The Asian black bear (Ursus thibetanus) inhabits two of the main islands, Honshu and Shikoku, in Japan. To determine how climatic oscillations during the Quaternary Era affected the genetic structure of the black bear populations in Japan, we examined their phylogeographic relationships and compared their genetic structure. We analysed an approximately 700-bp sequence in the D-loop region of mitochondrial DNA collected from 589 bears in this study with 108 bears from a previous study. We observed a total of 57 haplotypes and categorized them into three clusters (Eastern, Western and Southern) based on the spatial distribution of the haplotypes. All but 2 of the 41 haplotypes in the Eastern cluster were distributed locally. Genetic diversity was generally low in northern Japan and high in central Japan.Demographic tests rejected the expansion model in northern populations. Haplotypes of the Western and Southern clusters were unique to local populations. We conclude that the extant genetic structure of the Asian black bear populations arose as follows: first, populations became small and genetic drift decreased genetic diversity in the northern area during the last glacial period, whereas large continuous populations existed in the southern part of central Japan. These patterns were essentially maintained until the present time. In western and southern Japan, the effects of climatic oscillations were smaller, and thus, local structure was maintained.

No abstract

No abstract

Context Genetic diversity is one of the most important facets of biological diversity, and changes in the spatial pattern of habitats, often modified by human activity, are believed to have affected the genetic diversity of resident natural populations. Objectives We undertook a landscape genetic analysis in order to determine which landscape features influence gene flow within Asian black bear populations and to identify the underlying processes. Methods In our evaluation of gene flow, we estimated four parameters of resistance with regard to landscape elevation: the mean, the difference between the highest and lowest, the standard deviation, and the coefficient of variation of elevation among individuals. We then examined the resistance effect of different land use types. Results With the exception of mean elevation, we found that all parameters showed a significant relationship with genetic distance, indicating that unevenness in elevation provides functional resistance to gene flow. Although we found no evidence of landscape barriers (isolation‐by‐barrier), there was an indication of landscape resistance (isolation‐by‐resistance). Urban area and farmland are suggested to be the strong factors contributing to the resistance to gene flow, even though isolation‐by‐distance was also detected. When we examined gene flow for pairs of males and pairs of females, both isolation‐by‐distance and isolation‐by‐resistance were stronger in order of female pairs, male pairs, all individual pairs. Conclusions We conclude that landscape resistance was detectable with a high contrast in landscape heterogeneity and they are more influential on females than males. OPEN PRACTICES This article has been awarded Open Data badge. All materials and data are publicly accessible via the Open Science Framework at https://doi.org/10.5061/dryad.gn0qf16 . Learn more about the Open Practices badges from the Center for Open Science: https://osf.io/tvyxz/wiki .

Non-invasive DNA genotyping using hair samples has become a common method in population surveys of Asiatic black bears (Ursus thibetanus) in Japan; however, the accuracy of the genotyping data has rarely been discussed in empirical studies. Therefore, we conducted a large-scale pilot study to examine genotyping accuracy and sought an efficient way of error-checking hair-trapping data. We collected 2,067 hair samples, successfully determined the genotypes of 1,245 samples, and identified 295 individuals. The genotyping data were further divided into 3 subsets of data according to the number of hairs used for DNA extraction in each sample (1-4, 5-9, and C10 hairs), and the error rates of allelic dropout and false alleles were estimated for each subset using a maximum likelihood method. The genotyping error rates in the samples with C10 hairs were found to be lower than those in the samples with 1-4 and 5-9 hairs. The presence of erroneous genotypes among the identified individuals was further checked using a post hoc goodness-of-fit test that determined the match between the expected and observed frequencies of individual homozygotes at 0-6 loci. The results indicated the presence of erroneous genotypes, possibly as a result of allelic dropout, in the samples. Therefore, for improved accuracy, it is recommended that samples containing C10 hairs should be used for genotyping and a post hoc goodness-of-fit test should be performed to exclude erroneous genotypes before proceeding with downstream analysis such as capture-markrecapture estimation.