气候变化和人类活动正在不断地改变着滨海湿地栖息地，以及生物的分布格局，进 而影响物种之间的相互作用关系。生态学研究对于如何预测和管理气候变化和人类活动 变化对生态系统的影响。一个重要的目标则是了解这些变化如何影响上行和下行过程， 进而影响食物网的直接和间接作用。营养级联是指捕食者通过抑制被捕食者的种群数量 或者改变其行为，进而间接有利于较低营养级生物种群的生态过程，是研究食物网相互 作用关系的典型范例。虽然气候变化和人类活动改变的上行和下行限制因子可以同时发 生，进而影响营养级联的强度和结果，然而这些因子是否产生协同增效或者抵消的效应 还仍然不清晰。本研究以黄河口近海水域为研究区，通过现场管理实验，主要研究了营 养物质输入和碎屑物质输入的上行效应，和捕食者和中间消费者的下行效应对营养级联 的影响。最后，结合野外调查、野外受控实验、环境因子实验室分析、多元统计等方法， 确定上行控制因子和下行控制因子对营养级联强度和结果影响的相对对立和相互作用的 重要性。主要研究结果和结论为：

（1）外源性碎屑输入的上行效应决定营养级联的强度和方向

关于外源性物质输入的可利用量如何影响营养级联强度的空间变化的机制还不明确， 尤其是外源性物质输入量的可利用量与营养级联强度之间的关系仍然没有报道。通过在 黄河口三个潮沟区域，同时管理下行效应因子（捕食者存在与否）和上行效应（定量梯 度的外源性物质输入），定量研究资源输入量变化对食物网营养级联作用的影响，以及级 联强度和结果的变化对沉积环境初级生产者底栖藻类的影响。结果显示，营养级联强度 的非线性变异与外源性资源输入量的最终滞留量之间具有显著关联性。资源输入量的滞 留量显著地影响了中间捕食性无脊椎动物的生物量，成为解释营养级联空间变化的主要 原因。研究结果证明了级联强度对资源输入量最终可利用量的阈值型响应主要是由于消 费者的功能群响应变化。随着资源输入量滞留量的增加，营养级联强度从正值转化为负 值，表明营养级联更可能发生在低资源量的情况下。因此，本研究强调了识别消费者功 能群定居和觅食行为过程变化的重要性，尤其在理解人类活动改变的和自然环境梯度影 响下的资源输入量的空间异质性情况下，如何影响食物网直接和间接作用关系。

（2）捕食者多样性和碎屑输入的相互作用削弱营养级联强度

尽管生物多样性如何影响碎屑物质降解已有大量研究，但是碎屑物质输入量和生物 多样性之间的相互作用如何影响碎屑物质降解速率，消费者定居，初级生产者生长，以 及综合作用下的营养级联过程变化仍然不清晰，需要进一步研究。本研究通过管理捕食性螃蟹物种数目和定量梯度的机碎屑输入量，揭示下行调控过程和上行调控过程对食物 网相互作用的相对对立和交互作用的重要性。研究结果显示，外源性碎屑物质输入量的 增加，和生物多样性的增加，碎屑物质中的 C: N 和 C: P 比率具有显著性变化，显著促 进了碎屑生物内营养元素 N、P 的释放量，增加了水体-沉积物界面营养元素 DIN-N 和 PO43- -P 的含量，进而促进了底栖藻类生产量的累积。同时，捕食性消费者可以与外源性 碎屑资源输入量产生协同促进的作用，促进非捕食性消费者的定居，进而增加非捕食性 消费者对初级生产者底栖藻类的啃食强度，削弱食物网营养级联作用强度。本研究表明， 碎屑食物网中，增加捕食性消费者多样性可以促进非捕食性消费者对初级生产者的摄食 压力，削弱营养级联，进而显著地改变自然系统的生态系统功能。

（3）捕食者和营养物质输入对海草群落种间作用的调控作用

目前已经有研究证明营养级联对于提高植物生态系统对富营养化具有重要缓冲作用， 并能降低由于富营养化导致的负面生态效应。然而，在贫营养状态的原始（崭新的，未 被开发过干扰过的）生态系统内，营养物质输入和营养降级（捕食者数量降低导致的营 养级联效应）如何通过交互作用影响海草生态系统还未见报道。本研究通过野外现场管 理营养物质添加，控制一个三层营养级联的食物链营养级结构变化，以期探索上行和下 行效应因子如何通过交互作用影响贫营养化的海草生态系统。研究中，通过移除捕食者、 营养物质添加来揭示上行和下行效应的相对和相互作用如何影响海草生态系统对营养物 质的缓冲能力。结果发现，在原始的贫营养的河口海草生态系统，营养物质填加能够促 进海草的生长，提高其生产量，但是由于海草天蓬面积的增加，降低了底栖藻类对光的 利用度，使得底栖藻类的生物量降低。但是，捕食者移除导致的营养降级，由于释放了 植食性螃蟹，大大提高了捕食者螃蟹对海草的植食压力，因此，几乎完全抵消了营养物 质添加带来的对海草的促进效应。相反，移除螃蟹后，由于降低了植食效应，进一步的 促进了营养物质对海草生长的正向促进作用，能够促进对海草床的修复作用。研究结果 证明，一个三层食物网的营养级联是决定营养物质如何影响崭新的原始海草系统的一个 关键调节因子。

（4）捕食者-消费者大小和营养物质输入对海草群落中间作用的调控作用

食物网结构生物的体型大小以及适应性觅食是了解自然生态系统如何保持其稳定性 和生物多样性的两个核心科学问题。然而，有关捕食者-被捕食者个体大小如何影响捕食 者捕食以及中间消费者的觅食行为来调控营养级联过程，进而影响生态系统功能仍然不 清楚。本研究通过现场管理食物网的下行调控过程以及营养物质添加的上行调控过程， 揭示基于捕食者-被捕食者个体大小的营养级联下行效应和营养物质增加的上行效应对 海草生态系统的相对和相互作用的重要性。研究结果显示，在排除捕食者和消费者螃蟹 时，营养物质添加能够促进海草的生长，增加海草总体的生物量。然而，在不同个体大 小捕食者和中间消费者共存的情况下，营养级联主要变现为捕食者对植食性螃蟹中等和较小个体的抑制作用，却释放较大个体的螃蟹。在这种捕食者和中间消费者共存情况下， 营养物质填加对海草的积极作用完全消失，且海草总体生物量显著降低。说明即使捕食 者对中间消费者存在抑制作用，表现为捕食者对不同个体大小的植食性消费者的偏好型 抑制，发生的营养级联作用可以促进海草的生长；但植食性消费者的适应性觅食活动仍 然存在，进而降低了海草的生物量。很可能的原因是，第一，由于捕食者对较大个体消 费者的释放增加了其适应性觅食效应；第二，由于空间的密闭性，即使捕食者抑制了植 食性螃蟹的密度，但是仍然具有较大的植食压力。本研究表明，植食作用越强，越有利 于海草床中非优势底栖藻类的生长。 （5）基于营养级联过程调控的滨海湿地生态系统管理框架

滨海湿地食物网功能的完整性是维持滨海湿地生态系统的基础，同时也是维持与之 联系的陆地和海洋生态群落结构、动态过程的关键。当前关于滨海湿地栖息地结构的生 态修复并没有明确考虑食物网各营养级生物之间的相互作用关系，因此忽视了捕食者的 直接作用和营养级联作用对生态修复有效性的重要限制作用。本章通过整合营养级联理 论，提供基于生态系统管理调控的滨海湿地食物网修复和保护框架模式。通过识别影响 食物网组分结构以及营养级间相互作用关系的关键人为、自然驱动力和主要限制因子。 在辨识生态系统食物网生物状态及其变化趋势的同时，制定主要生态系统管理目标。然 后，基于当前生态系统环境特征，整合营养级联过程，判识营养级之间的潜在相互作用 关系，并进行生态系统风险预测分析。设定修复目标，预评估营养级联作用下的食物网 关键组分调整范围，建立理想目标状态下的稳定食物网结构及其相互作用关系。最终， 制定与实施相应的上行和下行过程调控的管理方案。

Climate Changing and anthropogenic habitat alteration are modifying the spatial habitats and species distribution in which species interactions occur. An important goal for ecological researches to predict and manage climate changes and human impacts on ecosystem is to understand how such alterations in coastal habitats affect the direct and indirect interactions in food webs, and how this in turn impacts emergence patterns of community structure. Trophic cascades are the archetypical example of examining outcome of food web interactions, and occur when predators directly consume or alter prey behaviour, indirectly facilitating populations at lower trophic levels. While climate change and anthropogenic alteration of bottom-up and top-down forces often co-occurs, whether these factors counteract or have additive or synergistic effects on strength and outcome of trophic cascades is poorly understood. In the Yellow River Estuary, China, in situ experimental manipulations were performed to determine the factors and processes of influencing food web interaction, with much emphasis given to the relative importance of bottom-up forces such as nutrition and subsidies input, and top-down forces such as the influences of herbivores (or mid-consumers) and their predators. Finally, I combined field surveys, in situ experimental manipulations, lab analysis and mathematical statistical analyses to detect the relative and interact role of bottom-up and topdown forces impact on trophic cascade, consequently altering the strength and outcome of trophic cascades in coastal waters. Main study contents and conclusions are as follows:

（1）Bottom-up forcing of subsidy controls the emergence and strength of trophic cacades

We have limited insights into mechanisms underlying spatial variation in cascade strength linked to availability of subsidy input across ecosystem gradients, especially, in terms of the quantitative relation, which is rarely explored. I simultaneously manipulated the top-down (predator presence) and bottom-up (a quantitative gradient of subsidies) interactive effects on the primary producers of benthic algae located at three tidal creeks. I found that non-linear variability in trophic cascade strength was strongly linked to the retention of subsidy input.

Retention of subsidy input influenced the colonization and foraging behaviour of intermediate
predatory invertebrates considerably, accounting for the spatial variation in trophic cascades.
The results demonstrated that there were thresholds of cascade strength to availability of subsidy
input due to consumer’s functional responses. Trophic cascade strength shifted from positive to
negative with increasing retention of subsidy input, suggesting that trophic cascades were more
likely to occur at lower subsidy conditions. Thus, the results highlight the importance of
identifying joint processes of colonisation and foraging behaviour of specific functional
consumers in understanding the effects of both human-altered and natural variations in subsidies
on the direct and indirect food web interactions.
(2) The interactive effects of predaotr diversity and susbsidy dampening trophic cascades
Ecologists have increasingly focused on how changes in the biodiversity of consumer affect ecological rates and biophysical processes in ecosystems. However, we have limited knowledge for the interactive effect of biodiversity and resources subsidy on decomposition, consumer recruitment, primary producers and the trophic cascades. I experimentally manipulated the number of predatory consumer (mud crabs) and quantitative gradient of subsidies, aiming to investigate the relative and interactive importance of top-down and bottom-up control of food web interactions. The results showed that detritus decay rates (decomposition), and loss of
nitrogen (N) and phosphorus (P) from detrital substrates significantly increased with increasing of subsidy input and biodiversity. Dissolved inorganic nitrogen (DIN) and soluble phosphorus (PO43--P) concentrations in the water column also increased with subsidy and biodiversity. Importantly, the synergic effect of subsidy and biodiversity increased the colonization of invertebrates (primary consumers) and grazing pressure on benthic microalgae, indicating that enhancing predator diversity dampens predator effects on invertebrates and weakens trophic cascades. Consequently, changes in diversity at higher trophic levels in benthic food webs can
impact on the subsidy effects, can significantly alter ecosystem functioning and processes in coastal ecosystems.
(3) Predator and nutrient meidate the interspecific competiton in seagrass system.
Understanding the buffering capacity of ecosystems to changing conditions is one of the most important goals of ecosystem science. In coastal vegetated ecosystems, a number of studies recently demonstrated that trophic cascades are particularly important in buffering or reducing negative effects of nutrient enrichment found under eutrophic conditions. Yet, it currently remains unclear how nutrient enrichment and trophic downgrading interact in pristine coastal ecosystems under oligotrophic conditions. I investigated how bottom-up and top-down forces  interact in an oligotrophic seagrass ecosystem by manipulating both nutrient loading and trophic interactions within a tri-trophic cascade in a field experiment. I factorially performed predator removal and nutrient addition to uncover the independent and interdependent roles of top-down and bottom up control in buffering nutrient addition in seagrass ecosystems. The results revealed that in pristine seagrass beds, nutrient addition stimulates seagrass production, while it reduced the growth of benthic microalgae, presumably because the thicker seagrass canopy reduced light availability. Trophic downgrading by excluding predators, however, almost completely negated these nutrient effects, as the release of herbivores from predation strongly enhanced grazing pressure on seagrass. The exclusion of grazers in turn restored seagrass biomass by allowing the nutrient addition to regain their effect, confirming that a tri-trophic cascade was an important mediator in how enhanced nutrient loading affected our system.

(4) Predator-prey body size and nutrient impact on interspecific competiton in seagrass system.

Body size and adaptive foraging in food webs, are two important science issues in ecology for understanding how ecosystem persist stability and biodiversity. However, we have limited knowledge into the mechanisms that predator-prey size impact predator predation and prey foraging behaviour, with nutrient loadings, consequently controlling trophic cascades. I examined how bottom-up forcing and size-dependent trophic cascade interact impact on seagrass system by manipulating both nutrient loading and trophic interactions in a field experiment. The results showed that in the absence of predator (fish) and herbivore (crabs), nutrient enrichment has positive effect on seagrass growth, and increases seagrass biomass. In the co-existence of predator and herbivores in different body size, trophic cascades occur indicating that top predators control of medium and small body size of crabs, releasing crabs in big body size. In addition, in the co-existence of predators and herbivores, herbivory counteracts the positive effects of nutrient enrichment on seagrass. Predator preferential consumption of consumers in smaller body size initiates trophic cascade facilitates positive effects of nutrient on seagrass growth, but this response is countered by a proportional increase in grazing. Adaptive foraging behaviour of released consumers in big body size, and higher density in experimental plot may account for the countervailing processes. Therefore, there is strong evidence that herbivory is capable of regulating competitive interactions between seagrass and benthic microalgae.

(5) A conceptual framework for integrating trophic cascades into ecosystem-based management.

Well-functionality of food webs is fundamental for sustaining coastal ecosystems and maintaining associated marine and terrestrial communities. The current emphasis on restoring coastal habitat structure without explicitly considering trophic interaction of food webs, thereby has been less successful than hoped in terms of enhancing the status of targeted species, Importantly, current emphasis has received less attention on the important constraints of trophic interaction on ecologically effective restoration. Here, trophic cascade principles (TCP) are integrated to ecosystem-based management (EBM), aiming to present a conceptual framework of management (TCP-EBM), which provides priority of food web-related issues that avoid the likely uncertainties may impede restoration effectiveness. I outline and illustrate step-by-step procedures that ensure ecosystem management will be guided toward sustainable actions: (1) actions should be scoping at a scale and identify the root driving forces and pressures of damaged ecosystem; (2) actions should develop ecosystem indicators and targets that partly are dependent on the structures and function of food webs; (3) actions should be prior to restoration and conducted necessary risk analysis in direct and indirect interactions of food web that predict the likely trends that are commensurate with current environmental problems; (4) actions should have clear assessment of ecosystem status relative to articulated expected outcomes of EBM goals and (5) conduct trophic cascade-based principles for manipulating top-down and bottomup processes of food webs for restoring coastal ecosystems. During the management practice, monitoring of ecosystem indicators and management effectiveness, and some adaptive management monitoring are needed. An integrated TCP for coastal restoration would complement current and ongoing shortcomings and uncertainties of restoration due to overlooking the biotic constraints on effectiveness. Thus, this presented conceptual frame TCPEBM will be helpful to promote the ecosystem integrity of structure and function of EBM.