Mapping and assessing spatial multiscale variations of birds associated with urban environments in metropolitan Taipei, Taiwan.

Environmental change can be monitored and assessed by analyzing changes in bird populations. This study employed multivariate factorial kriging (MFK) to determine the multiscale changes in the distribution of five commonly observed bird species to urban environmental change. Areas of metropolitan Taipei were analyzed to determine the extent and effect of land cover. A digital elevation model and normalized difference vegetation index were also constructed. Spatial patterns of variation in bird populations were analyzed by MFK at a regional scale (16 km) and local scale (2 km). Comparison of MFK results with those obtained by conventional principal component analysis (PCA) and correspondence analysis (CA) showed that the conventional PCA results were comparable to the MFK results at the local scale, and the CA results were comparable to MFK results at the regional scale. The MFK indicated that local-scale variation accounts for a larger proportion of the total variation than regional-scale variation, and local-scale variation is affected by more factors than regional scale variation. The MFK results also showed that heterogeneous elevation exerts a geographical influence on bird populations at the regional scale whereas the land cover of urban meadows in densely populated urban areas of Taipei exerts a local influence. The suburban trees and meadows in belt regions between suburbs have local and regional influences on bird distributions, respectively. The spatial patterns of bird populations in response to urban environments can be adequately explained by their diet categories and body sizes. The MFK, which simultaneously provides descriptive variations and spatial patterns in species at determinate scales, facilitates environmental interpretations of ordination results and identification of determinants for conservation efforts at multiple scales.