Until the 1980s, the deep sea was generally considered to be a particularly stable environment, free from major temporal variations (Sanders, 1968). Studies in the abyssal northeast Atlantic by Billett et al. (1983), and subsequently Lampitt (1985) discovered seasonal pulses of surface primary production-derived particulate organic matter (phytodetritus), and hence carbon, at abyssal depths. These early observations were subsequently extended to the central oceanic region of the NE Atlantic (Pfannkuche, 1993; Thiel et al., 1989), and prompted the establishment of more concerted time series studies in the Porcupine Abyssal Plain area. Today, the Porcupine Abyssal Plain Sustained Observatory (PAP–SO) is a multidisciplinary open-ocean time series site in the NE Atlantic (48°50′N 16°30′W, 4850 m water depth; Fig. 1), focused on the study of connections between the surface and deep ocean. In situ measurements of climatically and environmentally relevant variables have been made for more than 30 years. This represents an exceptionally long time series - a recent compilation of biological time series data, across terrestrial, freshwater, and marine realms, indicates an average duration of only 13-years (Dornelas et al., 2018). Long-term time series in the deep sea are rare, particularly those collecting data from surface to seabed. The PAP-SO is one of two abyssal long-term time series sites globally (Smith et al. 2015), the other being a thirty-year time series at Station M in the northeastern Pacific Ocean (34°50′N, 123°00′W, ~4000 m water depth), maintained by the Monterey Bay Aquarium Research Institute (Smith et al., 2020). This ‘sibling’ abyssal time series site also aims to understand the connections between the surface ocean and the seabed, using many similar techniques (Smith et al., 2017), facilitating comparisons between the two sites (e.g. Durden et al., 2019; Durden et al., 2020a; Laguionie-Marchais et al., 2013; Smith et al., 2009). Another source of extended comparison is the 21 year time series Long-Term Ecological Research Observatory HAUSGARTEN, Frontiers in Arctic Marine Monitoring (FRAM) in the Fram Strait between the North Atlantic and the central Arctic Ocean (78.5°N–80°N, 05°W–11°E, 250–5500 m water depth), maintained by the Alfred Wegener Institute for Polar and Marine Research (Soltwedel et al., 2016; Soltwedel et al., 2005). Much of our understanding of temporal variation in the deep sea, and connections between the surface ocean and the seabed have been derived from research conducted at these observatories.

直到1980年代，普遍认为深海是一个特别稳定的环境，没有重大的时间变化（Sanders，1968）。Billett等人在深海东北大西洋的研究。（1983），随后兰皮特（Lampitt，1985）发现了深部深度的表面初级生产来源的颗粒有机物（phytodetritus）的季节性脉冲，进而发现了碳。这些早期的观测随后扩展到了东北大西洋的中部海洋区域（Pfannkuche，1993； Thiel等，1989），并促使在豪猪深渊平原地区建立了更加协调一致的时间序列研究。今天，豪猪深渊平原持续观测站（PAP–SO）是东北大西洋（48°50′N 16°30′W，4850 m水深；图1）的多学科开放海洋时间序列地点，原位与气候和环境有关的变量的测量已经进行了30多年。这代表了一个非常长的时间序列-跨陆地，淡水和海洋领域的生物时间序列数据的最新汇编表明，平均持续时间仅为13年（Dornelas等，2018）。深海的长期时间序列很少见，尤其是那些从地表到海床收集数据的时间序列。PAP-SO是全球两个深海长期时间序列站点之一（Smith等人，2015），另一个是东北太平洋M站（34°50′N，123°）的三十年时间序列。 00'W，〜4000 m水深），由蒙特利湾水族馆研究所维护（Smith等，2020）。这个``同级''深渊时间序列站点还旨在使用许多类似技术（Smith等人，2017）来了解表层海洋与海床之间的联系，从而促进两个站点之间的比较（例如Durden等人，2019; Durden等，2020a； Laguionie-Marchais等，2013； Smith等，2009）。扩展比较的另一个来源是21年时间序列长期生态研究观测站HAUSGARTEN，北大西洋与北冰洋中部之间的弗拉姆海峡的北极海洋监测前沿（FRAM）（78.5°N–80°N，05） °W-11°E，水深250-5500 m），由阿尔弗雷德·韦格纳极地和海洋研究学院维持（Soltwedel等人，2016； Soltwedel等人，2005）。我们对深海时间变化的大部分了解，