南海北部次表层叶绿素最大值年际变化特征分析

次表层叶绿素最大值(subsurface chlorophyll maxima, SCMs)广泛存在于全球各海域, 该最大值层往往具有较高的海洋初级生产力和新生产力, 因此研究其年际变化特征对深入理解气候变化影响下海洋生态系统变化有重要意义。本文采用一维物理-生态耦合模型(one-dimensional physical-biological coupled model, MEM-1D)较好地模拟了1994—2019年南海北部海盆区海温、盐度、营养盐和叶绿素的垂向分布, 并采用3种统计方法, 分别从整体趋势、不同时间尺度及显著变化三方面分析了SCMs特征因子(强度、深度和厚度)的年际变化特征。总体而言1994—2019年SCMs强度整体减小趋势较弱(趋势斜率S<0), 具体表现为先减小(1994—2004年)后增大(2005—2012年)再减小(2013—2019年), 其中1999—2004年显著变小; SCMs深度呈变深趋势(趋势斜率S>0), 1994—2011年逐渐变深, 之后逐渐变浅, 但变化不显著; SCMs厚度整体呈增大趋势, 1999年起显著变大。相关分析发现, 海表面温度在年际变化上与SCMs特征因子间不存在相关性(P>0.05); 海表面温度对SCMs的影响主要表现在季节尺度上, SCMs深度和强度均与海表面温度呈一致性变化。季节性差分自回归滑动平均模型对SCMs三个特征因子的拟合效果较好, 平均绝对百分比误差分别为5.33%(强度)、0.62%(深度)、2.49%(厚度), 模型可用于对SCMs特征因子变化趋势的预测。

Interannual variation of subsurface chlorophyll maximum in the northern South China Sea

Subsurface chlorophyll maxima (SCMs) are widely found in the global oceans. The layer of SCM accounts for a great proportion of marine primary productivity and new productivity. Studying the interannual variation of SCMs can deepen our understanding of marine ecosystems under climatic change. In this study, a one-dimensional physical-biological coupled model (MEM-1D) was used to simulate the vertical distribution of ocean temperature, salinity, nutrients, and chlorophyll in the northern South China Sea from 1994 to 2019. Based on numerical modeling results, we analyzed the interannual variations of SCM characteristics (intensity, depth and thickness) using three statistical methods from different aspects, i.e., overall trend, different time scales and significant changes. In general, the intensity of SCMs from 1994 to 2019 shows a slightly decreasing trend (the trend slope S<0). Specifically, the SCM intensity decreased from 1994 to 2004 ( decreased significantly from 1999 to 2004), while it increased from 2005 to 2012 and then decreased until 2019. The depth of SCMs showed a deepening tendency between 1994 and 2019 (the trend slope S>0), gradually becoming deeper from 1994 to 2011, and then shoaling; but the changing trend was not significant. The thickness of SCMs increased as a whole, and it had increased significantly since 1999. Correlation analysis showed no relationship between annually averaged sea-surface temperature (SST) and the characteristics of SCMs (P>0.05). The influence of SST on SCMs was mainly on the seasonal scale, and the depth and intensity of SCMs changed consistently with SST. The seasonal, autoregressively integrated moving average model has a good fit for the three characteristics of SCMs, with the mean absolute percentage errors of 5.33% for intensity, 0.62% for depth and 2.49% for thickness, indicating that the model can be used to predict the trend of SCM characteristics.