The integration or co-application of traditional knowledge and scientific knowledge has been the subject of considerable research and discussion (see Johannes 1981; Johnson 1992; Stevenson 1996; McDonald et al. 1997; Huntington et al. 1999, 2002), with emphasis on various specific topics including environmental management and conservation (see Freeman & Carbyn 1988; Ferguson & Messier 1997; Ford & Martinez 2000; Usher 2000; Albert 2001). In most cases, examples of successful integration compare traditional and scientific observations at similar spatial scales to increase confidence in understanding or to fill gaps that appear from either perspective. We present a different approach to integration, emphasizing complementarity rather than concordance in spatial perspective, using two migratory species as examples.

Biological invasions are pervasive, alter ecosystem sustainability, and can reduce native biological diversity (Vitousek et al. 1996). Ecologists are scrambling to predict where and when new invaders may strike (Lodge 1993; Fagan et al. 2002). Urbanization has created a number of new ecological niches which, after remaining empty for some time, are increasingly being colonized or invaded by vertebrates (Erz 1966). One of the major disputes in evolutionary biology concerns the rate at which species evolve to occupy new niches (Diamond 1986). How quickly are animals adapting to these new habitats and what makes them successful colonists?

Implementation of the European Union (EU) Water Framework Directive (WFD 2000) and its planned supplement ‘Directive on the Protection of Groundwater against Pollution’ (EU GWD [European Union Groundwater Directive] 2003) demands an initial characterization of all groundwater bodies by national and regional authorities. The main criteria considered in the Directives to define the groundwater (GW) status are quality (chemical) and quantity, but there is the obvious omission of ecological perspective in these. A directive for a comprehensive policy dealing with GW protection at the level of the EU is a prerequisite for human welfare. Additionally, recognition of GW-dependent ecosystems (GDEs) in the EU is equally important for their sustainable management. GW management and policy should recognize the ecological functions of GW and their interactions with GDEs. As they can stretch across national and regional boundaries, political recognition of their importance and the necessity for ecological consistency of management plans have to be emphasized. During the last three decades, a number of researchers have focused their attention on the dynamics and functioning of GW ecosystems and this work could serve as a basis for identifying impacts of changes in key attributes of GDEs. Detailed aspects of GW ecosystems have been comprehensively dealt with (Gibert et al. 1994; Wilkens et al. 2000; Griebler et al. 2001; Danielopol et al. 2003). Changes in key attributes of GW ecosystems have had consequences for the environment in the EU and USA (Klijn & Witte 1999; Winter 1999; Sophocleous 2002; Winter et al. 2003).

To ensure that the best scientific evidence is available to guide conservation action, effective mechanisms for communicating the results of research are necessary. In medicine, an evidence-based approach assists doctors in applying scientific evidence when treating patients. The approach has required the development of new methods for systematically reviewing research, and has led to the establishment of independent organizations to disseminate the conclusions of reviews. (1) Such methods could help bridge gaps between researchers and practitioners of environmental conservation. In medicine, systematic reviews place strong emphasis on reviewing experimental clinical trials that meet strict standards. Although experimental studies are much less common in conservation, many of the components of systematic reviews that reduce the biases when identifying, selecting and appraising relevant studies could still be applied effectively. Other methods already applied in medicine for the review of non-experimental studies will therefore be required in conservation. (2) Using systematic reviews and an evidence-based approach will only be one tool of many to reduce uncertainty when making conservation-related decisions. Nevertheless an evidence-based approach does complement other approaches (for example adaptive management), and could facilitate the use of the best available research in environmental management. (3) In medicine, the Cochrane Collaboration was established as an independent organization to guide the production and dissemination of systematic reviews. It has provided many benefits that could apply to conservation, including a forum for producing and disseminating reviews with emphasis on the requirements of practitioners, and a forum for feedback between researchers and practitioners and improved access to the primary research. Without the Cochrane Collaboration, many of the improvements in research communication that have occurred in medicine over the last decade would not have been possible.

Urgent threats to the Gulf of California ecosystem from modern human activity obscure the fact that humans have interacted with native plants and animals for millennia. Archaeological and historical evidence suggests that indigenous peoples occupied both sides of the Gulf some 13 000 calendar years ago and that they eventually inhabited six major islands and visited most smaller ones. Biologists have increasingly realized that these peoples probably played a role in shaping island biotic communities extant today. How much of a role is unknown, but the best places to find evidence may be archaeological sites, which often contain remains of plants and animals directly used by prehistoric peoples. The opportunity to investigate the interaction between early humans and island biota may be lost because modern island visitors endanger sites. Many people, whether boaters, ecotourists, government officials, scientists or artefact collectors, enjoy picking up artefacts. Small surface sites, with exposed remains, can be completely denuded in minutes. Visitors to small islands can obliterate entire archaeological records, thereby creating the illusion of pristine islands. This problem is bound to worsen as Mexico implements Escalera Náutica, a chain of marinas specifically intended to multiply manyfold the boating population. The Mexican government's management plan for Gulf islands, published in 2000, recognizes only a general need to manage cultural resources. Specific mechanisms for protecting sites should be developed. These should educate visitors about the importance of the archaeological record and the destructiveness of collecting. They should also provide adequate enforcement of Mexico's existing antiquities laws. Conserving the archaeological resources may be the best way of preserving biological data essential to island biogeographers and ecologists.

The introduction of swine (Sus scrofa) has adversely affected the environment of many natural habitats throughout the world. Basin marshes are dwindling ecosystems in Florida that are especially vulnerable to damage by feral swine. In January 2003, the estimated amount of swine damage to the exposed portion of the last remnant of a basin marsh system in Savannas Preserve State Park (SPSP) was 19% (an area of 5 ha). Economic valuations for the swine damage were based on the monetary amounts that wetland regulators have allowed permit applicants to spend in mitigation attempts to replace lost wetland resources. In 2003, the area of natural habitat damaged by swine had a total value of US$ 1 238 760–4 036 290. The SPSP implemented a contract for swine control throughout 2003 in all areas of the Park. The damage to the basin marsh was re-estimated in January 2004, after swine removal. The damage sampling transects remained the same as the 2003 survey, but damage was significantly reduced, with 31% of sampling transects showing damage in January 2004 versus 92% in January 2003. Similarly, the total area of swine damage had decreased to 0.95 ha, and the value of the lost habitat had been reduced to US$ 235 355–766 865, an economic benefit of US$ 1 003 355–3 269 265. The benefit-cost ratio (BCR) relating the economic value of the damage reduction against the swine control costs (not all swine control in the park was carried out in the vicinity of the basin marsh) was conservatively estimated at 134–436. When estimating costs for swine control in the vicinity of the marsh based on the proportional number of swine removed there as compared to the Park as a whole, the BCR increased to 480–1562. Benefits of swine removal are very high relative to the costs of control.

Analysis of ecological baseline data collected for key resource species within the Galapagos Marine Reserve indicates that variation in animal density associated with the location of fully protected zones can be comparable to protected area effects. Even with a high level of interspersion between conservation, tourism and fishing management zones, major differences in densities of economically important species were evident between zone types prior to enforcement of fishing restrictions. Densities of the most valuable fishery resource, sea cucumbers, were three times higher in zones that remained open to fishing compared to ‘no-take’ conservation zones, and densities of sharks were five times higher in tourism zones than fishing or conservation zones. These results highlight bias in the socio-political processes that can accompany selection of marine protected areas, where fishers attempt to minimize perceived impacts on their livelihood by locating large protected zones in resource-poor areas, and tourism operators and sport divers argue for protection of areas containing atypically-interesting features. Bias in the location of fully protected zones can seriously confound ‘control-impact’ field investigations when data prior to prohibitions on fishing are lacking, including meta-analyses, which are dominated by such potentially systematically biased studies.

A conservation project aimed at ecosystem restoration had several unforeseen effects on a colony of the yellow-legged gull Larus michahellis in a small western Mediterranean island (Benidorm Island). The project included regulation of massive tourist visits to help restore the soil and autochthonous vegetation. However, gulls habituated rapidly to regulation of tourist activities, as nests located either close to or far from the main trail showed a similar hatching success. The quiet conditions produced by regulation seemingly facilitated a rapid colony increase. Partial removal of alien vegetation (Opuntia maxima) showed that gulls had a preference for sites with high vegetation cover because the growth of the colony was proportionally larger in well-vegetated plots. The pricking of a large number of gull eggs surprisingly coincided with a high reproductive success compared to the previous year, although indicators of food availability remained constant between years and the colony had decreased in numbers. Untreated nests were probably more successful because territory size for chicks increased and intraspecific predation decreased. Extreme care must be taken when planning ecosystem-wide management on islands with yellow-legged gull colonies, or other gull species locally considered as pests, to prevent unwanted effects.

There is increased international interest in religiously based restrictions on land and forest stand use. However, the extent to which so-called sacred groves represent earlier forest ecosystems, and their possible role in biodiversity conservation, are interrelated and complex issues, and neglected in the context of Ghanaian savannahs, which are believed to be in transition from a forested past. Geographical information system (GIS) analysis of time-series images, aerial photographs, statistical analysis of ecological field data and social surveys were used to investigate the Ghanaian savannah. Sacred groves were found to be similar in plant species content and structure to the deciduous forest of southern Ghana. Between 1960 and 1996, these groves experienced far fewer forest losses than unprotected tree stands (<20%, as opposed to up to 100%). Non-sacred groves also contained lower plant species richness and higher numbers of invasive species such as neem (Azadirachta indica A. Juss). Sacred groves, while dependent on respect for religion, local cultural structures and individual peer pressure, offer a role that may support and also be supported by official conservation efforts.

Loss of wildlife, encroachment on wild lands and conflicts between protected areas and neighbouring communities continue to threaten the integrity of protected areas (PAs) in Uganda. To increase support from local communities and ensure long-term sustainability for Uganda's PAs, a policy of community-based conservation was introduced in 1988 as a management approach in seven PAs. The effectiveness of the community-based approach for reducing threats was compared to conventional PA management by conducting a threat reduction assessment at 16 PAs, seven with community-based approaches and nine without. Additional data collected using document reviews, interviews with government officials, and surveys of PA wardens were compared with the threat reduction assessments. Twenty-three primary threats were identified at PAs. Local game poaching was the most common threat. The threat reduction assessment indices of community-based PAs (mean=49.0±12) were not significantly different from those of conventional PAs (mean=37.96±21.6). Some specific threats, such as bush burning, logging, encroachment and unclear boundaries, seemed to be better mitigated at community-based PAs. Management approaches at all PAs mitigated fewer than half of the identified threats. Management approaches are needed that directly target PA threats, go beyond PA boundaries by involving additional government departments, link people's livelihoods to conservation efforts and strengthen PA institutions.

Non-timber forest products (NTFPs) constitute the single largest determinant of livelihoods for scores of forest fringe communities and poor people in the tropics. In India over 50 million people are believed to be directly dependent upon NTFPs for their subsistence. However, such human dependence on NTFPs for livelihood gains (win) has most frequently been at a certain ecological cost (lose). If livelihoods are to be maintained, the existing ‘win-lose’ settings have to be steered to a ‘win-win’ mode, otherwise, there could be severe erosion of the biological resources and loss of livelihoods (‘lose-lose’). Examining the dependence of forest fringe communities on NTFPs at three sites in south India with contrasting human and ecological settings, three key factors (extent of dependence on NTFPs, indigenous ecological knowledge and market organization) are likely to constrain reaching the win-win situation. How these factors shape the ecological cost of harvesting NTFPs at the three sites is examined. Within the parameter space of these factors, it is possible to predict outcomes and associations that will conform to win-win or win-lose situations. Empirical data derived from the three study sites demonstrate the causality of the observed associations. The key for long-term livelihood gains lies in reducing the ecological cost. Certain interventions and recommendations that could optimize the balance between livelihood gains and ecological cost are proposed.

To elucidate effects of fragmentation and degradation of tropical rainforest on Negros Island (Philippines) on tropical herpetofauna, nine tropical rainforest fragments (eight limestone and one non-limestone), ranging in area from five to 122 ha (total area = 352 ha) in south-western Negros were surveyed using cruising and quadrat sampling methods, between December 2001 and October 2002. A 14-ha limestone plantation forest served as a reference site. Ten to fifteen of the total 61 species (16.4–24.6%) expected in the nine forest fragments (taken as a whole) had been lost in the previous 50 years. These species are known to occur in the tropical rainforest on Cuernos de Negros and environs in south-eastern Negros Island at altitudes <500 m above sea level. The factors responsible for the loss of these species appear to be (1) removal of large trees, which decreases canopy cover causing physical changes; and (2) forest fragmentation, which increases the edge areas. However, it is not possible to separate the effects of each of these two factors. The main factor that was correlated with the number of species in the limestone forest fragments appeared to be area. Two species of frogs appeared to be represented by relatively moderate population densities: the Negros cave frog (Platymantis spelaeus) and the common forest frog (Platymantis dorsalis). The burrowing skink (Brachymeles boulengeri taylori) also exhibited a relatively higher density than the other burrowing species of skinks. The other herpetofaunal species exhibited low population densities.