感数据发现，南海北部陆架POC浓度较高，平均为（33.34±8.02）mg/m3；吕宋海峡西南海域浓度较低，平均为（29.25±6.20）mg/m3；中央海盆区浓度最低，平均为（27.02±4.84）mg/m3。春夏季POC浓度较低，最低值一般出现在5月，冬季（12月至翌年1月）POC浓度达到最高。利用2003-2014年的长时间序列遥感叶绿素（Chl a）和海表温度（SST）、混合层深度（MLD）模式数据，以及实测数据对南海北部POC浓度的影响机制进行了分析。发现POC与Chl a在秋冬呈现较好的相关关系（R2=0.51），但在春夏季较离散，表明秋冬季生物作用对POC影响较大。2003-2014年期间，POC与Chl a、MLD及SST存在明显的年际变化，但并没有显著的上升或下降趋势。

Satellite views of the seasonal and interannual variations of the particulate organic carbon in the northern South China Sea

The particulate organic carbon (POC) plays an important role in the marine carbon cycle. It transports the organic carbon from the surface layer to the deep sea, and consequently removes the carbon dioxide (CO₂) from the air, thus it is an important component of biological pump. To study the temporal and spatial distribution and long-term variation of sea surface POC concentration in the northern South China Sea (nSCS), we used the in situ data of four cruises from June 2009 to May 2011 to validate and calibrate MODIS/AQUA satellite monthly average POC products. The results shown the satellite-derived POC and in situ data had a good linear relationship in the nSCS (R²=0.72), but with a systematic bias. With the validated satellite POC products, we found that POC concentration in the shelf of the bias was high (33.34±8.02) mg/m³; the POC values of the southwest area of Luzon Strait was low (29.25±6.20) mg/m³, and POC had lowest values in the northern basin of SCS (27.02±4.84) mg/m³. The POC concentration was lower in spring and summer than that in winter and autumn; the lowest values generally appeared in May, and it remained a relative low and stable value from June to August, then the POC concentration began to rise in September, and reached the highest value in December. We analyzed the controlling factors on the POC distribution in the nSCS using the satellite-derived chlorophyll a (Chl a) concentration, sea surface temperature (SST), the model-derived mixed layer depth (MLD), as well as the in situ Chl a and total suspended matter concentration (TSM). The results demonstrated that the POC and Chl a in the autumn and winter had a relatively good correlation (R²=0.51), but it was dispersed in the spring and summer, indicating that Chl a had significant influence on POC variation in the nSCS in autumn and winter. The POC, Chl a, MLD and SST had the significant inter-annual variability, but had no significant increase or decrease trend in the period of 2003-2014. The mechanism of controlling POC spatial and temporal variation was complex.