Representing overlapping polylines in QGIS

Representing overlapping polylines in QGIS

I have several shapefiles representing bus lines with different colors for each segment, however, because the lines are exactly superposed, only the shapefile on top can be seen. Is there a way to display the lines underneath for example side-by-side on these parts of segments where several lines converge (there might be 4 or 5 of them)? The polylines are made up of series of segments.

I know there are some tools in ArcGIS, but I'm using QGIS 2.8.0

In QGIS 2.8.0 you can do that with the offset option in the Style Layer Properties. In my example with three polylines (they are exactly superposed too):

Click in "Simple line" of line2 to display "Offset" option (I set 1 mm):

Click in "Simple line" of line5 to display "Offset" option (I set -1 mm):

The result obtained (Offset of line4 unchanged; 0 mm):

Editing note (answering to alpha-beta-soup comment):

With the next code, executed in the Python Console of QGIS, I have the same result:

from PyQt4.QtGui import * mc=iface.mapCanvas() layers=[] renderer = [] n = mc.layerCount() symbol = range(n) symbol[0] = QgsLineSymbolV2.createSimple({'color':'green', 'width':'1', 'offset':'1.0'}) symbol[1] = QgsLineSymbolV2.createSimple({'color':'red', 'width':'1', 'offset':'0.0'}) symbol[2] = QgsLineSymbolV2.createSimple({'color':'blue', 'width':'1', 'offset':'-1.0'}) for i in range(n): layers.append(mc.layer(i)) for layer in layers: renderer.append(layer.rendererV2()) for i in range(n): renderer[i].setSymbol(symbol[i]) iface.mapCanvas().refresh() for layer in layers: iface.legendInterface().refreshLayerSymbology(layer)

Assessing social-ecological vulnerability of coastal systems to fishing and tourism

Detecting areas with high social-ecological vulnerability (SEV) is essential to better inform management interventions for building resilience in coastal systems. The SEV framework, developed by the Intergovernmental Panel on Climate Change, is a robust method to identify SEV of tropical coastal systems to climate change. Yet, the application of this framework to temperate regions and other drivers of change remains underexplored. This study operationalizes the SEV framework to assess the social-ecological implications of fishing and tourism in temperate coastal systems. We spatially represented the SEV of coastal systems and identified the social and ecological vulnerability dimensions underpinning this SEV. Our results demonstrate that different dimensions contribute differently to the SEV, suggesting the need for distinctive management intervention to reduce the vulnerability of coastal systems. Our findings also highlight that livelihood diversification and the protection of marine areas may be plausible strategies to build resilience in temperate coastal systems that face fishing and tourism pressures. With this study, we hope to encourage the application of the SEV framework to other drivers of change for building more resilient coastal systems.


Climatic factors may act as ecological barriers that can determine the distribution of animal and plant species 1 . This may explain why incipient speciation processes often can be inferred from the analysis of patterns of niche divergence 2 . Genetic variation, species spatial structure resulting from landscape barriers to gene flow, and intraspecific evolutionary processes (e.g., local adaptation) are the major drivers of speciation. Therefore, it is not surprising that genetic studies combined with ecological niche models have been proven to be powerful tools to study evolutionary processes and resolve biodiversity conservation problems 3 . In addition, the characterization of environmental niches is essential to understanding species distributions and patterns of biological diversity. Correlative ecological niche modelling 33 is a common tool used to approach this characterization 3 . However, is the species level the most adequate level for this approach? Smith et al. 4 suggest that we must consider local adaptations as evolutionary factors affecting niche requirements, and therefore, inclusion of evolutionary relationships below and above the species level should be considered. For this reason, it is informative to compare niche modes across taxa in a separate way, considering species and local genetic lineages. Amongst vertebrates, amphibians are ideal organisms to analyse this question because their physiology is highly constrained by environmental factors 5,55 , and they have low dispersal abilities. This combination of characteristics is expected to promote the evolution of local adaptations to match the spatial complexity of environmental variation 7 .

It is well known that vicariant events can drive divergence by limiting genetic exchange among evolutionary units 8 , but currently, there is evidence that environmental factors can also play a key role in biological diversification 9,10,11 . Although habitat suitability models and niche similarity comparisons have been previously conducted at interspecific and intraspecific levels for the midwife toads (Alytes sp.), these models have been implemented only in A. obstetricans 9 . However, other Mediterranean species, such as A. cisternasii and A. dickhilleni, have not been studied using nested models (inter- and intraspecific climatic niche divergence schemes) despite the fact that their genetic and phylogenetic discontinuities are well documented 12,13 . Furthermore, an interspecific perspective of climatic niches for all the species in the genus is crucial for a better understanding of climatic determinants and the differentiation processes involved.

Recent studies on niche modelling have shown that environmental conditions can drive evolution across geographical ranges and affect patterns of genetic structure 14,9,15 . Moreover, genetic isolation and local adaptation can synergistically influence the fate of species 16 , and biogeographic and vicariant events can drive species differentiation (see Martínez-Solano et al. 17 ). However, the environmental factors involved in the maintenance of the current differenciation of species and evolutionary units have not been tested as a whole. Tectonic factors, with the formation of a mountain range in the Gibraltar Strait during the Upper Tortonian stage, played an important role in the speciation of the genus Alytes 18,17 . In fact, the geographic isolation of A. muletensis and A. maurus might have strongly affected their environmental niches. Finally, recent work suggests that a more complex geological scenario might have affected the evolutionary history of Alytes in its southern range, with a Pliocene volcanic archipelago between Cabo de Gata and the eastern Rif coast 19 .

The major aims of this study were (1) to characterize the realized niche differences and environmental factors that promote the differentiation and the observed distribution of Alytes species, and (2) to investigate the importance of niche evolution by testing the hypotheses niche conservatism, as the maintenance of ancestral requirements among species with a common ancestor 20 , and niche divergence, as the appearance of divergences among these species 2 . Additionally, we tested whether the observed patterns of niche environmental evolution were consistent at two phylogenetic scales: interspecific and intraspecific (i.e., genetic lineages from Dias et al. 12 Gonçalves et al. 13 and Maia-Carvalho et al. 9 ). We assumed the existence of differentiation at both inter- and intraspecific levels as a consequence of climatic niche differentiation. In this scheme, intra- and interspecific differentiation may be influenced by climate and geographic barriers throughout genetic variation and structure. Finally, We also aimed to determine whether the processes of niche and phylogenetic evolution were parallel by predicting whether subclades would show a phylogenetic signal.


Species distribution models have become increasingly noticeable in ecological and biogeographical research 33 . Mostly because ecologists need ways of rapidly assessing the impacts of climate change on large numbers of species for which the occurrence data are often the only source of information 34 . While several critical opinion on ENM analyses were presented in the last years 35 , MaxEnt seems to be the most reliable application for modeling species distribution. Its usefulness was also tested in case of rare organisms 36 ,37 . Noteworthy, some of the studies indicating inappreciable usefulness of ENM tools 38 (Beale et al.) were called into question by the subsequent researchers 39 .

In our research we tested three approaches to evaluate the past, present and future distribution of the suitable niches of poorly known lichen species. Some of the previous studies indicated that the correlations between the environmental data used in the modeling should be reduced and that the correlated variables should be excluded from the analysis 40 . Also it was suggested that using restricted area in ENM analysis is more reliable than calculating habitat suitability in the global scale 41 . According to the received AUC scores the most reliable model was created based on all available climatic and altitudinal data and it was constructed for the whole globe. We therefore would postulate that all potentially useful climatic variables and altitudinal data should be used in ENM studies, especially when ecological information about the studied taxon are poor and they do not allow to discriminate any climatic data as irrelevant. Noteworthy, despite slight differences in the trustworthiness of the three conducted analyses all created models indicated similar areas that could be occupied by Ochrolechia austroamericana. In our opinion lichen species distribution modeling with MaxEnt may be extended into new fields and it would be especially useful in reconstructing their past distribution and potential migration routes. While future habitat loss of the lichens became emerging question in lichenology 42 ,43 , so far MaxEnt was not implemented in any of those studies. While the limitations of such predictions are well-known and their verification is not possible in the present time, we believe that the vulnerability of specific areas indicated in the modeling should be taken into consideration prior to planning conservation actions.

Despite Ochrolechia austroamericana has been reported from rather limited number of localities 23 ,24 , the ENM method has shown its potential distribution range can be much wider.

The analyses have shown the suitability of habitats on almost every continent however its occurrence is highly improbable in most regions as the genus Ochrolechia A.Massal. was a subject of several taxonomic treatments over the past c. 30 years 24 ,25 ,44 ,45 ,46 ,47 ,48 and O. austroamericana was never found outside South America. In general, most Ochrolechia species have distributions rather restricted to one continent or region, and that can be possibly related to the limited dispersal of diaspores (ascospores), which are relatively large 49 . Only very few species, e.g. the tropical O. africana Vain., are known to have wider range 25 ,45 but if they truly represent one evolutionary lineage or several cryptic species, has not been settled yet. However, in the light of recent studies, which showed several lichenized fungi to represent numerous phylogenetically distinct clades 50 ,51 , we suspect this scenario to be more adequate in this case.

Concerning the most probable potential distribution range we consider that O. austroamericana occurs only South America from Venezuela to southern Argentina with, as our modeling has shown, the highest concentration of the suitable habitats in Central Andes. As it appears to be restricted to cooler climate conditions, it could not spread more northward due to the lack of appropriate spreading passages in Central America.

Apparently, the current distribution of the suitable habitats of O. austroamericana results from location of its glacial refugia and no long-distance dispersal of this lichen is observed. Its niches coverage decreased by almost 25% since LGM. Most of the loss is observed within the Pampas and in the high regions of Central and Southern Andes. We interpret the first issue as related with the coast line regression after Late Glacial that significantly affected local climatic conditions 52 . The loss of the habitats along the Andes may be caused by the warming of the high-Andean regions that was documented for numerous South American regions and which has been intensified in the last three decades, and, as the consequence, the vertical shift of vegetation belts in the altitudinal gradient, i.e. uppering the forest line 53 ,54 ,55 ,56 ,57 ,58 ,59 ,60 .

Surprisingly, the future distribution of O. austroamericana will not change much from the present range of the species. The difference between the three created models for 2080 differ in the suitability of the Atacama Desert and lower parts of the eastern slopes of Central Andes for the studied species occurrence. The studies of Boulanger et al. 61 indicated that in twenty-first century the amplitude of the seasonal cycle will tend to increase in southern South America, while in northern South America the amplitude of the seasonal cycle would be reduced that explains the relatively little changes in the general distribution of the species. However, while all emission paths tend to show the same pattern of warming, the highest warming is predicted in A2 storyline. This scenario is the only one which predicts increase of the suitable niches cover for O. austroamericana. We interpret it as a result of a tree-line location change 62 as the lowering of the forest limit due to the climate change will reduce vegetation cover and may expose additional areas appropriate for O. austroamericana.


The isoscapes revealed distinct spatial patterns in N concentration, δ 15 N and δ 13 C of C. album foliage ( Fig 1 ). In the invaded plots 1 and 2, N concentration ranged between 5.8�.5 g N*kg -1 and 6.3𠄹.9 g N*kg -1 , respectively, and increased considerably when comparing C. album individuals growing distant and adjacent to A. longifolia (hatched area, Fig 1A and 1B ). These alterations in N concentration corresponded to a more than two-fold increase for plants located close to the invader compared to the uninvaded vegetation at the borders of plot 1, and an enrichment by ca. 30% in plot 2. Similarly, δ 15 N of C. album became substantially enriched in the vicinity of A. longifolia, with background values of uninfluenced vegetation of ca. -11‰ (plot 1) and -8‰ (plot 2) increasing to values close to 0‰ ( Fig 1D and 1E ), even though C. album itself has no capacity of N2-fixation. For δ 13 C, a distinct spatial pattern with values enriched by ca. 2.5‰ for C. album growing close to the A. longifolia canopies was evident for plot 1, while for plot 2, pronounced small-scale variation only weakly related to the presence of the invader was observed ( Fig 1G and 1H ). Noticeably, enrichment in N, δ 15 N and δ 13 C associated with A. longifolia presence was evident not only in direct neighborhood of the invader, but exceeded the canopy by several meters.

Canopies of Acacia longifolia in plot 1 and 2 are indicated by white hatched polygons.

In plot 3, which is not invaded by A. longifolia, N concentration was extremely low (≤ 7.3 g N*kg -1 ) without much variation throughout the study plot ( Fig 1C ). δ 15 N showed weak spatial pattern, with enriched values mainly occurring in the southern part of the plot ( Fig 1F ). However, δ 15 N-enrichment was less in terms of absolute values and spatial extent compared to invaded plots 1 and 2. δ 13 C showed a similar pattern of enrichment, with depleted values in the northern and enriched values in the southern part of the plot and one peak showing maximal values of ca. -25.5‰ ( Fig 1I ).

Model-based cluster analysis resulted in optimal solutions with six clusters in plot 1 and four clusters in plot 2 and 3 according to the BIC (S1 Fig). Optimal models were ellipsoidal for all three plots, with variable volume, shape and orientation for plot 1, equal volume and shape for plot 2 and equal shape for plot 3 (S1 Fig). Fig 2 shows scatterplots of the three variables used for the cluster analyses, indicating cluster membership of each sample. In all three plots, N concentration and δ 15 N were correlated and clusters were arranged roughly along the concerted increase of these variables, with the exception of cluster II in plot 3 ( Fig 2A� ). Scatterplots of δ 15 N vs. δ 13 C and N concentration vs. δ 13 C similarly showed correlations for plot 3 ( Fig 2F and 2I ), with cluster II again accommodating a rather outlying group ( Fig 2F ). Plots 1 and 2 demonstrated a more complex relationship between δ 13 C and δ 15 N: for plot 1, in clusters I, II and III, a positive correlation between δ 15 N and δ 13 C was found, with cluster III being separated from clusters I and II by a higher enrichment in δ 15 N. In contrast, clusters IV, V and VI tended to decrease in δ 13 C with more enriched values in δ 15 N ( Fig 2D ). The same trend was evident in the scatterplot of N concentration vs. δ 13 C: clusters I, II and III showed increasing values in N concentration with enrichment in δ 13 C, while clusters IV, V and VI on average got depleted in δ 13 C with further increase in N concentration ( Fig 2G ). A similar pattern, though less pronounced, could be observed in plot 2 ( Fig 2E and 2H ). While across clusters I, II and III, values tended to correlate positively between δ 13 C and δ 15 N as well as δ 13 C and N concentration, cluster IV, with highest enrichment in δ 15 N and N concentration, did not show further enrichment in δ 13 C.

Cluster membership to cluster I to VI (for plot 1) or cluster I to IV (for plots 2 and 3) is indicated by different symbols and coloring, sorted by median N concentration, with blue to red representing low to high N. Ellipses denote the normal-probability contours at probability = 0.5. Please note different scaling of the axes.

Fig 3 shows boxplots of N concentration, δ 15 N and δ 13 C within the individual clusters. For all plots, median N concentration obviously increased with increasing cluster number, as clusters have been ordered by N concentration ( Fig 3A� ). In plot 1, δ 15 N similarly increased from cluster I to V, while cluster VI did not show additional enrichment ( Fig 3D ). Median δ 13 C increased from cluster I to II and then to IV, but decreased in clusters V and VI. Median δ 13 C of cluster III was intermediate between clusters I and II ( Fig 3G ). The distance to the closest A. longifolia canopy of each sample, although it was not a variable in the cluster analysis, mirrored the pattern of medians of δ 15 N, with members of cluster I showing the largest distance, then median distances decreased for clusters II through IV, and members of clusters IV, V and VI were located closest to A. longifolia.

Different letters indicate significant differences at P < 0.05 (Kruskal-Wallis test, non-parametric multiple comparisons corrected for α-inflation).

Patterns in plot 2, while less pronounced, were similar. N concentration increased with cluster number, though it did not differ significantly between clusters I and II ( Fig 3B ). For δ 15 N, median values were lowest in cluster II and increased in clusters III and IV ( Fig 3E ). δ 13 C increased from cluster I through III and was not significantly different between clusters III and IV, i.e. δ 13 C did not get further enriched with additional increase in N concentration ( Fig 3H ). Cluster II was the farthest and cluster IV the closest to the A. longifolia canopies, and again, the distances accurately mirrored the distribution of δ 15 N ( Fig 3K ).

In the uninvaded plot 3, i.e. without the influence of A. longifolia, δ 15 N and δ 13 C increased with increasing N concentration in clusters I, III and IV, however, in group II, relatively low N concentration was associated with the—on average—most enriched δ 15 N and δ 13 C values ( Fig 3C, 3F and 3I ).

Medians of the clusters from the model-based approach were submitted to a hierarchical cluster analysis with the aim to summarize clusters across the three plots. Hierarchical cluster analysis resulted in three final clusters ( Fig 4 ). Generally, the clustering approach yielded spatially homogeneous clusters, though no information on spatial location had been included in the analyses ( Fig 5 ). Final cluster one, which was characterized by low values in all variables and specifically, low δ 15 N with values close to -10‰, was represented at the northern part of plot 1 and covered roughly half of the area of uninvaded plot 3. Final cluster two showed medium δ 15 N and δ 13 C with low to medium N concentration and was present in all three plots, occupying the largest area overall. Final cluster three, with high N concentration and strongly enriched δ 15 N and δ 13 C, only occurred in invaded plots 1 and 2. In these plots, cells assigned to final cluster three spatially corresponded to the locations of A. longifolia canopies and their surroundings.

The optimal solution with k = 3 clusters as identified by highest Silhouette value is indicated by different coloring. Labels specify plot and initial cluster membership in the form plot.cluster using Arabic and Roman numerals, respectively.

Framed areas marked with Roman numerals illustrate membership of cells to clusters yielded by the initial model-based cluster analysis (within plots). Different colors indicate membership to clusters from the final hierarchical cluster analysis (across plots), which was calculated with median values derived from the model-based clustering. Canopies of Acacia longifolia in plots 1 and 2 are indicated by white polygons.


Regional scale, Nzeeu River only

Data from MOLUSCE analyses show a significant change in the land cover pattern between 1961, 1980 and 2015/2016, based on the detailed land cover mapping along Nzeeu River at the regional scale (Fig. 2). 59.1% of the riparian forest cover identified for the year 1961 was transformed into agricultural land by 2015/2016. In parallel, 36.6% of agricultural land of the year 1961 became invaded by L. camara until 2015/2016 (Table 1). Only 34.5% of the riparian forests assessed for the year 1961 remained throughout until 2015/2016. The number of human settlements increased from 36 to 189 between 1961 and 2015/2016 (i.e. by 425%).

Regional scale, Nzeeu River 1961– 1980 (1961–1980) 2015/2016 (1980–2015/16) – (1961–2015/16)
Riparian vegetation 51.24 35.75 (−30.2%) 34.57 (−3.3%) − (−32.5%)
Agricultural land 40.67 54.69 (+34.5%) 56.72 (+3.7%) − (+39.5%)
Settlements 0.92 1.76 (+91.3%) 1.79 (+1.7%) − (+94.6%)
  • First values indicate the area (ha). Gains and losses between time cohorts are given below as percentage in parenthesis.

FRAGSTAT results indicated a severe change in the habitat configuration along Nzeeu River between 1961 and 2015/2016. The overall number of patches (including all land cover categories) increased from 138 to 218. At the level of land cover category, the number of patches (NP) for agricultural land increased from 54 to 87, for human settlement from 7 to 29, and for riparian vegetation from 62 to 88. Effective Mesh Size (MESH) decreased from 0.70 to 0.62 (Table 3). At class level, the Mann–Whitney U test revealed significant increases of the number of patches (NP) of agricultural land between 1961 and 1980, and between 1980 and 2015/2016, and for riparian vegetation between 1980 and 2015/2016. Human settlements significantly increased between 1980 and 2015/2016.

Landscape scale, Nzeeu River and Kalundu River

Landscape scale data obtained from the randomly set plots along Nzeeu River and Kalundu River showed congruent trends as revealed at the regional scale (see above). Data from MOLUSCE analyses show a significant increase in the proportion of agricultural land from 48.2% to 64.2%, while riparian vegetation coverage decreased from 44.0% to 23.8% of the total land area between 1961 and 2015/2016 (Table 2). Furthermore, 58.5% of former pristine riparian thickets became transformed into agricultural land, but 18.8% of agricultural land was covered again by thicket afterwards. Only 28.9% of the pristine riparian thickets assessed in the year 1961 existed continuously until 2015/2016. Validation showed a 95.5% correctness of the comparison, an overall Kappa value of 0.92, 0.94 for Kappa histogram, and 0.98 Kappa location. Relative rasterization error ranged from −0.0012 to 0.00056 for land cover categories (Table 3).

Land cover category Nzeeu River (1961–2015/16) Kalundu River (1961–2015/16) Overall (1961–2015/16)
Riparian vegetation 42.90 (−43.63) 45.07 (−48.22) 43.98 (−45.95)
Agricultural land 49.93 (+26.15) 46.45 (+40:93) 48.20 (+33.20)
Settlements 0.60 (+195.74) 0.26 (+480.00) 0.43 (+280.60)
Land cover category Year
2014 1962
Agricultural land −0.00044308 −0.00007139851
Riverbed 0.00010825 0.00047055326
Settlement −0.00123494 0.00056181737
Riparian forest −0.00008747 −0.00004662649
  • RE was calculated based on Liao and Bai ( 2010 ). Values range between 0 and 1. Positive (negative) RE indicate that the grid area after rasterization is larger (smaller) than the vector area before rasterization for the land category. The larger the absolute value, the bigger is the error.

Data obtained from FRAGSTAT analyses indicated a severe change in the habitat configuration along the two rivers from 1961 to 2015/2016. At the landscape level, the overall number of patches (including all categories) increased from 487 to 985. At the class level, the number of patches (NP) for agricultural land increased from 185 to 196. The number of patches of riparian vegetation increased from 200 to 651. The Effective Mesh Size (MESH) decreased from 0.88 to 0.73 (Tables 3 and 4). At class level, the Mann–Whitney U test revealed significant differences of the landscape configuration between 1961 and 2015/2016 for agricultural land in terms of MESH, and for riparian vegetation in terms of NP and MESH. At land cover category (class) level, the analysis showed only significant differences between the two rivers for the agricultural land category in terms of NP in 2015/2016. Consequently, during the past five decades, total coverage of riparian vegetation strongly decreased, and the remaining patches have become strongly fragmented (Table 4).

Regional scale Landscape scale
Nzeeu River Nzeeu River Kalundu River Overall
Years / Land cover category NP MESH NP MESH NP MESH NP MESH
1961 138 0.7000 261 0.2315 226 0.8529 487 0.2224
1980 144 0.6460
2015/2016 218 0.6233 505 0.1786 479 0.7469 984 0.1827
Riparian vegetation 62 0.3881 200 0.4503 96 0.4599 200 0.4551
Settlements 7 0.0012 11 0.0036 3 0.0025 11 0.0032
Agricultural land 54 0.2128 185 0.4550 86 0.3688 185 0.4119
Riparian vegetation 54 0.2426
Settlements 13 0.0007
Agricultural land 63 0.3231
Riparian vegetation 88 0.2265 650 0.1015 322 0.0689 650 0.0852
Settlements 29 0.0003 67 0.0028 33 0.0018 67 0.0023
Agricultural land 87 0.3142 196 0.5644 85 0.6449 196 0.6046

Proportion of L. camara

The detailed land cover assessment of a 50 m strip along both sides of Nzeeu River and Kalundu River yielded the following proportions of land coverage: 43.8% agricultural land and 56.2% riparian vegetation. The latter land cover category consisted of 45.3% pristine riparian forests and 54.4% disturbed riparian thickets dominated by L. camara.

Arthropod abundance

Overall arthropod abundances across all three groups (herbivores, predators, ants) were lower in L. camara-dominated thickets than in pristine riparian forests (one-way PERMANOVA: P [F1,259] < 0.01). Combining data from both study years, pristine riparian forests yielded more herbivorous arthropods (P [F1,259] < 0.001) and ants (P [F1,259] < 0.01), but not predators (P [F1,259] > 0.2) (Fig. 3). Two-way PERMANOVA with study year and habitat as grouping variables identified significant year × habitat interactions (P [F1,258] < 0.05) in ant abundances. No interactions were observed for herbivores and predators (P [F1,258] > 0.1).

Availability and functionality

This methodology is implemented in a free and open-source software package, fuzzySim , which works under the r programming environment (R Core Team 2014 ). The package, including some sample data (Fontaneto et al. 2012 ), is available on the public platform R-Forge (, together with a reference manual and a step-by-step tutorial on its installation and usage. Most functionalities of fuzzySim (Fig. 2) are also being implemented as a graphical user interface extension for QGIS (QGIS Development Team 2014 ), which is also free and open-source.

The fuzzySim package allows a variety of methods for converting (multiple) species presence/absence data into continuous, fuzzy surfaces, including inverse distance to presence raised to any power (function distPres ), trend surface analysis of any given degree (function multTSA ) and generalized linear models based on presence–absence (function multGLM ). The former two methods can be useful for purposes other than comparing species distributions and assemblages, for example for defining putative geographical ranges (Takahashi et al. 2014 ) or for delimiting the geographical background in species distribution modelling (Acevedo et al. 2012 ). Besides several methods for selecting predictor variables (including information criteria and false discovery rate), multGLM includes an option to convert probability to prevalence-independent favourability values, which have proven appropriate for use within a fuzzy logic framework (Real, Barbosa & Vargas 2006 Acevedo & Real 2012 ). The package also allows using other continuous distribution data that users can obtain elsewhere, as long as they are bounded between 0 and 1, directly comparable among species and interpretable as fuzzy membership values (Zadeh 1965 Real, Barbosa & Vargas 2006 Barbosa & Real 2012 ).

The fuzzy similarity matrices produced with fuzzySim can then be plotted, compared, classified, clustered and converted into dendrograms depicting the fuzzy relationships between species distributions or between regional species compositions. r code for all these operations is provided in the tutorials available from the package homepage ( The fuzzy similarity matrices can also be entered in the RMACOQUI package (Olivero, Real & Márquez 2011 ) for a systematic analysis of chorotypes or biotic regions. The fuzzy versions of binary similarity indices can also be integrated within other software packages that currently compute these indices, such as vegan (Oksanen et al. 2013 ) or betapart (Baselga & Orme 2012 ).


Our results demonstrate that urbanization alters predator-avoidance behaviour in white-tailed deer and eastern cottontails. We found, across our study area, that neither white-tailed deer nor eastern cottontail were spatially segregated from coyotes, nor did we find a pronounced change in their daily activity patterns. While deer did not increase vigilance when coyotes were present, eastern cottontails had higher vigilance when coyotes were present in less urban sites, but lower vigilance when coyotes were present in more urban sites. Conversely, eastern cottontails had their highest vigilance rates in more urban sites when coyotes were absent, indicating an alternative source of fear that may be “forcing” cottontails to be vigilant even when coyotes are absent.

If all other factors are equal (e.g., food availability and social standing), prey species should display increased anti-predator behaviour where predation pressures exist (Caro, 2005 ). White-tailed deer in less urban ecosystems have been shown to avoid habitat occupied by coyotes (Jones et al., 2016 Lingle, 2002 ). Although coyotes depredate eastern cottontails and young white-tailed deer in the greater Chicago region (Gehrt, 2010a Morey et al., 2007 ), we found that neither white-tailed deer nor eastern cottontail demonstrated spatial segregation from coyotes across our study area (Figure 2a). These results are similar to previous findings in the Chicago metropolitan area (Magle et al., 2014 ). Coyote and deer occupancy rates were positively linked, and the authors suggested that the need for limited habitat may outweigh any spatial predator avoidance by deer (Magle et al., 2014 ). We too expect that habitat limitation constrained prey species’ ability to spatially avoid coyotes across the landscape.

Because white-tailed deer and cottontail did not spatially avoid coyotes, we expected to observe a change in prey species temporal activity patterns to minimize the frequency of interactions with coyotes. Yet, we found only minor changes in daily activity patterns of prey species in sites with or without coyote (Figure 2b). White-tailed deer slightly increased their activity to be active later in the morning and earlier in the evening when coyotes were present at a site (Figure 2b). However, peak activity of coyotes did not track high levels of deer activity. White-tailed deer are generally crepuscular or diurnal (Feldhamer, Thompson, & Chapman, 2003 ), and our results show that this behaviour does not change for urban deer (Figure 2b). Previous studies have also shown that urban coyotes become more nocturnal to avoid human activity (Gehrt, 2010a Grubbs & Krausman, 2009 Riley et al., 2003 Tigas, Van Vuren, & Sauvajot, 2002 ), and our results support this evidence (Figure 2b). Perhaps the risk from humans outweighs the benefits of overlapping activity patterns with deer and has decoupled the temporal interactions between deer and coyotes.

This potential decoupling may have important implications for urban predator–prey dynamics. Within human-dominated systems, especially cities, predators are often extirpated or removed from the system if they interact with humans (Curtis & Hadidian, 2010 ). Thus, urban apex predators, who previously faced little risk while hunting, now face new mortality risks when choosing when and where to forage given the presence of humans (Blecha, Boone, & Alldredge, 2018 ). For coyote, aligning their activity pattern with that of deer may be too risky at most times, especially given that other food sources are seemingly available (Newsome et al., 2015 ). Thus, our results may indicate the influential role of a “third player” (i.e., humans) in urban predator–prey dynamics.

Predation pressures may also be reduced due to a surplus of resources provided by anthropogenic sources (e.g., refuse, fruit trees, pet food). Coyotes are generalist predators (Morey et al., 2007 ), and urban coyotes have greater access to a variety of supplemental resources throughout the year (Morey et al., 2007 Murray & St. Clair, 2017 Newsome et al., 2015 Poessel, Mock, & Breck, 2017 ). A potential increase in prey items combined with the supplementation of anthropogenic food sources may reduce the overall predation rates on cottontails and deer by coyotes across the urban landscape, thus reducing the need for either species to increase their predator-avoidance behaviour.

We also expected prey species to express increased vigilance in the presence of coyote, and we found that eastern cottontail vigilance rates, in the presence of coyote, positively covaried with coyote occupancy. Vigilance rates were higher in less urban sites with coyote present where coyote occupancy is also high (Figure 2c). Conversely, eastern cottontail vigilance was highest in urban sites without coyotes. These varying results may suggest different sources that induce vigilance behaviour in eastern cottontails along an urbanization gradient. In Chicago, cottontail rabbits are likely to occupy green spaces in the urban core (e.g., city parks Gallo et al., 2017 ) where coyote occupancy is low (Figure 2c, Supporting Information Table S4) but visitation by humans and their pets is high. While these urban locations may provide potential refuge for eastern cottontails from coyote (i.e., human-shield effect), they potentially come with trade-offs in the form of increased interactions with humans and their pets. As a result, their vigilance rates are high in urban sites without coyote. As sites become less urban, we begin to see more expected anti-predator behaviour towards a native predator as cottontails have higher vigilance rates in less urban sites with coyote than those without (Figure 2c).

In regards to white-tailed deer, our results may indicate that white-tailed deer are at their limits of behavioural plasticity in urban ecosystems, and cannot afford to change their predator-avoidance behaviour (Lowry et al., 2013 ). Deer vigilance rates at sites with coyote did not vary as a function of urbanization. However, vigilance rates appeared higher in less urban sites without coyote, though this trend was not significant (Figure 2c). Alternatively, the separation in temporal interaction between coyote and deer due to changes in urban coyote daily activity (Figure 2b) may have reduced the need for deer to be vigilant of coyote.

It is important to note that detecting within-patch dynamics was not possible within our study design, as we had only one camera placed within each habitat patch. However, we found that detection rates were higher for both prey species when coyotes were present (Supporting Information Table S1). These findings could indicate that coyotes are influencing within-patch activity and movement patterns of both prey species (Bowers & Dooley, 1993 ). Future research should assess within-patch spatial avoidance or behavioural changes by including multiple sampling sites within a single habitat patch or collecting fine-scale movement data using GPS collars on individual animals.

Our results add to a growing body of literature that indicates interactions between predators and prey in human-dominated landscapes may be better understood by considering the interplay between three players instead of two: predators, prey and people (Berger, 2007 Blecha et al., 2018 Magle et al., 2014 ). Given our findings, we believe it would be of value to assess fine-scale behaviours at the individual level (e.g., forager's perception of risk, internal state of predator and prey). Conducting fine-scale diet and behavioural analysis, such as measuring giving up densities, (Brown & Kotler, 2004 Kotler, Brown, & Bouskila, 2004 ) would be a natural progression to further assess our postulates about behavioural plasticity, the use of anthropogenic resource by predators and novel threats within urban ecosystems. Additionally, we assessed patterns of predator-avoidance behaviour based on the presence of coyote, yet predator abundance may have a stronger influence on predator–prey dynamics (Power, Matthews, & Stewart, 1985 ). Future studies that measure population dynamics, such as mark-recapture, would help elucidate the influence predator abundance has on prey behaviour in novel ecosystems.

Our findings have greater implications for urban wildlife management. Conservation and management actions are often based on predictions about population dynamics and species interactions generated from ecological models derived in more natural or rural settings. Changes in the impact that urban predators have on the behaviour of urban prey can alter or reduce top-down trophic effects causing further changes in prey behaviour (Waser et al., 2014 ). Changes in predator-avoidance behaviour may have further cascading effects on both plant and animal communities, ultimately changing the community composition of urban ecosystems (Kuijper et al., 2016 Waser et al., 2014 ). For example, urban and suburban white-tailed deer often modify understory plant communities and alter forest ecology through uncontrolled herbivory (Côté, Rooney, Tremblay, Dussault, & Waller, 2004 DeNicola, VerCauteren, Curtis, & Hyngstrom, 2000 ). Thus, recognizing how urbanization alters the behaviour of both urban predators and urban prey is a key component to understanding urban wildlife communities and managing and conserving biodiversity on an urbanizing planet.


This is the first coronavirus detected from a bat in Singapore, and our study provides the first evidence of cave nectar bat E. spelaea harbouring a lineage D betacoronavirus. Coronaviruses were previously detected in E. spelaea in the Philippines, but no sequence data were generated (Watanabe et al., 2010 ). Eonycteris spelaea has also previously been reported to harbour a paramyxovirus in China (Yuan et al., 2014 ), Phnom-Penh bat virus (Queen et al., 2015 ), Issyk-kul virus (Calisher et al., 2006 ) and were also seropositive for Nipah virus in Malaysia (Yob et al., 2001 ).

In South-East Asia, lineage D betacoronaviruses (also referred to as Ro-BatCoV HKU9: Woo et al., 2012a , b ) have been detected in eight different bat species from three bat families belonging to two suborders, Yinpterochiroptera and Yangochiroptera: C. sphinx, D. moluccensis, E. spelaea, H. lekaguli, P. jagori, R. leschenaultii, Scotophilus heathi and Scotophilus kuhli (Woo et al., 2007 Tsuda et al., 2012 Anindita et al., 2015 ). A next-generation sequencing approach in Yunnan province China detected HKU9 coronavirus sequences in a community dominated by Hipposideros armiger, but the host identity of the faeces tested was not confirmed (Ge et al., 2012 ). The bat species known to harbour lineage D Betacoronavirus are widely distributed across South Asia and South-East Asia, and there are large areas where these distributions overlap (Fig 3, Table S2). Lineage D betacoronaviruses were also detected in two bat species in Kenya, Rousettus aegyptiacus and H. commersoni, and one in Madagascar, Pteropus rufus (Tong et al., 2009 Razanajatovo et al., 2015 ).

Interestingly, the pooled urine samples where coronavirus was detected in the Illumina NGS were negative with traditional RT-PCR, indicating a low amount of virus. These screening results possibly indicate that there is a low prevalence in the bats in Singapore, low virus titres in the samples and/or we missed shedding foci during our sampling. The only bat species we detected coronaviruses in were the cave nectar bats (E. spelaea), and this was using deep sequencing where the majority of the reads were detected in the urine sample. We did not detect this in the other insectivorous or frugivorous bat species with this well-validated primer set however, we did not perform NGS on samples from other species in Singapore. The majority of coronavirus detections have been from insectivorous bats, but this may have been spurred by the discovery of SARS-like CoV in Rhinolophus bat species in China (Drexler et al., 2014 ). Different coronaviruses can infect the same species of bats, even very divergent viruses (Wacharapluesadee et al., 2015 ). The close evolutionary relationship of lineage D betacoronaviruses detected from geographically separated bat species indicates that the spread of lineage D infections in bats is not host-restricted.

In Germany, bat parturition peaks appear correlated with bat coronavirus shedding (Drexler et al., 2011 ) it would be beneficial to determine the pregnancy cycles of bats in the tropics, which is largely unknown or based on studies with limited geographical scope. In the tropics, the constant availability of resources means there is less dependence on temporal windows of food. In Malaysia, E. spelaea has been found with dependent young in 8 of 12 months, indicating that individuals are reproducing throughout the year (Kingston et al., 2006 ), with two peak periods of pregnancy in June and September. In the Philippines, it was also found that E. spelaea has two annual peaks in reproduction although with some variation (Krutzsch, 2005 ). The first peak centred on March or April while the second centred on or near August (Heideman and Utzurrum, 2003 ). These bats typically roost in large numbers in caves, although the studied colony in Singapore roosts under an overpass and numbers over 3000 individuals (Lee, unpublished data). This opens up the possibility that there is a persistent source of virus in the colonies and immunologically naïve individuals have the opportunity for exposure from infected bats as they are tightly packed and may fight for specific sites in a roosting area. This is especially true for larger and older males because it is hypothesized that E. spelaea exhibits a resource defence polygynous mating system, where males spend significantly more time and energy for roost defence and surveillance (Bumungsri et al., 2013 ).

In addition, E. spelaea, R. leschenaultii and H. armiger co-roost in caves in India and in SE Asia and multispecies roosts may facilitate the transmission and recombination of coronaviruses (Mendenhall, unpublished data Struebig et al., 2005 Furey et al., 2011 ). Hipposideros lekaguli is thought to be associated with Rousettus species and even E. spelaea in SE Asia (Alviola et al., 2015 ). These species also have a large, overlapping distribution across SE Asia, which may facilitate the emergence of novel coronaviruses (Fig. 2). The foraging ecology of these bats E. spelaea is known to feed on a wide range of plant species and their high visitation frequency to the inflorescence of a few keystone plant species such as Durio zibethinus and Parkia speciosa (Heideman and Utzurrum, 2003 Krutzsch, 2005 ). Both C. sphinx and R. leschnaulti have also been found to feed on the same tree (Singaravelan and Marimuthu, 2004 , 2008 ).

The discovery of this novel coronavirus demonstrates the unsampled diversity of this virus family in bats. Betacoronaviruses are responsible for two of the major coronavirus cross-species spillover events (SARS and MERS virus) from bats to incidental hosts, ultimately resulting in sustained transmission chains. The bat species where lineage D betacoronaviruses have been detected share roosting sites, do so in large numbers and high density and also forage on the same plants. These may be factors that ultimately lead to the emergence of novel virus strains (Table 2). Even though the lineage D betacoronaviruses are not implicated as viruses capable of infecting humans, understanding their evolution can help us understand the evolution and ecological aspects of this medically important virus family. Furthermore, as little is known about the evolutionary factors affecting the transmission of CoVs between hosts, additional studies, such as host cell receptor usage, are required to assess the potential risks of zoonotic transmission to humans.

Bat species Roost site (cave/tree) Food (no. of plant species) Numbers in single roost References
Eonycteris spelaea Caves in forested areas and man-made structure in urban areas Flowers/nectar Several thousands. Co-habits with other bat species (Francis et al., 2008 )
Cynopterus sphinx Foliage (tree and palms) only Fruits (wild and cultivated) 3–7 bats per roost (Bates et al., 2008b )
Dobsonia moluccensis Caves, sinkholes, boulder piles, old mines, disused buildings, dense vegetation Fruits Several thousands (Hutson et al., 2008 )
Hipposideros lekaguli Mainly limestone caves Insects 300 individuals (per colony) (Bumrungsri per comm.) (Csorba et al., 2008 )
Rousettus leschenaultii Caves, old buildings, forts and disused tunnels Fruits and flowers/nectar A few to several thousand individuals (Bates and Helgen, 2008 )
Hipposideros armiger Caves and man-made structures Insects Hundreds of individuals. Co-habits with Rhinolophus and other bat species (Bates et al., 2008a )
Ptenochirus jagori Mainly tree cavities, also caves and sheltered rock crevices Fruits (145), flowers/nectar (2) and leaves (7) Singly or in small groups in caves (Mickleburgh, 1992 Reiter and Curio, 2001 Ong et al., 2008 )
Scotophilus kuhli Tree cavities, foliage and man-made structures (in forest and human-modified environments) Insects (hymenopterans and dipterans) Several hundred individuals (Bates et al., 2008d )
Scotophilus heathii Tree cavities, foliage and man-made structures (in forest and human-modified environments) Insects Singly or in colonies of up to 50 bats (Bates et al., 2008c )

Data available from the Dryad Digital Repository: (Duffy, 2018 ).

Filename Description
jec13032-sup-0001-FigS1.pdfPDF document, 33.6 KB
jec13032-sup-0002-FigS2.pdfPDF document, 22.8 KB
jec13032-sup-0003-FigS3.pdfPDF document, 22.8 KB
jec13032-sup-0004-FigS4.pdfPDF document, 26.6 KB
jec13032-sup-0005-TableS1.docxWord document, 12.7 KB
jec13032-sup-0006-FigureCaptions.docxWord document, 12.5 KB

Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Electronic supplementary material is available online at

Published by the Royal Society. All rights reserved.


Carver S, Kilpatrick AM, Kuenzi A, Douglass R, Ostfeld RS, Weinstein P

. 2010 Environmental monitoring to enhance comprehension and control of infectious diseases . J. Environ. Monit. 12, 2048-2055. (doi:10.1039/c0em00046a) Crossref, PubMed, Google Scholar

. 2012 Effects of host diversity on infectious disease . Annu. Rev. Ecol. Evol. Syst. 43, 157182. (doi:10.1146/annurev-ecolsys-102710-145022) Crossref, ISI, Google Scholar

Keesing F, Holt RD, Ostfeld RS

. 2006 Effects of species diversity on disease risk . Ecol. Lett. 9, 485-498. (doi:10.1111/j.1461-0248.2006.00885.x) Crossref, PubMed, ISI, Google Scholar

Johnson PT, Preston DL, Hoverman JT, Richgels KL

. 2013 Biodiversity decreases disease through predictable changes in host community competence . Nature 494, 230233. (doi:10.1038/nature11883) Crossref, PubMed, ISI, Google Scholar

Ezenwa VO, Godsey MS, King RJ, Guptill SC

. 2006 Avian diversity and West Nile virus: testing associations between biodiversity and infectious disease risk . Proc. R. Soc. B 273, 109-117. (doi:10.1098/rspb.2005.3284) Link, ISI, Google Scholar

Huang ZY, van Langevelde F, Prins HH, de Boer WF.

2015 Dilution versus facilitation: impact of connectivity on disease risk in metapopulations . J. Theor. Biol. 376, 6673. (doi:10.1016/j.jtbi.2015.04.005) Crossref, PubMed, ISI, Google Scholar

Ferguson H, Brock P, Torr S

. 2018 Host species diversity and the transmission of vector-borne disease in low income countries . In Ecology and evolution of infectious diseases: effective strategies for pathogen control and public health management in low-income countries (eds

Roche B, Broutin H, Simard F

), pp. 180-197. Oxford, UK : Oxford University Press . Google Scholar

LoGiudice K, Duerr STK, Newhouse MJ, Schmidt KA, Killilea ME, Ostfeld RS

. 2008 Impact of host community composition of Lyme disease risk . Ecology 89, 2841-2849. (doi:10.1890/07-1047.1) Crossref, PubMed, ISI, Google Scholar

Gottdenker NL, Chaves LF, Calzada JE, Saldaña A, Carroll CR

. 2012 Host life history strategy, species diversity, and habitat influence Trypanosoma cruzi vector infection in changing landscapes . PLoS Negl. Trop. Dis. 6, e1884. (doi:10.1371/journal.pntd.0001884) Crossref, PubMed, ISI, Google Scholar

Oda E, Solari A, Botto-Mahan C

. 2014 Effect of mammal host diversity and density on the infection level of Trypanosoma cruzi in sylvatic kissing bugs . Med. Vet. Entomol. 28, 384-390. (doi:10.1111/mve.12064) Crossref, PubMed, ISI, Google Scholar

. 2005 Species coexistence and pathogens with frequency-dependent transmission . Am. Nat. 166, 112-118. (doi:10.1086/430674) Crossref, PubMed, ISI, Google Scholar

. 2016 Epidemiological implications of host biodiversity and vector biology: key insights from simple models . Am. Nat. 187, 405-422. (doi:10.1086/685445) Crossref, PubMed, ISI, Google Scholar

2015 Bottlenecks in domestic animal populations can facilitate the emergence of Trypanosoma cruzi, the aethiological agent of Chagas disease . Proc. R. Soc. B 282, 20142807. (doi:10.1098/rspb.2014.2807) Link, ISI, Google Scholar

. 1986 The structure of Neotropical mammal communities: an appraisal of current knowledge . Rev. Chil. Hist. Nat. 59, 121134. ISI, Google Scholar

Bacigalupo A, Segura JA, García A, Hidalgo J, Galuppo S, Cattan PE

. 2006 First finding of Chagas disease vectors associated with wild bushes in Metropolitan Region of Chile . Rev. Med. Chil. 134, 1230-1236. (doi:10.4067/S0034-98872006001000003) PubMed, ISI, Google Scholar

Botto-Mahan C, Cattan PE, Canals M, Acuña M

. 2005 Seasonal variation in the home range and host availability of the blood-sucking insect Mepraia spinolai in wild environment . Acta Trop. 95, 160-163. (doi:10.1016/j.actatropica.2005.05.001) Crossref, PubMed, ISI, Google Scholar

Coronado X, Rozas M, Botto-Mahan C, Ortiz S, Cattan PE, Solari A

. 2009 Molecular epidemiology of Chagas disease in the wild transmission cycle: the evaluation in the sylvatic vector Mepraia spinolai from an endemic area of Chile . Am. J. Trop. Med. Hyg. 81, 656-659. (doi:10.4269/ajtmh.2009.09-0053) Crossref, PubMed, ISI, Google Scholar

Correa JP, Bacigalupo A, Fontúrbel FE, Oda E, Cattan PE, Solari A, Botto-Mahan C

. 2015 Spatial distribution of an infectious disease in a native small mammal community . Naturwissenschaften 102, 51. (doi:10.1007/s00114-015-1304-5) Crossref, PubMed, ISI, Google Scholar

2019 Spatio-temporal characterization of Trypanosoma cruzi infection and discrete typing units infecting hosts and vectors from non-domestic foci of Central Chile . PLoS Negl. Trop. Dis. 13, e7170. (doi:10.1371/journal.pntd.0007170) Crossref, ISI, Google Scholar

. 2008 Mamíferos de Chile . Barcelona, Spain : Editorial LYNX . Google Scholar

Canals M, Cruzat L, Molina MZ, Ferreira A, Cattan PE

. 2001 Blood host sources of Mepraia spinolai (Heteroptera: Reduviidae), wild vector of Chagas Disease in Chile . J. Med. Entomol. 38, 303-307. (doi:10.1603/0022-2585-38.2.303) Crossref, PubMed, ISI, Google Scholar

Chacon F, Bacigalupo A, Quiroga JF, Ferreira A, Cattan PE, Ramirez-Toloza G

. 2016 Feeding profile of Mepraia spinolai, a sylvatic vector of Chagas disease in Chile . Acta Trop. 162, 171-173. (doi:10.1016/j.actatropica.2016.06.027) Crossref, PubMed, ISI, Google Scholar

Rozas M, Botto-Mahan C, Coronado X, Ortiz S, Cattan PE, Solari A

. 2007 Co-existence of Trypanosoma cruzi genotypes in wild and peridomestic mammals in Chile . Am. J. Trop. Med. Hyg. 77, 647-653. (doi:10.4269/ajtmh.2007.77.647) Crossref, PubMed, ISI, Google Scholar

Botto-Mahan C, Acuña-Retamar M, Campos R, Cattan P, Solari A

. 2009 European rabbits (Oryctolagus cuniculus) are naturally infected with different Trypanosoma cruzi genotypes . Am. J. Trop. Med. Hyg. 80, 944-946. (doi:10.4269/ajtmh.2009.80.944) Crossref, PubMed, ISI, Google Scholar

Galuppo S, Bacigalupo A, Garcia A, Ortiz S, Coronado X, Cattan PE, Solari A

. 2009 Predominance of Trypanosoma cruzi genotypes in two reservoirs infected by sylvatic Triatoma infestans of an endemic area of Chile . Acta Trop. 111, 90-93. (doi:10.1016/j.actatropica.2009.02.010) Crossref, PubMed, ISI, Google Scholar

Rojo G, Sandoval-Rodríguez A, López A, Ortiz S, Saavedra M, Botto-Mahan C, Cattan PE, Solari A

. 2017 Within-host temporal fluctuations of Trypanosoma cruzi discrete typing units: the case of the wild reservoir rodent Octodon degus . Parasit. Vectors 10, 380. (doi:10.1186/s13071-017-2314-2) Crossref, PubMed, ISI, Google Scholar

Yefi-Quinteros E, Muñoz-San Martin C, Bacigalupo A, Correa JP, Cattan PE

. 2018 Trypanosoma cruzi load in synanthropic rodents from rural areas in Chile . Parasit. Vectors 11, 171. (doi:10.1186/s13071-018-2771-2) Crossref, PubMed, ISI, Google Scholar

Botto-Mahan C, Ortiz S, Rozas M, Cattan PE, Solari A

. 2005 DNA evidence of Trypanosoma cruzi in the Chilean wild vector Mepraia spinolai (Hemiptera: Reduviidae) . Mem. Inst. Oswaldo Cruz 100, 237-239. (doi:10.1590/S0074-02762005000300003) Crossref, PubMed, ISI, Google Scholar

Botto-Mahan C, Campos R, Acuña-Retamar M, Coronado X, Cattan PE, Solari A

. 2010 Temporal variation of Trypanosoma cruzi infection in native mammals in Chile . Vector-Borne Zoonot. Dis. 10, 317-319. (doi:10.1089/vbz.2009.0006) Crossref, PubMed, ISI, Google Scholar

Botto-Mahan C, Bacigalupo A, Correa JP, Oda E, Solari A

. 2012 Field assessment of Trypanosoma cruzi infection and host survival in the native rodent Octodon degus . Acta Trop. 122, 164-167. (doi:10.1016/j.actatropica.2011.12.003) Crossref, PubMed, ISI, Google Scholar

Lima M, Marquet PA, Jaksic FM

. 1996 Extinction and colonization processes in subpopulations of five neotropical small mammal species . Oecologia 107, 197-203. (doi:10.1007/bf00327903) Crossref, PubMed, ISI, Google Scholar

Botto-Mahan C, Cattan PE, Canals M

. 2002 Field tests of carbon dioxide and conspecifics as baits for Mepraia spinolai, vector of Chagas disease . Acta Trop. 82, 377-380. (doi:10.1016/S0001-706X(02)00040-2) Crossref, PubMed, ISI, Google Scholar

Meserve PL, Gutiérrez JR, Yunger JA, Contreras LC, Jaksic FM

. 1996 Role of biotic interactions in a small mammal assemblage in semiarid Chile . Ecology 77, 133-148. (doi:10.2307/2265662) Crossref, ISI, Google Scholar

Cattan PE, Botto-Mahan C, Correa JP, Bacigalupo A, Cornejo-Villar B

. 2019 Trapping and blood-sampling small mammals in semi-arid environments. ( Google Scholar

. 2007 The animal care and use committee of the American Society of Mammalogists. Guidelines of the American Society of Mammalogists for the use of Wild Mammals in Research . J. Mammal. 88, 809823. (doi: 10.1644/06-MAMM-F-185R1.1) Crossref, ISI, Google Scholar

Arenas M, Campos R, Coronado X, Ortiz S, Solari A

. 2012 Trypanosoma cruzi genotypes in insect vectors and patients with Chagas of Chile studied by means of cytochrome b gene sequencing, minicircle hybridization, and nuclear gene polymorphism . Vector-Borne Zoonot. Dis. 12, 196-205. (doi:10.1089/vbz.2011.0683) Crossref, PubMed, ISI, Google Scholar

. 2013 Retrospective molecular integrated epidemiology of Chagas disease in Colombia . Infect. Genet. Evol. 20, 148-154. (doi:10.1016/j.meegid.2013.08.028) Crossref, PubMed, ISI, Google Scholar

Wincker P, Britto C, Pereira JB, Cardoso MA, Oeleman O, Morel CM

. 1994 Use of a simplified polymerase chain reaction procedure to detect Trypanosoma cruzi in blood samples from chronic chagasic patients in a rural endemic area . Am. J. Trop. Med. Hyg. 51, 771-777. (doi:10.4269/ajtmh.1994.51.771) Crossref, PubMed, ISI, Google Scholar

. 2006 The Martyrdom of St. Lucie: decimation of a meiofauna . Bull. Mar. Sci. 79, 341352. ISI, Google Scholar

Zuur A, Ieno EN, Walker N, Saveliev AA, Smith GM

. 2009 Mixed effects models and extensions in ecology with R . New York, NY : Springer . Crossref, Google Scholar

Bates D, Maechler M, Bolker B, Walker S

. 2015 Fitting linear mixed-effects models using lme4 . J. Stat. Softw. 67, 1-48. (doi:10.18637/jss.v067.i01) Crossref, ISI, Google Scholar

. 2016 Smoothing parameter and model selection for general smooth models (with discussion) . J. Am. Stat. Assoc. 111, 1548-1575. (doi:10.1080/01621459.2016.1180986) Crossref, ISI, Google Scholar

. 2017 gamm4: Generalized Additive Mixed Models using ‘mgcv’ and ‘lme4’. R package version 0.2–5. See Google Scholar

Lima M, Stenseth NC, Leirs H, Jaksic FM

. 2003 . Population dynamics of small mammals in semi-arid regions: a comparative study of demographic variability in two rodent species . Proc. R. Soc. Lond B 270, 1997-2007. (doi:10.1098/rspb.2003.2448) Link, ISI, Google Scholar

Jiménez C, Fontúrbel FE, Oda E, Ramírez PA, Botto-Mahan C

. 2015 Parasitic infection alters rodent movement in semiarid ecosystem . Mamm. Biol. 80, 255-259. (doi:10.1016/j.mambio.2015.01.006). Crossref, ISI, Google Scholar

Ramirez-Sierra MJ, Dumonteil E

. 2016 Infection rate by Trypanosoma cruzi and biased vertebrate host selection in the Triatoma dimidiata (Hemiptera: Reduvidae) species complex . J. Med. Entomol. 53, 20-25. (doi:10.1093/jme/tjv157) Crossref, PubMed, ISI, Google Scholar

Gürtler RE, Ceballos LA, Ordóñez-Krasnowski P, Lanati LA, Stariolo R, Kitron U.

2009 Strong host-feeding preferences of the vector Triatoma infestans modified by vector density: implications for the epidemiology of Chagas disease . PLoS Negl. Trop. Dis. 3, e447. (doi:10.1371/journal.pntd.0000447) Crossref, PubMed, ISI, Google Scholar

Orozco MM, Enriquez GF, Alvarado-Otegui JA, Cardinal MV, Schijman AG, Kitron U, Gürtler RE

. 2013 New sylvatic hosts of Trypanosoma cruzi and their reservoir competence in the humid Chaco of Argentina: a longitudinal study . Am. J. Trop. Med. Hyg. 88, 872-882. (doi:10.4269/ajtmh.12-0519) Crossref, PubMed, ISI, Google Scholar

Orozco MM, Piccinali RV, Mora MS, Enriquez GF, Cardinal MV, Gürtler RE

. 2014 The role of sigmodontine rodents as sylvatic hosts of Trypanosoma cruzi in the Argentinean Chaco . Inf. Genet. Evol. 22, 12-22. (doi:10.1016/j.meegid.2013.12.020) Crossref, PubMed, ISI, Google Scholar

Brémond P, Salas R, Waleckx E, Buitrago R, Aliaga C, Barnabé C, Depickère S, Dangles O, Brenière SF

. 2014 Variations in time and space of an Andean wild population of T. infestans at a microgeographic scale . Parasit. Vectors 7, 164. (doi:10.1186/1756-3305-7-164) Crossref, PubMed, ISI, Google Scholar

. 1989 Chagas in a community of small sympatric mammals from Las Chinchillas National Reserve, IV Region, Chile . Parasitol. Dia 13, 15-20. Google Scholar

Gurgel-Gonçalves R, Cura C, Schijman AG, Cuba CA

. 2011 Infestation of Mauritia flexuosa palms by triatomines (Hemiptera: Reduviidae), vectors of Trypanosoma cruzi and Trypanosoma rangeli in the Brazilian savanna . Acta Trop. 121, 105-111. (doi:10.1016/j.actatropica.2011.10.010) Crossref, PubMed, ISI, Google Scholar

Abrahan LB, Gorla DE, Catalá SS

. 2011 Dispersal of Triatoma infestans and other Triatominae species in the arid Chaco of Argentina - Flying, walking or passive carriage? The importance of walking females . Mem. Inst. Oswaldo Cruz 106, 232-239. (doi:10.1590/S0074-02762011000200019) Crossref, PubMed, ISI, Google Scholar

Watch the video: TOPOLOGICAL EDITING in QGIS evitiamo sovrapposizioni, buchi ed errori topologici