Rocky shores occur at the interface of the land and sea. Typically they are open ecosystems, with steep environmental gradients. Their accessibility to man has rendered them susceptible to a variety of impacts since prehistoric times. Access can be regulated, however, and they are more amenable to management than open ocean habitats. This review uses examples from throughout the world to demonstrate the extent to which rocky shores have been, and are currently, affected by pollution (examples used are endocrine disrupters, oil, eutrophication), over-collection of living resources, introduced alien species, modification of coastal processes (coastal defences, siltation) and global change (climate, sea level). These impacts are put into the context of natural fluctuations in time and variability in space of both the environment and the organisms. The relative magnitudes of some anthropogenic disturbances differ between the industrialized, developed world and the developing world. For example, in developed, industrialized countries pollution based impacts should diminish over the next 25 years due to improved regulation and a reduction in older ‘dirtier’ heavy industry. Conversely, in many developing countries pollution will increase as a consequence of growth in the human population and industrialization. Except for large-scale disasters such as oil spills, pollution tends mainly to influence embayed coastlines. Chronic effects such as eutrophication can have broader-scale impacts over whole coastlines and elevated nutrient levels have also been implicated in a trend of increasing frequency of catastrophic kills due to harmful algal. Direct removal of living resources has had major effects on coastlines at both local and regional scales and is likely to increase over the next 25 years, especially in developing countries where rapidly expanding human populations will put further pressure on resources. Impacts from recreational activities are likely to increase with greater leisure time in wealthier regions of the world, and cheaper travel will spread these impacts to poorer regions. Invasions by alien species have increased in frequency during the last 20 years leading to some dramatic effects on native assemblages. Problems associated with alien species, especially pathogens, will continue to increase over the next few decades. The proportion of the coastline modified by artificial structures (breakwaters, seawalls, groynes) will increase because of coastal development and defences against sea-level rise and the greater frequency of storms. This will increase connectivity between areas of rocky habitat. Siltation will continue to increase due to urbanization of catchments and estuaries, and changes in agricultural practice. This may have considerable impacts at local and regional scales, favouring sediment tolerant organisms such as turf algae and anemones. In the future, greater frequency of environmental extremes is likely, including large-scale events such as the El Niño Southern Oscillation (ENSO). Global change in temperature, sea-level rise and increases in the frequency of storms will affect rocky shores throughout the world, but this will occur over long time scales; over the next 25 years most of the responses by rocky shore communities will mostly be quite subtle. Thus rocky shores will be subject to increasing degradation over the next 25 years. They are, however, less vulnerable than many other aquatic habitats due to their hard substratum (rock), their relative lack of large biogenic structures and to their generally open nature. They are also remarkably resilient, and recovery can occur rapidly due to recruitment from unaffected areas. Their susceptibility to both terrestrial and marine disturbances does make them more vulnerable than sublittoral and offshore habitats. There are considerable gaps in knowledge, particularly of certain microhabitats such as crevices, boulders, sand-scoured areas and rock pools. These have been much less studied than more accessible assemblages on open, freely draining rock. More research is needed to establish the effects of increasing sediment loads, ultraviolet radiation and introduced species on rocky shore communities. Strategic and applied research programmes should integrate field experiments and carefully selected monitoring programmes to verify management regimes. Hindcasting from the palaeo-record would be valuable, to compare rates of predicted change with periods when change was rapid in the past. This information could, in principle, be used to help conserve rocky shores through networks of marine protected areas and a general reduction of environmental pollution.

Given currently high rates of extinction, it is critical to be able to predict how ecosystems will respond to loss of species and consequent changes in community structure. Much previous research in this area has been based on terrestrial systems, using synthetically assembled communities. There has been much less research on inter-trophic effects in different systems, using in situ removal experiments. Problems with the design of early experiments have made it difficult to determine whether reductions in ecosystem functioning in low diversity treatments were due to the number of species present or merely to the reduced likelihood of including particular (''key'') species or functional groups. We established a field experiment, using cage enclosures, to test whether the number or identity of grazing gastropods present would affect the diversity and productivity of intertidal algal communities. We found that there was no relationship between ecosystem functioning and diversity per se, but that different species had idiosyncratic effects. This appears to be a common feature of intertidal systems, which often contain strongly interacting species. In this case, the limpet Patella ulyssiponensis was the most influential primary consumer. Additional treatments enabled us to test the potential for other grazers to compensate for its loss. Early results (after four months) suggested that compensation was possible, implying potential redundancy of P. ulyssiponensis. This effect disappeared after 13 months, however, highlighting the context dependence of so-called redundancy and underlining the importance of long-term field-based experiments in this area.

Habitat-forming species sustain biodiversity and ecosystem functioning in harsh environments through the amelioration of physical stress. Nonetheless, their role in shaping patterns of species distribution under future climate scenarios is generally overlooked. Focusing on coastal systems, we assess how habitat-forming species can influence the ability of stress-sensitive species to exhibit plastic responses, adapt to novel environmental conditions, or track suitable climates. Here, we argue that habitat-former populations could be managed as a nature-based solution against climate-driven loss of biodiversity. Drawing from different ecological and biological disciplines, we identify a series of actions to sustain the resilience of marine habitat-forming species to climate change, as well as their effectiveness and reliability in rescuing stress-sensitive species from increasingly adverse environmental conditions.

Pacific oysters are now one of the most 'globalised' marine invertebrates. They dominate bivalve aquaculture production in many regions and wild populations are increasingly becoming established, with potential to displace native species and modify habitats and ecosystems. While some fishing communities may benefit from wild populations, there is now a tension between the continued production of Pacific oysters and risk to biodiversity, which is of particular concern within protected sites. The issue of the Pacific oyster therefore locates at the intersection between two policy areas: one concerning the conservation of protected habitats, the other relating to livelihoods and the socio-economics of coastal aquaculture and fishing communities. To help provide an informed basis for management decisions, we first summarise evidence for ecological impacts of wild Pacific oysters in representative coastal habitats. At local scales, it is clear that establishment of Pacific oysters can significantly alter diversity, community structure Communicated by David Hawksworth.

Larval transport is a key process in the life-history and population dynamics of many marine species. It is strongly influenced by ocean currents, but the influence of behavioural traits of larvae (e.g. vertical migration) on their advective dispersal is poorly understood. In the absence of field data, predictions of population connectivity are often based on hydrodynamic models and assume that larvae behave as passive particles. Variations in the vertical distribution of early and late stage larvae of Mytilus spp. were investigated in a field study in the Irish Sea, collecting discrete water samples by Niskin sampler at a range of depths at different tidal states and phases at 2 sites and on 2 dates during summer 2005. Larvae were homogenously distributed throughout the water column during flood tide, whereas larvae we more densely aggregated in middle and bottom waters during ebb tide. Highest densities were greatest near the bottom during low flow conditions. Larval size had no effect on the patterns of vertical distribution of larvae in the water column. The relative lack of larvae in the water column during ebb tides compared to flood tides could lead to significant net transport in the direction of flood tides. If this is a widespread phenomenon, models based on the assumption that larvae behave as passive particles are unlikely to predict dispersal distances correctly.