Local spatial autocorrelation in biological variables

Spatial statistics is a comprehensive tool for quantifying cell neighbor relationships and biological processes via tissue image analysis

Automated microscopy and computational image analysis has transformed cell biology, providing quantitative, spatially resolved information on cells and their constituent molecules from the sub-micron to the whole-organ scale. Here we explore the application of spatial statistics to the cellular relationships within tissue microscopy data and discuss how spatial statistics offers cytometry a powerful yet underused mathematical tool set for which the required data are readily captured using standard protocols and microscopy equipment. We also highlight the often-overlooked need to carefully consider the structural heterogeneity of tissues in terms of the applicability of different statistical measures and their accuracy and demonstrate how spatial analyses offer a great deal more than just basic quantification of biological variance. Ultimately, we highlight how statistical modeling can help reveal the hierarchical spatial processes that connect the properties of individual cells to the establishment of biological function.

Spatiotemporal tradeoffs and synergies in vegetation vitality and poverty transition in rocky desertification area

Vegetation recovery and poverty alleviation are critical problems in the karst national designed poor counties (NPDC) in southwest China. However, little information is available about the relationship between poverty and vegetation dynamics in these areas. In this study, we used remote sensing and statistical datasets from 2000 to 2015 to identify the relations between vegetation dynamics and poverty among the NPDC in southwest rocky desertification areas. We estimated the vegetation dynamics using the Normalized Difference Vegetation Index and poverty with the rural per capita net income. Local indicator of spatial association and the space-time transition type of poverty were applied to identify spatial patterns of the poverty spatial distribution relationship and transition. Also, poverty, natural and ecological governance factors were assessed using the Geo-detector method to uncover the driving factors of karst vegetation. The results showed that vegetation increased significantly (p < 0.05) in karst NPDC (82.82%) and rocky desertification control counties (78.77%). The karst NPDC was significantly clustered. The hot spots of rural per capita net income changed from west and north (2000) to only north (2015) and cold spots changed from east and south (2000) to only south (2015). The rural per capita net income spatiotemporal transition was higher in 2000 than in 2015. We found a weak synergy between vegetation change and poverty type transition in 42.86% of the browning counties, 45.45% in the slowly greening counties, and 43.65% in stable greening counties. However, 57.50% of counties in the quick greening counties showed a tradeoff relationship with the poverty type transition. The rocky desertification rate and ecological engineering measures affected vegetation dynamics importantly. The results will help decision-makers to understand the interdependence between vegetation and poverty. This will contribute to better policies formulation to tackle poverty in the karst rocky desertification area.

Land Cover Classification in Mangrove Ecosystems Based on VHR Satellite Data and Machine Learning—An Upscaling Approach

Mangrove forests grow in the inter-tidal areas along coastlines, rivers, and tidal lands. They are highly productive ecosystems and provide numerous ecological and economic goods and services for humans. In order to develop programs for applying guided conservation and enhancing ecosystem management, accurate and regularly updated maps on their distribution, extent, and species composition are needed. Recent advances in remote sensing techniques have made it possible to gather the required information about mangrove ecosystems. Since costs are a limiting factor in generating land cover maps, the latest remote sensing techniques are advantageous. In this study, we investigated the potential of combining Sentinel-2 and Worldview-2 data to classify eight land cover classes in a mangrove ecosystem in Iran with an area of 768 km2. The upscaling approach comprises (i) extraction of reflectance values from Worldview-2 images, (ii) segmentation based on spectral and spatial features, and (iii) wall-to-wall prediction of the land cover based on Sentinel-2 images. We used an upscaling approach to minimize the costs of commercial satellite images for collecting reference data and to focus on freely available satellite data for mapping land cover classes of mangrove ecosystems. The approach resulted in a 65.5% overall accuracy and a kappa coefficient of 0.63, and it produced the highest accuracies for deep water and closed mangrove canopy cover. Mapping accuracies improved with this approach, resulting in medium overall accuracy even though the user’s accuracy of some classes, such as tidal zone and shallow water, was low. Conservation and sustainable management in these ecosystems can be improved in the future.

Population genetic structures at multiple spatial scales: importance of social groups in European badgers

Population viability and metapopulation dynamics are strongly affected by gene flow. Identifying ecological correlates of genetic structure and gene flow in wild populations is therefore a major issue both in evolutionary ecology and species management. Studying the genetic structure of populations also enables identification of the spatial scale at which most gene flow occurs, hence the scale of the functional connectivity, which is of paramount importance for species ecology. In this study, we examined the genetic structure of a social, continuously distributed mammal, the European badger (Meles meles), both at large spatial scales (among populations) and fine (within populations) spatial scales. The study was carried out in 11 sites across France utilizing a noninvasive hair trapping protocol at 206 monitored setts. We identified 264 badgers genotyped at 24 microsatellite DNA loci. At the large scale, we observed high and significant genetic differentiation among populations (global Fst = 0.139; range of pairwise Fst [0.046–0.231]) that was not related to the geographic distance among sites, suggesting few large-scale dispersal events. Within populations, we detected a threshold value below which badgers were genetically close (&lt; 400 m), highlighting that sociality is the major structuring process within badger populations at the fine scale.

Spatiotemporal Variation and Hotspot Detection of the Avian Influenza A(H7N9) Virus in China, 2013⁻2017

This study aims to describe the spatial and temporal characteristics of human infections with H7N9 virus in China using data from 19 February 2013 to 30 September 2017 extracted from Centre for Health Protection of the Department of Health (CHP) and electronic databases managed by China's Center for Disease Control (CDC) and provincial CDCs synthetically using the Geographic Information System (GIS) software ArcMap™ 10.2 and SaTScan. Based on the multiple analyses of the A(H7N9) epidemics, there was a strong seasonal pattern in A(H7N9) virus infection, with high activity in the first quarter of the year, especially in January, February, and April, and a gradual dying out in the third quarter. Spatial distribution analysis indicated that Eastern China contained the most severely affected areas, such as Zhejiang Province, and the distribution shifted from coastline areas to more inland areas over time. In addition, the cases exhibited local spatial aggregation, with high-risk areas most found in the southeast coastal regions of China. Shanghai, Jiangsu, Zhejiang, and Guangdong were the high-risk epidemic areas, which should arouse the attention of local governments. A strong cluster from 9 April 2017 to 24 June 2017 was also identified in Northern China, and there were many secondary clusters in Eastern and Southern China, especially in Zhejiang, Fujian, Jiangsu, and Guangdong Provinces. Our results suggested that the spatial-temporal clustering of H7N9 in China is fundamentally different, and is expected to contribute to accumulating knowledge on the changing temporal patterns and spatial dissemination during the fifth epidemic and provide data to enable adequate preparation against the next epidemic.

Spatial genetic structure in wild cardoon, the ancestor of cultivated globe artichoke: Limited gene flow, fragmentation and population history

Nuclear and chloroplast markers and phenotypic characters were integrated to analyse the population genetic structure of wild cardoon, Cynara cardunculus var. sylvestris, the ancestor of cultivated globe artichoke, Cynara cardunculus var. scolymus on the island of Sardinia, Italy. The spatial scale ranged from a few metres to ∼200km. Wild cardoon appears to be genetically fragmented, with significant genetic divergence at various scales, indicating that gene flow is insufficient to counterbalance the effects of genetic drift or founder effects. Divergence between populations was higher for chloroplast (40%) than for nuclear markers (15%), suggesting that gene flow via seed was lower than via pollen. Two main genetic groups were detected; these correlated with differences in flowering time, capitula size, glossiness, and anthocyanin pigmentation. A complex population structure of wild cardoon emerged over small spatial scales, likely resulting from the interplay between gene dispersal, colonisation history and selective forces. Indeed, Sardinia appears to be a 'hybrid zone' of different gene pools. The island has unique diverse germplasm that has originated from hybridisation among different gene pools. The sampling of seeds from a few plants but from many sites is suggested as the best strategy to harvest the genetic diversity of wild cardoon.

Patterns of Species Richness and Turnover for the South American Rodent Fauna

Understanding the spatial distribution of species sheds light on the group’s biogeographical history, offers clues to the drivers of diversity, and helps to guide conservation strategies. Here, we compile geographic range information for South America’s diverse rodents, whose 14 families comprise ~50% of the continent’s mammalian species. The South American rodent fauna is dominated by independent and temporally staggered radiations of caviomorph and sigmodontine groups. We mapped species richness and turnover of all rodents and the principal clades to identify the main predictors of diversity patterns. Species richness was highest in the Andes, with a secondary hotspot in Atlantic Forest and some regions of considerable richness in Amazonia. Differences in richness were evident between the caviomorphs and sigmodontines, the former showing the greatest richness in tropical forests whereas the latter show—and largely determine—the all-rodent pattern. Elevation was the main predictor of sigmodontine richness, whereas temperature was the principal variable correlated with richness of caviomorphs. Across clades, species turnover was highest along the Andes and was best explained by elevational relief. In South America, the effects of the familiar latitudinal gradient in species richness are mixed with a strong longitudinal effect, triggered by the importance of elevation and the position of the Andes. Both latitudinal and elevational effects help explain the complicated distribution of rodent diversity across the continent. The continent’s restricted-range species—those seemingly most vulnerable to localized disturbance—are mostly distributed along the Andes and in Atlantic Forest, with the greatest concentration in Ecuador. Both the Andes and Atlantic Forest are known hotspots for other faunal and floral components. Contrasting patterns of the older caviomorph and younger sigmodontine radiations underscore the interplay of both historical and ecological factors in determining present-day diversity patterns.

Niche suitability affects development: skull asymmetry increases in less suitable areas

For conservation purposes, it is important to take into account the suitability of a species to particular habitats; this information may predict the long-term survival of a species. In this sense, morphological measures of developmental stress, such as fluctuating asymmetry, can be proxies for an individual’s performance in different regions. In this study, we conducted tests to determine whether areas with different levels of suitability for a species (generated by ecological niche models) were congruent with morphological markers that reflect environmental stress and morphological variance. We generated a Maxent niche model and compared the suitability assessments of several areas with the skull morphology data (fluctuating asymmetry and morphological disparity) of populations of the Atlantic forest endemic to Brazil rodent Akodon cursor. Our analyses showed a significant negative relationship between suitability levels and fluctuating asymmetry levels, which indicates that in less suitable areas, the individuals experience numerous disturbances during skull ontogeny. We have not found an association between morphological variance and environmental suitability. As expected, these results suggest that in environments with a lower suitability, developmental stress is increased. Such information is helpful in the understanding of the species evolution and in the selection of priority areas for the conservation of species.

Dispersal, mating events and fine-scale genetic structure in the lesser flat-headed bats

Population genetic structure has important consequences in evolutionary processes and conservation genetics in animals. Fine-scale population genetic structure depends on the pattern of landscape, the permanent movement of individuals, and the dispersal of their genes during temporary mating events. The lesser flat-headed bat (Tylonycteris pachypus) is a nonmigratory Asian bat species that roosts in small groups within the internodes of bamboo stems and the habitats are fragmented. Our previous parentage analyses revealed considerable extra-group mating in this species. To assess the spatial limits and sex-biased nature of gene flow in the same population, we used 20 microsatellite loci and mtDNA sequencing of the ND2 gene to quantify genetic structure among 54 groups of adult flat-headed bats, at nine localities in South China. AMOVA and F(ST) estimates revealed significant genetic differentiation among localities. Alternatively, the pairwise F(ST) values among roosting groups appeared to be related to the incidence of associated extra-group breeding, suggesting the impact of mating events on fine-scale genetic structure. Global spatial autocorrelation analyses showed positive genetic correlation for up to 3 km, indicating the role of fragmented habitat and the specialized social organization as a barrier in the movement of individuals among bamboo forests. The male-biased dispersal pattern resulted in weaker spatial genetic structure between localities among males than among females, and fine-scale analyses supported that relatedness levels within internodes were higher among females than among males. Finally, only females were more related to their same sex roost mates than to individuals from neighbouring roosts, suggestive of natal philopatry in females.

Spatial autocorrelation analysis of Chinese inter-provincial industrial chemical oxygen demand discharge

A spatial autocorrelation analysis method is adopted to process the spatial dynamic change of industrial Chemical Oxygen Demand (COD) discharge in China over the past 15 years. Studies show that amount and intensity of industrial COD discharges are on a decrease, and the tendency is more remarkable for discharge intensity. There are large differences between inter-provincial discharge amount and intensity, and with different spatial differentiation features. Global spatial autocorrelation analysis reveals that Global Moran’s I of discharge amount and intensity is on the decrease. In space, there is an evolution from an agglomeration pattern to a discretization pattern. Local spatial autocorrelation analysis shows that the agglomeration area of industrial COD discharge amount and intensity varies greatly in space with time. Stringent environmental regulations and increased funding for environmental protections are the crucial factors to cut down industrial COD discharge amount and intensity.

Do common eiders nest in kin groups? Microgeographic genetic structure in a philopatric sea duck

We investigated local genetic associations among female Pacific common eiders (Somateria mollissima v-nigrum) nesting in a stochastic Arctic environment within two groups of barrier islands (Simpson Lagoon and Mikkelsen Bay) in the Beaufort Sea, Alaska. Nonrandom genetic associations were observed among nesting females using regional spatial autocorrelation analyses for distance classes up to 1000 m in Simpson Lagoon. Nearest-neighbour analyses identified clusters of genetically related females with positive lr values observed for 0-13% and 0-7% of the comparisons in Simpson Lagoon and Mikkelsen Bay, respectively, across years. These results indicate that a proportion of females are nesting in close proximity to more genetically related individuals, albeit at low frequency. Such kin groupings may form through active association between relatives or through natal philopatry and breeding site fidelity. Eiders nest in close association with driftwood, which is redistributed annually by seasonal storms. Yet, genetic associations were still observed. Microgeographic structure may thus be more attributable to kin association than natal philopatry and site fidelity. However, habitat availability may also influence the level of structure observed. Regional structure was present only within Simpson Lagoon and this island group includes at least three islands with sufficient driftwood for colonies, whereas only one island at Mikkelsen Bay has these features. A long-term demographic study is needed to understand more fully the mechanisms that lead to fine-scale genetic structure observed in common eiders breeding in the Beaufort Sea.

A review of techniques for spatial modeling in geographical, conservation and landscape genetics

Most evolutionary processes occur in a spatial context and several spatial analysis techniques have been employed in an exploratory context. However, the existence of autocorrelation can also perturb significance tests when data is analyzed using standard correlation and regression techniques on modeling genetic data as a function of explanatory variables. In this case, more complex models incorporating the effects of autocorrelation must be used. Here we review those models and compared their relative performances in a simple simulation, in which spatial patterns in allele frequencies were generated by a balance between random variation within populations and spatially-structured gene flow. Notwithstanding the somewhat idiosyncratic behavior of the techniques evaluated, it is clear that spatial autocorrelation affects Type I errors and that standard linear regression does not provide minimum variance estimators. Due to its flexibility, we stress that principal coordinate of neighbor matrices (PCNM) and related eigenvector mapping techniques seem to be the best approaches to spatial regression. In general, we hope that our review of commonly used spatial regression techniques in biology and ecology may aid population geneticists towards providing better explanations for population structures dealing with more complex regression problems throughout geographic space.

Characterizing the spatial structure of songbird cultures

Recent advances have shown that human-driven habitat transformations can affect the cultural attributes of animal populations in addition to their genetic integrity and dynamics. Here I propose using the song of oscine birds for identifying the cultural spatial structure of bird populations and highlighting critical thresholds associated with habitat fragmentation. I studied song variation over a wide geographical scale in a small and endangered passerine, the Dupont's Lark Chersophilus duponti, focusing on (1) cultural population structure, to determine a statistical representation of spatial variation in song and identify cultural units, and (2) the minimum patch size needed for an individual to develop a stable repertoire. I found that overall song diversity depends on variation among populations (beta-cultural diversity). Abrupt thresholds occurred in the relationships between individual song dissimilarity and geographic distance, and between individual song diversity and patch area. Spatial autocorrelation analysis showed that populations located as little as 5 km apart may have independently evolved their song traditions. Song diversity stabilized in patches as small as 100 ha supporting as few as 8-20 males. Song repertoires of smaller patches were significantly poorer. Almost one-quarter of the study populations inhabited patches <100 ha, and their cultural traditions appear to have eroded. The analysis of spatial patterns in birdsong may be a useful tool for detecting subpopulations prone to extinction.

Detection and visualization of spatial genetic structure in continuous Eucalyptus globulus forest

Visualizing the pattern of variation using microsatellites within a Eucalyptus globulus forest on the island of Tasmania provided surprising insights into the complex nature of the fine-scale spatial genetic structure that resides in these forests. We used spatial autocorrelation and principal coordinate analysis to compare fine-scale genetic structure between juvenile and mature cohorts in a study area, 140 m in diameter, located within a typical, continuous E. globulus forest. In total, 115 juvenile and 168 mature individuals were genotyped with eight highly polymorphic microsatellite loci. There was no significant difference in the level of genetic diversity between cohorts. However, there were differences in the spatial distribution of the genetic variation. Autocorrelation analysis provided clear evidence for significant spatial genetic structure in the mature cohort and significant, but weaker, structure in the juvenile cohort. The spatial interpolation of principal coordinate axes, derived from ordination of the genetic distance matrix between individuals, revealed a spatially coherent family group which was evident in both cohorts. Direct comparison of the genetic structure within each cohort allowed visualization of a shift in the spatial distribution of genetic variation within the population of approximately 10 m. As the shift coincided with the direction of prevailing winds, it is hypothesized that this phenomenon is due to downwind dispersal of seeds and is indicative of the important role of prevailing winds in forcing eastward gene flow in these high-latitude forests.

Oceanographic habitat of an endangered Mediterranean procellariiform: implications for marine protected areas

Marine protected areas (MPAs) require ecologically meaningful designs capable of taking into account the particularities of the species under consideration, the dynamic nature of the marine environment, and the multiplicity of anthropogenic impacts. MPAs have been most often designated to protect benthic habitats and their biota. Increasingly, there is a need to account for highly mobile pelagic taxa, such as marine birds, mammals and turtles, and their oceanic habitats. For breeding seabirds foraging from a central place, particular attention should be paid to distant foraging grounds and movement corridors, which can often extend to hundreds of kilometers from breeding colonies. We assessed the habitat use by the most threatened Mediterranean seabird, the Balearic Shearwater, Puffinus mauretanicus, using vessel-based surveys during the chick-rearing period (May-June). We used a hierarchical modeling approach to identify those environmental variables that most accurately reflected the oceanographic habitat of this species by (1) delineating its foraging range using presence/ absence data and (2) identifying important foraging grounds where it concentrates in dense aggregations. The foraging range comprised the frontal systems along the eastern Iberian continental shelf waters (depth <200 m) and areas close to the breeding colonies in the Balearic Islands. Shearwaters aggregated in productive shelf areas with elevated chlorophyll a concentrations. Following the model of a core-buffer MPA, we envisioned those areas of dense aggregation (i.e., the area of influence of the Ebro River discharge and Cape La Nao regions) as the core regions deserving elevated protection and more stringent management. More diffuse protective measures would be applied within the larger buffer region, delineated by the foraging range of the species. Marine zoning measures can greatly benefit the conservation of the Balearic Shearwater and other far-ranging seabirds by extending protective measures beyond their breeding colonies during both the breeding and non-breeding seasons.

Techniques for accuracy assessment of tree locations extracted from remotely sensed imagery

Remotely sensed imagery is becoming a common source of environmental data. Consequently, there is an increasing need for tools to assess the accuracy and information content of such data. Particularly when the spatial resolution of imagery is fine, the accuracy of image processing is determined by comparisons with field data. However, the nature of error is more difficult to assess. In this paper we describe a set of tools intended for such an assessment when tree objects are extracted and field data are available for comparison. These techniques are demonstrated on individual tree locations extracted from an IKONOS image via local maximum filtering. The locations of the extracted trees are compared with field data to determine the number of found and missed trees. Aspatial and spatial (Voronoi) analysis methods are used to examine the nature of errors by searching for trends in characteristics of found and missed trees. As well, analysis is conducted to assess the information content of found trees.

Population structure inferred by local spatial autocorrelation: An example from an Amerindian tribal population

Spatial autocorrelation (SA) methods were recently extended to detect local spatial autocorrelation (LSA) at individual localities. LSA statistics serve as useful indicators of local genetic population structure. We applied this method to 15 allele frequencies from 43 villages of a South American tribe, the Yanomama. Based on a network of links <or=51 km between neighboring villages, we calculated LSA statistics for Moran, Geary, and Getis-Ord coefficients. We also developed two new, rescaled indices of local SA. Local indicators of positive SA highlight villages surrounded by genetically similar near neighbors. Negative LSA statistics indicate sharp genetic differences from near neighbors. Markedly positive LSA was found for all 11 outlier villages. The most negatively LSA villages are in the central, densely connected cluster. The Getis-Ord coefficients of suitably transformed allele frequencies point to clusters of villages with unusually high or low allelic polymorphisms. The most homozygous villages are all in the four geographically isolated village clusters. The most polymorphic villages are all in the large, densely settled Yanomame dialect group. An ad hoc linguistic isolation index between neighboring villages showed that villages in isolated pairs and triplets have linguistically similar neighbors, whereas nine villages with notably negative LSA are all near dialect and kinship boundaries. The location of a village with respect to the graph structure of its neighborhood affects its LSA and genetic polymorphism. The implications of these findings for the population structure of the Yanomama are compatible with those from an earlier study of global SA in these villages.

Spatial structure of shell polychromatism in populations of Cepaea nemoralis: new techniques for an old debate

A conspicuous shell polychromatism is observed in colonies of Cepaea nemoralis from western France (Brittany). The present study is intended to search for a spatial structure of shell features at this scale and to infer evolutionary processes from the observed patterns. We used a database of morph frequencies (six composite phenotypes were retained) measured on 213 samples regularly distributed on the whole studied area. Data analysis was based on two distinct multivariate methods leading to the following steps: (i) to search for a structure without reference to environmental conditions with a method (global principal components analysis: GPCA) which takes into account the spatial information by means of a neighbouring relationship between sampling points (Delaunay triangulation); (ii) to test the structuring power of environmental conditions by means of two explanatory factors (distance from the sea, altitude) involved in a redundancy analysis (RDA); (iii) to search for a spatial structure using residuals of the previous analysis, ie, after removing effects of environmental conditions. Global covariance accounted for 26.4% of the total variance, leading to a highly significant autocorrelation for each phenotype (step 1). Geographical mapping of factorial scores resulting from global analysis showed a well structured littoral zone and a strong southern-northern inland differentiation. Sixteen percent of the total variance was expressed in RDA but all morphs were not equally concerned. After removing environmental effects, a significant spatial structure still remains but was essentially caused by random processes. We argue for the importance of these last phenomena.

Reproductive performance links to fine-scale spatial patterns of female grey seal relatedness

Fine-scale spatial patterns of female relatedness throughout the established grey seal breeding colony of North Rona, Scotland, were investigated by accurate mapping and spatially explicit analyses of a large sample (n = 262) of mothers using variation at nine microsatellite DNA loci. Local spatial autocorrelation analyses identified locations where seals were more highly related to the colony than average. These locations were also areas where the more successful females bred, were occupied first during each breeding season, were centrally placed locations of preferred habitat types and were likely to be the locations which were the first to be colonized historically. Mothers occupying such sites achieved higher than average pup growth rates, suggesting a founder fitness benefit.