Robustness of reserve selection procedures under temporal species turnover

Complementarity-based algorithms for the selection of reserve networks emphasize the need to represent biodiversity features efficiently, but this may not be sufficient to maintain those features in the long term. Here, we use data from the Common Birds Census in Britain as an exemplar data set to determine guidelines for the selection of reserve networks which are more robust to temporal turnover in features. The extinction patterns found over the 1981-1991 interval suggest that two such guidelines are to represent species in the best sites where they occur (higher local abundance) and to give priority to the rarer species. We tested five reserve selection strategies, one which finds the minimum representation set and others which incorporate the first or both guidelines proposed. Strategies were tested in terms of their efficiency (inversely related to the total area selected) and effectiveness (inversely related to the percentage of species lost) using data on eight pairs of ten-year intervals. The minimum set strategy was always the most efficient, but suffered higher species loss than the others, suggesting that there is a trade-off between efficiency and effectiveness. A desirable compromise can be achieved by embedding the concerns about the long-term maintenance of the biodiversity features of interest in the complementarity-based algorithms.     

GPS measurement error and resource selection functions in a fragmented landscape

Advances in technology have allowed ecologists to employ remote observations of individual organism's spatial location. These data are used to model species distributions and habitat associations, which inform conservation efforts and management plans. These data are not without error. To illustrate the consequences of not considering measurement error, I introduce measurement error to a habitat selection model, using three different distributions. I show how measurement error can confound inferences made about a hypothetical organism's true habitat selection. By simulating different initial strengths of selection I show the introduction of measurement error results in the largest reduction in habitat selection strength (from truth) for very selective individuals (habitat specialists). Not surprisingly, the inclusion of error in very weakly selective individuals (habitat generalists) can result in a switching from true selection to observed avoidance. Researchers need to be aware that, first, there is measurement error in remotely observed data, and second, a tradeoff occurs between measurement error and landscape fragmentation. Landscapes with a high degree of fragmentation require spatially accurate (low measurement error) data in order to make reliable estimates of habitat selection or species distribution. The results of this study are discussed in light of the conservation of species threatened by habitat fragmentation and the management suggestions arising from selection studies.     

The value of information in reserve site selection

The reserve site selection problem is to select sites for the establishment of biological reserves with the goal to maximize the number of species contained in the reserves. When species distributions are known, this corresponds to the maximal coverage problem. In practice, knowledge of species distributions may be incomplete and only incidence probabilities are available. In this case, the goal is to maximize the expected number of species contained in the reserves. This is called the maximal expected coverage problem. This paper describes and illustrates a formal approach to assess the value of information, such as site surveys or species surveys, in this problem.     

Host preferences and behaviour of oxpeckers: co-existence of similar species in a fragmented landscape

The host preferences and behaviour of red-billed and yellow-billed oxpeckers (Buphagus erythrorynchus and Buphagus africanus) were studied at two locations in Kenya and combined with previously published data from East and South Africa. Red-billed oxpeckers are generally more common and have a greater niche breadth with respect to host preferences than yellow-billed oxpeckers, possibly due to the behavioural dominance of the latter. Otherwise, the host preferences are remarkably similar: both prefer larger species of ungulates and are found predominantly on species with manes. There are also no significant differences in the time–activity budgets of the two species controlling for site and host species, and differences in the proportion of time spent on different body parts of hosts are small in absolute magnitude even when statistically significant. Despite these and other basic ecological similarities, no evidence for competitive displacement was detected by comparing host preferences in areas of sympatry with areas of allopatry, even though each species appears to depress the population size of the other when sympatric. Traditional single-community competition theory cannot explain the geographical ecology of these two species. Instead, co-existence appears to be dependent on the patterns of extinction and colonization characterized by metapopulations distributed among a fragmented habitat.     

Effects of kin-structure on disease dynamics in raccoons (Procyon lotor) inhabiting a fragmented landscape

The spatial aggregation of related individuals can affect disease dynamics by altering contact rates between individuals and/or groups of individuals. However, kin-structure likely affects contact rates in a scale-dependent manner: increasing contact rates within, but decreasing contact rates amongst spatial groups. Thus, the effect of kin-structure on disease exposure risk should depend upon the spatial scale relevant for pathogen transmission. This study was undertaken to test the effects of kin-structure on disease dynamics in raccoons (Procyon lotor) inhabiting a highly fragmented agricultural ecosystem. Raccoons (n = 566) were trapped at 31 spatially discrete habitat patches (local populations). Trapped individuals were genotyped at 13 microsatellites and their serum screened for antibodies to canine distemper virus (CDV) and multiple Leptospira spp. serovars. Our analyses revealed that the effects of kin-structure on transmission dynamics are dependent on both spatial scale and the pathogen under consideration. Thus, in the directly transmitted pathogen (CDV), the probability of pathogen co-exposure (the proportion of individual pairs where both individuals were seropositive for a particular pathogen) was positively correlated with genealogical proximity within local populations. Conversely, in the environmentally transmitted leptospires co-exposure was homogeneous within local populations and was positively correlated with gene flow among these populations. Our analyses also revealed that, after controlling for confounding effects of landscape and demographic variables, the effects of kin-structure on disease exposure risk was pathogen-specific. In particular, increasing kin-structure was associated with increased disease exposure risk in the directly transmitted pathogen (CDV), and alternatively with decreased disease exposure risk in environmentally transmitted leptospires. The potential, scale-dependent, effects of kin-structure on disease exposure risk revealed by our analyses could play an important role in elucidating disease dynamics in natural populations and in planning effective disease mitigation strategies.     

Recruitment limitations in Primula vulgaris in a fragmented landscape

Plant recruitment is limited by dispersal, if seeds cannot arrive at potential recruitment sites, and by establishment, due to a low availability of safe sites for recruitment. Seed-sowing experiments, scarcely applied along gradients of landscape alteration, are very useful to assess these limitations. Habitat loss and fragmentation may foster recruitment limitations by affecting all the processes from seed dispersal to seedling establishment. In this study, we perform a seed-sowing experiment to disentangle the importance of dispersal and establishment limitations in different stages of recruitment of the perennial herb Primula vulgaris in fragmented forests of the Cantabrian Range (Northwestern Spain). We evaluated the influence of ecological gradients resulting from habitat loss and fragmentation (modifications of habitat amount at the landscape and microhabitat scales, changes in the species’ population size, changes in seed predation and seedling herbivory) on seedling emergence, survival and early growth. We found strong evidence of dispersal limitation, as seedling emergence was very low in experimental replicates where no seeds were added. This limitation was independent of landscape alterations, as we found no relation with any of the ecological gradients studied. Establishment limitations at the germination phase were also unrelated to ecological gradients, probably because these limitations are more related to fine-scale environmental gradients. However, further monitoring revealed that seedling survival after summer and winter periods and seedling growth were conditioned by landscape alteration, as we found effects of habitat amount at the landscape and microhabitat scales, of presence of populations of P. vulgaris and of seedling herbivory. These effects were complex and sometimes opposite to what can be expected for adult plants, revealing the presence of different requirements between life stages.     

Extinction,turnover and species diversity in an experimentally fragmented California annual grassland

Summary Many natural populations are subdivided among partially isolated habitat patches, but the influence of habitat patchiness per se on species immigration, extinction, and the resulting patterns of species diversity, has received virtually no experimental study. In an experiment designed to test the effects of habitat subdivision on local community structure, we compare the diversity and annual turnover of flowering plant species in 3 treatments of the same total area, but subdivided to different degrees. We experimentally fragmented a California winter annual grassland into isolated plots, two of 32 m², eight of 8 m², and 32 of 2 m², each treatment representing a combined area of 64 m². Insularization of the experimental habitat fragments is provided by grazing sheep. The effects of plot area on species diversity, extinction, and turnover are consistent with the MacArthur-Wilson model. Species richness increases with the degree of habitat subdivision. Extinction, immigration, and turnover, however, are relatively independent of the degree of subdivision. These experimental results contrast with predictions that habitat subdivision necessarily results in greater rates of extinction accompanied by reduced species diversity.     

A land snail's view of a fragmented landscape

Habitat fragmentation may influence the genetic structure of populations, especially of species with low mobility. So far, these effects have been mainly studied by surveying neutral markers, and much less by looking at ecologically relevant characters. Therefore, we aimed to explore eventual patterns of covariation between population structuring in neutral markers and variation in shell morphometrics in the forest‐associated snail Discus rotundatus in relation to habitat fragment characteristics. To this end, we screened shell morphometric variability and sequence variation in a fragment of the mitochondrial 16S rDNA gene in D. rotundatus from the fragmented landscape of the Lower Rhine Embayment, Germany. The 16S rDNA of D. rotundatus was highly variable, with a total of 118 haplotypes (384 individuals) forming four clades and one unresolved group. There was a geographic pattern in the distribution of the clades with the river Rhine apparently separating two groups. Yet, at the geographic scale considered, there was no obvious effect of fragmentation on shell morphometrics and 16S rDNA variation because G_ST often was as high within, as between forests. Instead, the age of the habitat and (re‐)afforestation events appeared to affect shell shape and 16S rDNA in terms of the number of clades per site. The ecologically relevant characters thus supported the presumably neutral mitochondrial DNA markers by indicating that populations of not strictly stenecious species may be (relatively) stable in fragments. However, afforestation after large clearcuts and habitat gain after the amendment of deforestation are accompanied by several, seemingly persistent peculiarities, such as altered genetic composition and shell characters (e.g. aperture size). © 2009 The Linnean Society of London, Biological Journal of the Linnean Society, 2009, 98 , 839–850.     

Effects of population size and pollinator limitation on seed-set of Primula sieboldii populations in a fragmented landscape

Female reproductive success, fruit and seed-set and factors affecting them were measured for 3 years (1995–1997) for 24 populations of Primula sieboldii E. Morren, a threatened Japanese plant species in a highly fragmented landscape in southern Hokkaido. The smaller populations (genets <=; 3) set almost no seeds consistently during the study years. In larger populations (genets >= 7), proportions of failed flowers without damage from fungi or herbivores were relatively high, but fruit set per flower varied among the populations in 1995. Correlations of fruit (r = 0.589, P = 0.011) and seed- set (r = 0.688, P < 0.01) with population pollinator availability were highly significant. In 1995, pollination failure would be the most plausible reason for the variation in reproductive success among the populations. However, pollinator availability was relatively high in all populations in 1996 and 1997. In these years, higher proportions of fruit failed because of antagonistic biological interactions. In the populations with poor pollinator availability in 1995, mean fruit and seed-sets were greater in the long-styled morph than in the short-styled morph. Such between-morph differences are thought to be caused by partial self-compatibility of the long-styled morph and the results of hand pollination experiments support this hypothesis.     

Habitat use by beech martens in a fragmented landscape

We investigated key habitat features used by beech martens occupying a fragmented agricultural landscape. Locations of eight martens were radio-monitored daily for 4–12 months each. We evaluated the selection of home ranges within a geographic area, as well as the selection of habitat features within home ranges, using random simulations for comparison. Home ranges included more wood and scrub vegetation, and less arable land, than random simulations. They also included a higher proportion of watercourses with continuous vegetation along their verges, and were closer to these watercourses than random simulations. Within home ranges, beech martens used wood and scrub vegetation, and farm buildings, more intensively than expected, and arable land less than expected. On average, beech martens stayed close to watercourses (where most of the wood vegetation occurred) but not to farm buildings, indicating that only the former shaped beech marten movements. By these findings, arable land appeared to act as a seldom traversed barrier, making farm buildings uneasy to reach if they are located far from watercourses.     

Migration, coherence and persistence in a fragmented landscape

The chance of local extinction is high during periods of small population size. Accordingly, a metapopulation made of local communities that support internal population cycling may face the threat of regional extinction if the local dynamics is coherent (synchronized). These systems achieve maximum sustainability at an intermediate level of migration that allows recolonization but prevents synchronization. Here we implement an individual-based simulation technique to examine the maximum persistence condition for a system of patch habitats connected by passive migration. The models discussed in this paper take into consideration realistic elements of metapopulations, such as migration cost, disordered spatial structure, frustration and environmental noise. It turns out that the state with maximum anti-correlation between neighboring patches is the most sustainable one, even in the presence of these complications. The results suggest, at least for small systems, a model independent conservation strategy: coherence between neighboring local communities has, in general, a negative impact, and population will benefit from intervention that increases anti-correlations.     

Influence of environmental and spatial variables on regional distribution of forest plant species in a fragmented and changing landscape

During the past several centuries, forests in Europe and large parts of North America have been subject to extensive forest clearance. The last several decades, however, numerous new forest patches have been established onto former agricultural land. As a result, the present forest area often consists of a mixture of small forest patches of different age, area, habitat quality and connectivity embedded within a hostile agricultural landscape. In these patchy landscapes, distribution patterns of plant species may be affected by both regional and local factors, although the relative importance of both is still poorly understood. In this study, we investigated distribution patterns of 113 forest plant species in a fragmented landscape. Species abundances at the regional scale conformed to a clearly unimodal abundance distribution which we believe to be related to 1) environmental heterogeneity due to succession and 2) inequality in migration rates. Patch incidence was significantly related to life form, which in turn was correlated to seed mass and dispersal mechanism. Multiple logistic regressions showed that presence/absence of 59 species studied was significantly affected by patch connectivity, patch area and age for 35, 30 and 34 species, respectively. The results of this study indicate that distribution patterns of forest plant species are influenced by both local and regional factors. Moreover, they also demonstrate that next to spatial aspects of fragmentation, temporal patterns of landscape change may have far-reaching effects on presence/absence patterns of plant species and therefore should be incorporated in studies dealing with regional population structures of plants.     

Roe deer population structure in a highly fragmented landscape

Northern Belgium (Flanders) is one of the most densely populated and urbanized regions in Europe. Many species are therefore likely to suffer from anthropogenic pressure and habitat destruction and fragmentation. Although many large mammals are recolonizing in parts of Europe, including Belgium, due to adaptation, a relaxation of persecution and habitat restoration, we have little actual data concerning the effects of landscape features on their population structure. We analysed the genetic structure of discrete roe deer (Capreolus capreolus) populations in the Eastern part of Flanders, with special emphasis on the impact of habitat fragmentation and anthropogenic barriers. The sampled populations were clearly genetically differentiated. Genetic structure could be explained by purely distance-based landscape modelling, but a simpler model focusing solely on barrier effects of large transportation infrastructure explained nearly as much genetic variance. In contrast, analyses based on least-cost landscape modelling failed to yield a significant effect. Overall, the results suggest considerable landscape-level effects of transportation infrastructure.     

Host-parasitoid extinction and colonization in a fragmented prairie landscape

Few field studies of natural populations have examined the factors influencing local extinctions and colonization of empty habitat patches for a prey species and its predator. In this study, I carried out a census of planthopper (Prokelisia crocea; Hemiptera: Delphacidae) and egg parasitoid (Anagrus columbi; Hymenoptera: Mymaridae) incidence and densities in 147 host-plant patches (Spartina pectinata; Poaceae) over seven planthopper generations in a tall-grass prairie landscape. For both species, the likelihood of going extinct in a patch was related to a number of patch-specific variables: density, temporal variability in density, proportion of hosts parasitized (planthopper only), host-plant density, patch size, patch isolation, and composition of the surrounding matrix. Colonization likelihood was related only to the physical attributes of the patch. There was high patch turnover in this prairie landscape. On average, planthoppers went extinct in 23% of the patches and A. columbi went extinct in 51% of the patches in each generation. For the planthopper, extinction likelihood increased with a decrease in patch size and the proportion of the matrix composed of mudflat. Parasitism of eggs had no effect on the extinction likelihood of local P. crocea populations, suggesting that A. columbi may not play a major role in the patch dynamics of its host. The likelihood of extinction for A. columbi was dependent on factors that spanned three trophic levels. An increase in plant density, decrease in host density and decrease in parasitoid density all increased the likelihood of A. columbi extinction within a patch. The dependency on multiple trophic levels may explain the higher extinction risk for the parasitoid than its host. A. columbi extinction was also affected by the matrix habitat surrounding the patch—the effect was the opposite of that for P. crocea. Finally, vacant patches were colonized at rates of 53% and 34% per generation for the planthopper and parasitoid, respectively. For both species, colonization probabilities decreased with an increase in patch isolation. High host densities in a patch also favored high rates of colonization by A. columbi. I discuss how anthropogenic changes to the prairie landscape can affect the metapopulation dynamics and persistence time of this host-parasitoid interaction.     

Population genetics of orchid bees in a fragmented tropical landscape

The prospects for persistence of bees living in fragmented landscapes is a topic of considerable interest due to bees’ importance as pollinators of agricultural crops and wild plants, coupled with the ubiquity of native habitat loss and evidence that bees may be declining worldwide. Population persistence in fragmented areas depends on dispersal potential and maintenance of gene flow among fragments of habitat. Here we used population genetic techniques to characterize, for two equally abundant orchid bee species that differ in their physiology and ecology, levels of genetic differentiation among fragments of tropical forest in southeastern Costa Rica in a ~200 km² landscape. We measured population differentiation with ϕ_PT (an analogue to the traditional summary statistic Fst), as well as two measures that may more accurately reflect the level of differentiation when highly variable loci are used: G’st and D_est. We also calculated pairwise genetic distances among individuals and conducted Mantel tests to test the correlation of genetic and geographic distance, for each species. We found strong differences in genetic structure between the species. Contrary to our expectations, each measure of genetic structure revealed that the larger-bodied species, Eulaema bombiformis, had higher levels of differentiation than the smaller species, Euglossa championi. Furthermore, for Eulaema bombiformis there was a significant positive correlation of genetic and geographic distance while for Euglossa championi there was no significant positive correlation. Our results demonstrate that bee species can have strikingly different levels of gene flow in fragmented habitats, and that body size may not always act as a useful proxy for dispersal, even in closely related taxa.     

Mermithid parasitism of Hawaiian Tetragnatha spiders in a fragmented landscape

Hawaiian Tetragnatha spiders inhabiting small forest fragments on the Big Island of Hawaii are parasitized by mermithid nematodes. This is the first report of mermithid nematodes infecting spiders in Hawaii, and an initial attempt to characterize this host-parasite interaction. Because immature mermithids were not morphologically identifiable, a molecular identification was performed. A phylogenetic analysis based on 18S small ribosomal subunit nuclear gene sequences suggested that Hawaiian spider mermithids are more closely related to a mainland presumptive Aranimemis species that infects spiders, than to an insect-infecting mermithid collected on Oahu, HI, or to Mermis nigrescens, also a parasite of insects. Measured infection prevalence was low (ranging from 0 to 4%) but differed significantly among forest fragments. Infection prevalence was associated significantly with fragment area, but not with spider density nor spider species richness. Results suggest that mermithid populations are sensitive to habitat fragmentation, but that changes in infection prevalence do not appear to affect spider community structure.