Massive structural and compositional changes over two decades in forest fragments near Kampala,Uganda

Private forests harbor considerable biodiversity, however, they are under greater threat than reserved areas, particularly from urbanization, agriculture, and intense exploitation for timber and fuel wood. The extent to which they may act as habitats for biodiversity and how level of protection impacts trends in biodiversity and forest structure over time remain underresearched. We contribute to filling this research gap by resampling a unique data set, a detailed survey from 1990 of 22 forests fragments of different ownership status and level of protection near Kampala, Uganda. Eleven of the 22 fragments were lost over 20 years, and six of the remnants reduced in size. Forest structure and composition also showed dramatic changes, with six of the remnant fragments showing high temporal species turnover. Species richness increased in four of the remaining forests over the resample period. Forest ownership affected the fate of the forests, with higher loss in privately owned forests. Our study demonstrates that ownership affects the fate of forest fragments, with private forests having both higher rates of area loss, and of structural and compositional change within the remaining fragments. Still, the private forests contribute to the total forest area, and they harbor biodiversity including IUCN “vulnerable” and “endangered” species. This indicates the conservation value of the fragments and suggests that they should be taken into account in forest conservation and restoration.     

Land snail diversity in post extraction secondary forest reserves in Edo State,Nigeria

Knowledge of the fauna of tropical lowland rainforest is urgently necessary because of the high rate of biodiversity loss and global extinction of species as a result of deforestation. We studied land molluscs species richness and diversity in four heavily degraded secondary forest reserves and one old‐growth forest reserve in Edo State, Nigeria using a combination of direct fixed‐time search and litter‐sieving techniques in twelve plots of 400 m² each per reserve. A total of 43 species and 2570 individuals were collected from all the reserves. Local species richness ranged from 19 to 39 species while number of specimens from 203 to 971. Molluscan species richness and diversity is significantly higher in the old‐growth forest than in the disturbed forest reserves. Land molluscs family composition was similar in all the forests with respect to the common and wide‐ranging species while rare and narrow‐range species are restricted to the old‐growth forest in Okomu. The carnivorous streptaxids and detritivorous subulinids dominate species richness and abundance respectively in all the sites. Species turnover is moderately high within and between the forest reserves indicating the uniqueness of the faunal composition of each forest and the need for adequate protection.     

Liana habitat associations and community structure in a Bornean lowland tropical forest

Lianas (woody vines) contribute substantially to the diversity and structure of most tropical forests, yet little is known about the importance of habitat specialization in maintaining tropical liana diversity and the causes of variation among forests in liana abundance and species composition. We examined habitat associations, species diversity, species composition, and community structure of lianas at Sepilok Forest Reserve, Sabah, Malaysia in northeastern Borneo among three soil types that give rise to three distinct forest types of lowland tropical rain forest: alluvial, sandstone hill, and kerangas (heath) forest. Alluvial soils are more nutrient rich and have higher soil moisture than sandstone soils, whereas kerangas soils are the most nutrient poor and drought prone. Lianas ≥0.5-cm in diameter were measured, tagged, and identified to species in three square 0.25-ha plots in each forest type. The number of lianas ≥0.5 cm did not differ significantly among forest types and averaged 1348 lianas ha⁻¹, but mean liana stem diameter, basal area, estimated biomass, species richness, and Fisher’s diversity index were all greater for plots in alluvial than sandstone or kerangas forests. Liana species composition also differed greatly among the three habitats, with 71% of species showing significant positive or negative habitat associations. Sandstone forests were intermediate to alluvial and kerangas forests in most aspects of liana community structure and composition, and fewer species showed significant habitat associations with this forest type. Ranking of forest types with respect to liana density, biomass, and diversity matches the ranking in soil fertility and water availability (alluvial > sandstone hill > kerangas). These results suggest that edaphic factors play an important role in maintaining liana species diversity and structuring liana communities.     

Avifauna of two upland evergreen forest reserves,the Atewa range and Tano Offin,in Ghana

Ntiamoa-Baidu, Y., Asamoah, S.A., Owusu, E.H. & Owusu-Boateng, K. 2000. Avifauna of two upland evergreen forest reserves, the Atewa Range and Tan0 Offin, in Ghana. Ostrich 71 (1 & 2): 277–281.

Data are presented on the avifauna of two Upland Evergreen forest reserves in Ghana: Atewa Range Forest Reserve and the Tano Offin Forest Reserve. The data obtained for the two reserves using three survey techniques: transect counts, timed species counts and mist-netting are compared. A total of 105 species were recorded for the Atewa Range Forest Reserve while the Tano-Offin Forest Reserve yielded 110 species. Mean species encounter rate in the transect counts at Atewa was 24.2 per km while individual bird encounter rate was 52.6 per km; encounter rates at Tano Offin were 31.2 per km. and 98.2 per km for species and individual birds respectively. The most common species encountered in Atewa was the Yellow-whiskered Greenbul Andropadus latimstris (7.1% of observations in transect counts and 6.7% in spot counts), while the Green-crested Touraco (6.6% of transect counts) and Green Hylia (5.7% of spot counts) were the most abundant at Tan0 Offin. The records for Atewa included four globally threatened species while Tano yielded three threatened species. Two species of birds, the Equatorial Akalat Sheppardia aequatorialis and the yellow-bellied Greenbul Phyllastrepus flavostn'atus were recorded for the first time in Ghana from the two reserves. The bird community at Tan0 Ofin was significantly more diverse than that at Atewa. In both reserves, spot counts yielded more species (67.6% and 74.5% of total recorded at Atewa and Tan0 respectively), than transect counts and mist nets. 31 species recorded in both spot counts and transect counts at Atewa were not captured in mist nets, while nine species caught in mist nets were not recorded by the two methods. The figures for the Tano Offin reserve were 43 and six respectively. We conclude that of the three survey techniques, the timed species counts offer the most comprehensive results for rapid assessment of forest avifauna, but a combination of all three methods is necessary for complete coverage. The need for management of the two reserves to maintain their unique biodiversity in the face-of threats from mining and logging is discussed.     

Habitat fragmentation rather than habitat amount or habitat split reduces the diversity and abundance of ground-dwelling anurans within forest remnants of the Brazilian Cerrado

Rampant deforestation has caused the loss and fragmentation of natural habitats, which has precipitated a global biodiversity crisis. Research on how land-use change contributes to a loss of biodiversity is urgently needed, especially in ecosystems that have undergone rapid anthropogenic changes. We sought to investigate the extent to which habitat loss, fragmentation, and habitat split (the separation of forest and aquatic habitats) negatively influenced taxonomic diversity, functional diversity, total abundance, and the individual abundances of five anuran species in the Brazilian Cerrado. We sampled anurans between December 2017 and March 2018 using pitfall traps at sites distributed along a gradient of habitat fragmentation/habitat split: unfragmented forest, forest fragments without habitat split, and forest fragments with habitat split. Forest cover was measured within a 1-km radius of each site. Sites within unfragmented forests had higher taxonomic and functional diversities than either fragment type. Taxonomic diversity was highly correlated with functional diversity, but we did not find a pattern to the loss of functional traits. Total anuran abundance and the abundances of Chiasmocleis albopunctata, Physalaemus cuvieri, and Rhinella diptycha were higher in unfragmented forests compared to forest fragments. No species was more abundant in fragments than in unfragmented forests. Our results indicate that the fragmentation of forests by agricultural land use is directly and indirectly responsible for the loss of taxonomic and functional diversity, as well as for reducing population sizes of ground-dwelling anurans. Although we did not find a distinct effect of habitat split on ground-dwelling anurans, our study underscores the importance of preserving continuous forest habitats for the maintenance of anuran diversity in the Cerrado.     

Distribution and abundance of forest birds in Ghana

Ntiamoa-Baidu, Y., Owusu, E.H., Asamoah, S. & Owusu-Boateng, K. 2000. Distribution and abundance of forest birds in Ghana. Ostrich 71 (1 & 2): 262–268.

Forest reserves within the Ghanaian rain forest have been classified into categories (Condition I-VI) on the basis of their botanical importance and status of the vegetation. We present data on the distribution and abundance of avifauna of 28 Condition II and III forest reserves in southern Ghana based on line transect counts and mist-net captures. A total of 227 species were recorded for all the sites; species records for individual sites ranged from 78–119. The dominant groups in the overall survey were Muscicapidae (represented by 40 species), Pcynonotidae (21) and Cuculidae (12). The most abundant bird species in the Ghanaian forests were Yellow-whiskered Greenbul, Green Hylia, Green Pigeon and Olive Sunbird. Species encounter rates ranged from 16.7 to 50.7 species per km with individual bird encounter rates of 27.9 to 172.0 birds per km. Capture rates in mist-nets ranged from 1.2 to 3.9 birds per 100 metre net-hr (mnh). A total of 183 species were recorded in Condition II forest as compared with 167 in Condition III forests. The effects of forest condition and vegetation types on the total number of species recorded were, however, not statistically significant. The Similarity Index for bird communities within Condition II and III forests was 0.72. Bird species composition in the two forest types also differed: e.g., seven primary forest species, as well as 40% of the species associated with primary and matured secondary forest which were recorded in Condition II forests, were absent from Condition III forests. Bird communities of Moist Evergreen, Upland Evergreen and Moist semi-Deciduous were more similar than communities in the Wet Evergreen forest type. The implication of the results for conservation of the Ghana's forests is discussed.     

Butterfly populations in two forest fragments at the Kenya coast

Species richness, diversity and composition of butterflies in two Kenya coastal forest remnants, Muhaka and Mrima hill, were investigated. Sixty‐three species were recorded from each forest remnant from a total of 1329 individuals. Species accumulation curves for both forests did not reach an asymptote. High species similarity was recorded between the forest interior and the surrounding matrix, primarily due to invasion of the forest interior clearings by the savanna species. Despite their small sizes, these forest remnants were found to maintain viable populations of true forest butterflies. However, the number of species was less than half that recorded from the larger forest reserve of Arabuko‐Sokoke, located in the same geographical area. Records from Muhaka forest show species unique to it, not found in the larger forest reserves, underscoring the importance of small remnants in the preservation of forest biodiversity. The high species similarity between the forest remnants implied that if habitat corridors were created, gene flow between these remnants and other larger forest reserves would be possible. This would reduce the isolation of true forest butterfly populations within the remnants and potential local extinction.     

The role of native forest plantations in the conservation of Neotropical birds: the case of the Andean alder

The gradual increase in reforested areas worldwide, as a strategy for mitigating native forest loss, has stressed the need of assessing their real value as habitat for native species. Forest plantations, particularly those based on native species, could be valuable for conservation purposes, especially in heavily fragmented and disturbed ecosystems. We evaluated the value of a monoculture of a native tree species, the Andean alder (Alnus acuminata), for the conservation of avifauna in the Central Andes region, which is considered a bird species diversity hotspot but also suffers from high anthropogenic disturbance levels. Our results suggest that alder plantations are valuable for conservation from three points of view: (1) they have similar or greater bird species richness and abundance than secondary native forests; (2) low community similarities are found between this type of forest compared to secondary forest stands (with 27 species exclusive to alder plantations); and (3) three near threatened species (Odontophorus hyperythrus, Eriocnemis derbyi, and Cyanolyca viridicyanus). Further, 27 out of the 85 species found at the alder plantations were of least concern but showing decreasing population trends. While forest plantations do not replace native forests, they offer habitat for many bird species, some of them being of conservation concern (i.e., included in an IUCN threat category) or with decreasing populations. Hence establishing native species plantations among native forest remnants – especially in heavily fragmented landscapes – could have a positive effect in the conservation of threatened avifauna.     

Avian use of man-disturbed forest habitats in the Eastern Ghats, India

Short-term, fixed-line surveys using mist-nets and visual censuses were made of bird populations on five plots in the Eastern Ghats of Andhra Pradesh. Plots included two segments of moderately-disturbed hill forest, a remnant patch of humid ravine forest surrounded by young coffee, a mature coffee plantation with remnant natural forest overstory, and a sub-mature teak monoculture. Species-number and diversity (H') were measured to be highest in the two hill forest plots. Equitability was highest in the Teak plot. Population numbers were highest in one of the forest plots and the Coffee plot. 'Over-dominance' by tramp species was exhibited in the Coffee plot, in which the three most common bird species (two bulbuls and a white-eye) comprised more than half of the individuals censused. All censuses were dominated by non-forest and forest-edge species; few raptors and species larger than 100 g were recorded in censuses. The patch of Ravine Forest was species-poor and yet supported several unusual species (e.g., Abbott's Babbler Malacocincla abbotti. Blue Chat Erithacus brunneus and Little Spiderhunter Arachnothera longirostris ) found on no other plot. We suggest that: (1) the few remaining undisturbed tracts of moist deciduous and semi-evergreen forest should be declared natural reserves with complete protection from disturbance; (2) future forestry development in upland southeastern India should exclude teak monocultu re except on badly disturbed or clear-felled areas. Preferred development would feature coffee/ pepper plantation under maximum natural overstory, interspersed with remnant forest tracts for faunal and watershed protection. Such a combination would appear to have the least effect on the original forest avifauna.     

Seasonal use of remnant forest fragments by understorey forest birds in the Uluguru Mountains,Tanzania: a conservation priority

Forest fragmentation can lead to extinctions of some species at local levels and is eroding bird diversity at an increasing rate. While there is information on the distribution of forest bird species in most of the Eastern Arc Mountain forests, some forests, particularly the smaller fragments, have not been adequately surveyed. Using mist netting we surveyed avifauna in some of the poorly known forests (12.5–25 ha) located 320–1 300 m above sea level in the Uluguru Mountains in order to address their conservation importance. Proportions of seasonal altitudinal migrants were significantly higher in these lower-altitude forests during the cold season than the hot season. The results suggest that these forests support bird species of conservation concern, most of which are forest dependent and some of which make seasonal movements between high-altitude montane forests and lowland/ foothill forests. These forests are important cold-season habitat of altitudinal migrants and further fragmentation should be halted as a matter of regional and global priority.     

Linking Land‐Use Scenarios,Remote Sensing and Monitoring to Project Impact of Management Decisions

Large‐scale modifications of natural ecosystems lead to mosaics of natural, semi‐natural and intensively used habitats. To improve communication in conservation planning, managers and other stakeholders need spatially explicit projections at the landscape scale of future biodiversity under different land‐use scenarios. For that purpose, we visualized the potential effect of five forest management scenarios on the avifauna of Kakamega Forest, western Kenya using different measures of bird diversity and GIS data. Future projections of bird diversity combined: (1) remotely sensed data on the spatial distribution of different forest management types; (2) field‐based data on the biodiversity of birds in the different management types; and (3) forest management scenarios that took into account possible views of various stakeholder groups. Management scenarios based on the species richness of forest specialists were very informative, because they reflected differences in the proportions of near‐natural forest types among the five scenarios. Projections based on community composition were even more meaningful, as they mirrored not only the proportions of near‐natural forest types, but also their perimeter to area ratios. This highlights that it is important to differentiate effects of the total area of available habitat and the degree of habitat fragmentation, both for species richness and community composition. Furthermore, our study shows that an approach that combines land‐use scenarios, remote sensing and field data on biodiversity can be used to visualize future biodiversity. As such, visualizations of alternative scenarios are valuable for successful communication about conservation planning considering different groups of stakeholders in species‐rich tropical forests.     

Avian diversity in a Kenyan agroecosystem: effects of habitat structure and proximity to forest

Tropical agroecosystems cover an increasingly large proportion of the Earth’s terrestrial surface. Yet, relatively little is known about the factors that influence their avifauna, especially in areas of high human population density. The potential of tropical farmland for sustaining bird biodiversity, including forest birds, can be influenced by habitat structure and the distance from the nearest forest. We investigated the effect of these two factors on the bird community in the farmland near Kakamega Forest, Kenya. Using point counts, we assessed the number of bird species and individuals on 56 study plots in distances up to about 2,100 m from the forest. We observed a total of 96 bird species in the farmland, 22 of which were forest, 58 shrub-land, and 16 open-country species. High vertical vegetation heterogeneity and a large number of woody plant individuals were related to high species richness of forest and shrub-land birds, whereas open-country birds avoided such areas. The species richness and total number of forest birds declined with increasing distance to the forest. A comparison with the bird community within Kakamega Forest indicated that only a fraction of the forest species could be sustained in the farmland. This suggests that agroecosystems with a diverse habitat structure can support a high diversity of birds, but have only a limited capacity to compensate for forest loss. Electronic supplementary material The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

Small montane cloud forest fragments are important for conserving tree diversity in the Ecuadorian Andes

Montane tropical cloud forests, with their complex topography, biodiversity, high numbers of endemic species, and rapid rates of clearing, are a top global conservation priority. However, species distributions at local and landscape scales in cloud forests are still poorly understood, in part because few regions have been surveyed. Empirical work has focused on species distributions along elevation gradients, but spatial variation among forests at the same elevation is less commonly investigated. In this study, the first to compare tree communities across multiple Andean cloud forests at similar elevations, we surveyed trees in five ridge‐top forest reserves at the upper end of the ‘mid‐elevation diversity bulge’ (1900–2250 masl) in the Intag Valley, a heavily deforested region in the Ecuadorian Andes. We found that tree communities were distinct in reserves located as close as 10 to 35 km apart, and that spatially closer forests were not more similar to one another. Although larger (1500 to 6880 ha), more intact forests contained significantly more tree species (108–120 species/0.1 ha) than smaller (30 to 780 ha) ones (56–87 species/0.1 ha), each reserve had unique combinations of more common species, and contained high proportions of species not found in the others. Results thus suggest that protecting multiple cloud forest patches within this narrow elevational band is essential to conserve landscape‐level tree diversity, and that even small forest reserves contribute significantly to biodiversity conservation. These findings can be applied to create management plans to conserve and restore cloud forests in the Andes and tropical montane cloud forests elsewhere.     

Effects of forest continuity on flying saproxylic beetle assemblages in small woodlots embedded in agricultural landscapes

Ancient forests are of considerable interest for strategies for biodiversity conservation. However, in European forest landscapes fragmented and harvested for a long time forest continuity might be no longer a key driver for flying organisms such as saproxylic beetles. In a study based on paired samples (n = 60 stands, p = 180 traps) of ancient and recent forests, we investigated the effects of forest continuity on saproxylic beetle assemblages in two French regions. Mean species richness was significantly related with deadwood volume in ancient forests, but not in recent forests. This loss of relationship between assemblages and their environment suggests that dispersal limitation is at work, at least for some species. Forest continuity had a significant effect on mean species richness and on the mean number of common species, but not on rare species. Forest continuity had a significant effect on assemblage composition in one out of the four cases tested. In both regions, we identified species associated with either recent or ancient forests. Finally, mean body size of species was significantly smaller in recent forests compared with ancient ones, as was their tree diameter preference, despite a higher volume of large deadwood in recent forests. These results lend support to using forest continuity as a criterion to identify sites of conservation importance, even in highly fragmented landscapes.     

Biodiversity as a solution to mitigate climate change impacts on the functioning of forest ecosystems

Forest ecosystems are critical to mitigating greenhouse gas emissions through carbon sequestration. However, climate change has affected forest ecosystem functioning in both negative and positive ways, and has led to shifts in species/functional diversity and losses in plant species diversity which may impair the positive effects of diversity on ecosystem functioning. Biodiversity may mitigate climate change impacts on (I) biodiversity itself, as more‐diverse systems could be more resilient to climate change impacts, and (II) ecosystem functioning through the positive relationship between diversity and ecosystem functioning. By surveying the literature, we examined how climate change has affected forest ecosystem functioning and plant diversity. Based on the biodiversity effects on ecosystem functioning (B→EF), we specifically address the potential for biodiversity to mitigate climate change impacts on forest ecosystem functioning. For this purpose, we formulate a concept whereby biodiversity may reduce the negative impacts or enhance the positive impacts of climate change on ecosystem functioning. Further B→EF studies on climate change in natural forests are encouraged to elucidate how biodiversity might influence ecosystem functioning. This may be achieved through the detailed scrutiny of large spatial/long temporal scale data sets, such as long‐term forest inventories. Forest management strategies based on B→EF have strong potential for augmenting the effectiveness of the roles of forests in the mitigation of climate change impacts on ecosystem functioning.     

Comparing forest structure and biodiversity on private and public land: secondary tropical dry forests in Costa Rica

Secondary forests constitute a substantial proportion of tropical forestlands. These forests occur on both public and private lands and different underlying environmental variables and management regimes may affect post‐abandonment successional processes and resultant forest structure and biodiversity. We examined whether differences in ownership led to differences in forest structure, tree diversity, and tree species composition across a gradient of soil fertility and forest age. We collected soil samples and surveyed all trees in 82 public and 66 private 0.1‐ha forest plots arrayed across forest age and soil gradients in Guanacaste, Costa Rica. We found that soil fertility appeared to drive the spatial structure of public vs. private ownership; public conservation lands appeared to be non‐randomly located on areas of lower soil fertility. On private lands, areas of crops/pasture appeared to be non‐randomly located on higher soil fertility areas while forests occupied areas of lower soil fertility. We found that forest structure and tree species diversity did not differ significantly between public and private ownership. However, public and private forests differed in tree species composition: 11 percent were more prevalent in public forest and 7 percent were more prevalent in private forest. Swietenia macrophylla, Cedrela odorata, and Astronium graveolens were more prevalent in public forests likely because public forests provide stronger protection for these highly prized timber species. Guazuma ulmifolia was the most abundant tree in private forests likely because this species is widely consumed and dispersed by cattle. Furthermore, some compositional differences appear to result from soil fertility differences due to non‐random placement of public and private land holdings with respect to soil fertility. Land ownership creates a distinctive species composition signature that is likely the result of differences in soil fertility and management between the ownership types. Both biophysical and social variables should be considered to advance understanding of tropical secondary forest structure and biodiversity.     

A recent history of forest fragmentation in southwestern Ivory Coast

This study provides monitoring and spatial analysis of forest fragmentation based on high-resolution remote sensing, in order to give a framework for assessing residual biodiversity in the Guiglo-Taï region. Massive deforestation has taken place, mainly in the last 20 years, 79% of unprotected forests have been eliminated on the 100 × 100 km study area. Deforestation is also beginning to encroach on protected areas. Forest fragments are mainly found close to the deforestation front. In areas which were first deforested, forest fragments are very scarce and are restricted to less productive soils; they are often severely degraded. Young secondary forests develop on abandoned fields, but primary forest regeneration seems unlikely in light of the distance from seed sources and the increase in population pressure. Conservation initiatives in this zone of high biodiversity should focus on protected areas, taking into account short and long term benefits for the local population.     

Plant extinction in New Caledonia: protection of sclerophyll forests urgently needed

The sclerophyll forests which once extended over the lowlands of the west coast of New Caledonia are now reduced to small fragments representing about 2% (10 000 ha) of their original area. Much of the remaining forests are degraded. Threats to sclerophyll forests come from land clearance, grazing by cattle or deer, and fire. In sclerophyll forests, 223 endemic phanerogam species occur and 59 of these are specific to this forest type. Several of the 59 specific species are known only from a few plants at a single locality and are critically endangered. Pittosporum tanianum sp. nov. became extinct shortly after its discovery in 1988, and becomes the first documented plant extinction in New Caledonia. A further 15 species of New Caledonian plants, not recorded for several decades, are discussed, and it is concluded that between 4 and 9 of them may be extinct. The existing reserves containing sclerophyll forests are inadequate to protect the remaining biodiversity of the forests. Four immediate steps needed to protect sclerophyll forests are (i) restoration of Leprédour Island; (ii) purchase and restoration of selected privately owned forests; (iii) management of publicly owned forest near Népoui; and (iv) ex situ conservation of certain species.     

The effect of clearing on plant composition in mulga (Acacia aneura) dry forest,Australia

Despite high deforestation rates in dry forests there are few assessments of the impacts of this transformation on biodiversity. This study addresses the impact of clearing on plant species composition and diversity in mulga (Acacia aneura) forests of Queensland, Australia using data collected from either side of 58 clearing boundaries, incorporating data on site and management conditions. The physical and climatic conditions of the study sites were narrowly constrained but had greater influence on plant composition and richness than clearing per se. In general, the abundance of perennial forbs and shrubs was higher in cleared forest compared with adjacent intact forest and there was no significant difference in abundance of other plant life forms. Perennial forb richness diminished with the age of clearings, particularly where acid sub‐soils had been exposed and shrubs proliferated. Where deforestation in the mulga dry forests is not accompanied by the establishment of exotic pasture it has a relatively benign effect on plant diversity, mulga itself remains in high density as small trees and the integrity of the forest could be maintained with restoration.     

Do riparian reserves support dung beetle biodiversity and ecosystem services in oil palm-dominated tropical landscapes?

Agricultural expansion and intensification are major threats to global biodiversity, ecological functions, and ecosystem services. The rapid expansion of oil palm in forested tropical landscapes is of particular concern given their high biodiversity. Identifying management approaches that maintain native species and associated ecological processes within oil palm plantations is therefore a priority. Riparian reserves are strips of forest retained alongside rivers in cultivated areas, primarily for their positive hydrological impact. However, they can also support a range of forest‐dependent species or ecosystem services. We surveyed communities of dung beetles and measured dung removal activity in an oil palm‐dominated landscape in Sabah, Malaysian Borneo. The species richness, diversity, and functional group richness of dung beetles in riparian reserves were significantly higher than in oil palm, but lower than in adjacent logged forests. The community composition of the riparian reserves was more similar to logged forest than oil palm. Despite the pronounced differences in biodiversity, we did not find significant differences in dung removal rates among land uses. We also found no evidence that riparian reserves enhance dung removal rates within surrounding oil palm. These results contrast previous studies showing positive relationships between dung beetle species richness and dung removal in tropical forests. We found weak but significant positive relationships between riparian reserve width and dung beetle diversity, and between reserve vegetation complexity and dung beetle abundance, suggesting that these features may increase the conservation value of riparian reserves. Synthesis and applications: The similarity between riparian reserves and logged forest demonstrates that retaining riparian reserves increases biodiversity within oil palm landscapes. However, the lack of correlation between dung beetle community characteristics and dung removal highlights the need for further research into spatial variation in biodiversity–ecosystem function relationships and how the results of such studies are affected by methodological choices.