Impact of human action is being felt in all ecosystems. Traditionally, urban landscapes and agricultural fields have been considered to be dominated by humans, but human alteration has also been substantial in tropical forests. The common forms of acute forest disturbance are ones that involve logging selectively or clear-cutting, allowing forest to regenerate by natural means. In contrast to the acute forms of disturbance, the chronic form of disturbance, which is subtle and slowly creeping, but equally destructive, is a much less-recognized problem. In the chronic form of disturbance people remove only a small fraction of forest biomass at a given time, generally as head-loads of firewood, or in the form of fodder, leaf litter and other non-timber forest products (NTFPs). The problem with the chronic form of forest disturbance is that plants or ecosystems often do not get time to recover adequately, because the human onslaught never stops, and can cause adverse changes in the forest, even if rates of biomass removal are within the carrying capacity of the forest. In a similar fashion, the cumulative effects of low but chronic exposure to air pollution are now cause for concern all over the planet (Pitelka 1994). In the case of chronic forest disturbance, the effect somewhat resembles that of persistent insect herbivory at a moderate scale. Fuelwood, which is still the main source of cooking energy in most developing countries, is reported to be in under supply for 1.4 thousand million people, and these may rise to 2.5 thousand million by 2010 (Food and Agriculture Organization 1994). Therefore, more areas are likely to come under regimes of chronic disturbance in the next century.

A model of disturbance equilibration

The contemporary ‘non-equilibrium’ paradigm in ecology acknowledges that ecosystems and their component floral and faunal communities show historic dynamism, not unchanging constancy (Shugart & West 1981; Pickett et al. 1992; Hengeveld 1994). In formulating a hierarchical view of landscape ecology, Urban et al. (1987) modelled broad classes of forest disturbances and corresponding landscape scales. A reference level is the scale at which a disturbance is observed as an ‘interesting event’. At the next, lower landscape scale the disturbance has a significant impact and explanatory mechanisms may be investigated. At the level above the reference scale, the ‘event’ may be viewed in a broader ecological context where its impact has been buffered and successfully incorporated. Urban et al. (1987) suggested that ‘a disturbance regime that can be incorporated is not disturbing at all’. Extending this perspective to anthropogenic perturbations, Urban et al. (1987) praised the ‘wisdom’ of foresters for discovering clear-cut harvesting as a forest management strategy that mimics natural disturbances. It might be possible for a reader with a pro-exploitation agenda to misinterpret the meaning of the model presented in Urban et al. (1987) and attempt to use it in support of forest management that is destructive to overall ecosystem health, thereby detracting from Urban et al.'s (1987) and Shugart and West's (1981) pioneering contributions to the field of landscape ecology. It could be argued that we only need look to a higher landscape scale in the model of equilibration to find a level at which clear-cuts need no longer be considered forest disturbances.

Increasingly, natural areas are exposed to people who come to view, study or photograph wildlife. In order to develop appropriate management plans for both avian and human use of natural environments it is essential to understand how people affect foraging birds. The foraging behaviour of five species of water-birds at Loxahatchee (Arthur B. Marshall National Wildlife Refuge), part of the Everglades, in Southern Florida was observed, between 1992 and 1994, from a dike that received many visitors. Species examined included common gallinule (Gallinula chloropus), sora rail (Porzana carolina), glossy ibis (Plegadis falcinellus), little blue heron (Egretta caerulea) and Louisiana heron (E. tricolor). These birds were observed before people were near, while people were present, and following the departure of people. Variation in feeding behaviour was largely explained by whether people were present, the number of people present, and the amount of noise made by the people. For all species, time devoted to feeding and number of strikes or pecks decreased while people were present. The percentage of time spent foraging and the number of strikes decreased as the noise made by people increased. Birds that were closer to the path flew away from people more often than birds that were further away. Birds usually swam or flew away from the path while people were present.

New legislation has encouraged regulatory biologists in the UK to examine the impacts of eutrophication in rivers for the first time. The principal tools for this have been new indices based on macrophyte and diatom communities. The use of such indices is placed within an appropriate theoretical framework. The importance of recognizing the upper limit of sensitivity of such indices, as well as factors such as organic pollution, which can confuse interpretation, is discussed. Despite their limitations, community-based indices are valuable tools for reconnaissance studies and, in the long-term, as indicators of the extent to which nutrient reductions have led to an ecological response. Some general guidelines for monitoring eutrophication in rivers are proposed.

Madagascar's biodiversity is of extremely high international significance, yet comprehensive efforts to assess current knowledge and set priorities have been absent until recently. Beginning in April 1995, a major participatory effort to assess the country's scientific and conservation priorities was undertaken in Madagascar. This process laid important groundwork for the revision of Madagascar's National Environmental Action Plan. The first stage of the process was a scientific priority-setting workshop. Over one hundred experts, organized in thematic groups, reached consensus on biodiversity priorities for the island, based on cross-discipline comparisons. A principal finding of the workshop is that many areas of outstanding biodiversity and research importance are located outside protected areas. Participants also agreed that corridors needed to be created between the high-priority protected areas in order to maintain gene flow and exchange of species. The second stage of the process was a stakeholder consultation which integrated scientific findings, national priorities, local stakeholder views, and donor input. The stakeholder consultation concluded that a collaborative, regional approach was needed to augment site-based conservation activities. Participants also emphasized that institutional strengthening in forestry and parks agencies needed much higher priority. The net result of the process was the adoption of a landscape approach to conservation which integrates regional planning, biodiversity monitoring and institutional strengthening.

Definition of appropriate tropical forest policies must be supported by better information about forest distribution. New information technologies make possible the development of advanced systems which can accurately report on tropical forest area issues. The European Commission TREES (Tropical Ecosystem Environment observation by Satellite) project has produced a consistent map of the humid tropical forest cover based on 1 km resolution satellite data. This base-line reference information can be further calibrated using a sample of high-resolution data, in order to produce accurate forest area estimates. There is good general agreement with other pantropical inventories (Food & Agriculture Organization of the United Nations Forest Resources Assessment 90, World Conservation Union Conservation Atlas of Tropical Forests, National Aeronautics & Space Administration [USA] Landsat Pathfinder) using different approaches (compilation of existing data, statistical sampling, exhaustive survey with satellite data). However, for some countries, large differences appear among the assessments. Discrepancies arising from this comparison are here analysed in terms of limitations associated with each approach and they are generally associated with differences in forest definition, data source and processing methodology. According to the different inventories, the total area of closed tropical forest is estimated at 1090–1220 million hectares with the following continental distribution: 185–215 million hectares in Africa, 235–275 million hectares in Asia, and 670–730 million hectares in Latin America. A proposal for improving the current state of forest statistics by combining the contribution of the various methods under review is made.

Tropical moist forests are declining rapidly world-wide. Reforestation efforts in Trinidad have generally been insufficient and focused on non-native species. Some of these species (e.g. Pinus caribaea), however, appear ecologically inappropriate but have been used extensively on sandy soils after fires or where harvesting has degraded the original vegetation. Little research attention has been given to natural regeneration as a potential for future timber production. The purpose of this study was to determine whether or not post-fire regeneration provided sufficient species of commercial importance that could influence forest management planning. The field research was conducted at the Melajo Nature Reserve in north-east Trinidad, where there has been a particular focus on non-native species plantation in adjacent areas. A fire in 1987 devastated much of the remaining mature mora (Mora excelsa Benth.) forest at the Melajo Nature Reserve. The successional species were identified and recorded four years after the fire, and the abundances of commercially-important regenerating species were determined. The total number of species recorded was 201, representing 60 families with 153 genera. There were 14 species of timber value, with Mora excelsa, Terminalia amazonia, Byrsonima spicata and Sterculia caribaea being the most abundant. The abundance of commercially-important timber species in the most fire-damaged plot was 93% less than that for the least fire-damaged plot. The density of these species at the post-fire site in Melajo was on average higher than that of the same species in the mature mora forest. In 1991, the post-fire site at Melajo contained about 54% of the species recorded for a mature mora forest at Matura, north of Melajo (over 50 years ago).

The Red Sea is a large marine ecosystem in which biological research has been considerable but integrated environmental assessment insubstantial. Approximately 1400 coastal and offshore (i.e. island) sites in the Saudi Arabian Red Sea were examined and an analysis of ordinal data on the abundance of ecosystems and magnitude of human uses/environmental impacts was conducted. Mangroves, seagrasses, other floral groups and terrestrial mammals were significantly more abundant at the coastal sites than offshore. The coastal sites were also impacted most heavily, while reefs, birds, turtles and marine mammals were significantly more abundant in offshore areas. Latitudinal trends include significantly increased abundance of mangroves and seagrasses (and other flora) towards the southern Red Sea, and a decrease in abundance of reefs. Significantly higher levels of beach oil were encountered towards the northern Red Sea, probably reflecting its greater proximity to the Gulf of Suez. Cluster analysis using all biological data revealed distinctive groupings which separated according to latitude. The biogeographic patterns are comparable to those observed in previous studies for seagrasses and other communities.

Using a relational database, applications of the findings to coastal management include creation of environmental profiles for particular sites or sectors, identification of resource-use conflict areas, and selection of representative sites for protected areas. Comparison with data from a complementary investigation in the Arabian Gulf indicates that the Red Sea is less perturbed by human activities than the Arabian Gulf. However, it is also evident that the Red Sea is no longer a pristine environment.