Scripps Institution of Oceanography, La Jolla, California.
Oregon State University, Corvallis, Oregon.
Glob Chang Biol. 2018 Jul;24(7):3065-3078. doi: 10.1111/gcb.14161. Epub 2018 May 3.
The western Antarctic Peninsula (WAP) is a bellwether of global climate change and natural laboratory for identifying interactions between climate and ecosystems. The Palmer Long-Term Ecological Research (LTER) project has collected data on key ecological and environmental processes along the WAP since 1993. To better understand how key ecological parameters are changing across space and time, we developed a novel seascape classification approach based on in situ temperature, salinity, chlorophyll a, nitrate + nitrite, phosphate, and silicate. We anticipate that this approach will be broadly applicable to other geographical areas. Through the application of self-organizing maps (SOMs), we identified eight recurrent seascape units (SUs) in these data. These SUs have strong fidelity to known regional water masses but with an additional layer of biogeochemical detail, allowing us to identify multiple distinct nutrient profiles in several water masses. To identify the temporal and spatial distribution of these SUs, we mapped them across the Palmer LTER sampling grid via objective mapping of the original parameters. Analysis of the abundance and distribution of SUs since 1993 suggests two year types characterized by the partitioning of chlorophyll a into SUs with different spatial characteristics. By developing generalized linear models for correlated, time-lagged external drivers, we conclude that early spring sea ice conditions exert a strong influence on the distribution of chlorophyll a and nutrients along the WAP, but not necessarily the total chlorophyll a inventory. Because the distribution and density of phytoplankton biomass can have an impact on biomass transfer to the upper trophic levels, these results highlight anticipated links between the WAP marine ecosystem and climate.
西南极半岛(WAP)是全球气候变化的晴雨表,也是识别气候与生态系统相互作用的天然实验室。自 1993 年以来,帕尔默长期生态研究(LTER)项目一直在该地区收集关键生态和环境过程的数据。为了更好地了解关键生态参数在空间和时间上的变化情况,我们开发了一种新的基于现场温度、盐度、叶绿素 a、硝酸盐+亚硝酸盐、磷酸盐和硅酸盐的景观分类方法。我们预计这种方法将广泛适用于其他地理区域。通过自组织映射(SOMs)的应用,我们在这些数据中确定了 8 个常见的景观单元(SU)。这些 SU 与已知的区域水团具有很强的一致性,但具有额外的生物地球化学细节层,使我们能够在几个水团中识别出多个不同的营养剖面。为了识别这些 SU 的时空分布,我们通过原始参数的目标映射,在帕尔默 LTER 采样网格上对它们进行了映射。自 1993 年以来,对这些 SU 的丰度和分布的分析表明,有两种年份类型,其特征是叶绿素 a 分配到具有不同空间特征的 SU 中。通过为相关的、时滞的外部驱动因素开发广义线性模型,我们得出结论,早春海冰条件对 WAP 沿线叶绿素 a 和营养物质的分布有强烈影响,但不一定对总叶绿素 a 储量有影响。由于浮游植物生物量的分布和密度可能会对生物量向上层营养级的转移产生影响,因此这些结果突出了 WAP 海洋生态系统与气候之间的预期联系。