Quantifying heterogeneous responses of fish community size structure using novel combined statistical techniques.
Quantifying heterogeneous responses of fish community size structure using novel combined statistical techniques.
复制标题
DOI：
10.1111/gcb.13190
复制
发表时间：
2016-05
影响因子：
11.6
通讯作者：
Blanchard JL
中科院分区：
文献类型：
--
作者：
Marshall AM;Bigg GR;van Leeuwen SM;Pinnegar JK;Wei HL;Webb TJ;Blanchard JL
为了了解生态系统的变化，必须确定研究这些变化的适当尺度。大型海洋生态系统覆盖数千平方公里，是生态系统监测和评估的一个有用分类办法。然而，在大海洋生态系统之间取平均值可能会掩盖其内部复杂的动态。本研究的目的是在数学上确定本地和区域模式的生态变化在LME使用经验正交函数（EOFs）。在使用EOFs定义具有不同变化模式的区域后，应用源自控制理论的统计模型（具有外源输入的非线性自回归移动平均- NARMAX）来评估这些区域内的潜在变化驱动因素。我们选择了北海渔业研究调查（1980-2008年）中鱼类丰度的空间数据集（0.5°纬度× 1°经度）以及温度，氧气，净初级生产量和捕捞压力代理，我们应用了NARMAX和NARMAX方法。自1980年以来，有两个区域发生了显著变化：北海中部的群落规模结构减少，NARMAX模型认为这与渔业的变化有关;挪威海沟区域的群落规模结构增加，NARMAX结果表明，这主要与海底温度的变化有关。这些地区进行了比较，没有变化的区域沿着苏格兰东部海岸的模型确定的社区规模结构是最密切相关的净初级生产。这项研究强调了北海不同区域环境变化和捕捞压力的多方面影响。此外，通过强调这种空间异质性的社区规模结构的变化，重要的本地空间动态往往被忽视，当北海被认为是一个大规模的，同质的生态系统（通常是在政治海洋战略框架指令的情况下）。
To understand changes in ecosystems, the appropriate scale at which to study them must be determined. Large marine ecosystems (LMEs) cover thousands of square kilometres and are a useful classification scheme for ecosystem monitoring and assessment. However, averaging across LMEs may obscure intricate dynamics within. The purpose of this study is to mathematically determine local and regional patterns of ecological change within an LME using empirical orthogonal functions (EOFs). After using EOFs to define regions with distinct patterns of change, a statistical model originating from control theory is applied (Nonlinear AutoRegressive Moving Average with eXogenous input – NARMAX) to assess potential drivers of change within these regions. We have selected spatial data sets (0.5° latitude × 1°longitude) of fish abundance from North Sea fisheries research surveys (spanning 1980–2008) as well as of temperature, oxygen, net primary production and a fishing pressure proxy, to which we apply the EOF and NARMAX methods. Two regions showed significant changes since 1980: the central North Sea displayed a decrease in community size structure which the NARMAX model suggested was linked to changes in fishing; and the Norwegian trench region displayed an increase in community size structure which, as indicated by NARMAX results, was primarily linked to changes in sea‐bottom temperature. These regions were compared to an area of no change along the eastern Scottish coast where the model determined the community size structure was most strongly associated to net primary production. This study highlights the multifaceted effects of environmental change and fishing pressures in different regions of the North Sea. Furthermore, by highlighting this spatial heterogeneity in community size structure change, important local spatial dynamics are often overlooked when the North Sea is considered as a broad‐scale, homogeneous ecosystem (as normally is the case within the political Marine Strategy Framework Directive).