Dataset Information

Classification of Tree Species in Overstorey Canopy of Subtropical Forest Using QuickBird Images.

ABSTRACT: This paper proposes a supervised classification scheme to identify 40 tree species (2 coniferous, 38 broadleaf) belonging to 22 families and 36 genera in high spatial resolution QuickBird multispectral images (HMS). Overall kappa coefficient (OKC) and species conditional kappa coefficients (SCKC) were used to evaluate classification performance in training samples and estimate accuracy and uncertainty in test samples. Baseline classification performance using HMS images and vegetation index (VI) images were evaluated with an OKC value of 0.58 and 0.48 respectively, but performance improved significantly (up to 0.99) when used in combination with an HMS spectral-spatial texture image (SpecTex). One of the 40 species had very high conditional kappa coefficient performance (SCKC ? 0.95) using 4-band HMS and 5-band VIs images, but, only five species had lower performance (0.68 ? SCKC ? 0.94) using the SpecTex images. When SpecTex images were combined with a Visible Atmospherically Resistant Index (VARI), there was a significant improvement in performance in the training samples. The same level of improvement could not be replicated in the test samples indicating that a high degree of uncertainty exists in species classification accuracy which may be due to individual tree crown density, leaf greenness (inter-canopy gaps), and noise in the background environment (intra-canopy gaps). These factors increase uncertainty in the spectral texture features and therefore represent potential problems when using pixel-based classification techniques for multi-species classification.

SUBMITTER: Lin C

PROVIDER: S-EPMC4433356 | biostudies-literature | 2015

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Classification of Tree Species in Overstorey Canopy of Subtropical Forest Using QuickBird Images.

Lin Chinsu C Popescu Sorin C SC Thomson Gavin G Tsogt Khongor K Chang Chein-I CI

PloS one 20150515 5

This paper proposes a supervised classification scheme to identify 40 tree species (2 coniferous, 38 broadleaf) belonging to 22 families and 36 genera in high spatial resolution QuickBird multispectral images (HMS). Overall kappa coefficient (OKC) and species conditional kappa coefficients (SCKC) were used to evaluate classification performance in training samples and estimate accuracy and uncertainty in test samples. Baseline classification performance using HMS images and vegetation index (VI) ...[more]

PMID: 25978466

Similar Datasets

Project description:As a transitional vegetation type between evergreen broadleaved forest and deciduous broadleaved forest, evergreen-deciduous broadleaved mixed forest is composed of diverse plant species. This distinctive forest is generally distributed in mountainous areas with complex landforms and heterogeneous microenvironments. However, little is known about the roles of environmental conditions in driving the species diversity patterns of this forest. Here, based on a 15-ha plot in central China, we aimed to understand how and to what extent topographical characteristics and soil nutrients regulate the number and relative abundance of tree species in this forest. We measured environmental factors (terrain convexity, slope, soil total nitrogen, and phosphorus concentrations) and species diversity (species abundance distribution and species richness) in 20 m × 20 m subplots. Species abundance distribution was characterized by skewness, Berger-Parker index, and the proportion of singletons. The generalized additive model was used to examine the variations in diversity patterns caused by environmental factors. The structural equation model was used to assess whether and how topographical characteristics regulate species diversity via soil nutrients. We found that soil nutrients had significant negative effects on species richness and positive effects on all metrics of species abundance distribution. Convexity had significant positive effects on species richness and negative effects on all metrics of species abundance distribution, but these effects were mostly mediated by soil nutrients. Slope had significant negative effects on skewness and the Berger-Parker index, and these effects were almost independent of soil nutrients. Soil nutrients and topographical characteristics together accounted for 9.5-17.1% of variations in diversity patterns and, respectively, accounted for 8.9-13.9% and 3.3-10.7% of the variations. We concluded that soil nutrients were more important than topographical factors in regulating species diversity. Increased soil nutrient concentration led to decreased taxonomic diversity and increased species dominance and rarity. Convexity could be a better proxy for soil nutrients than slope. Moreover, these abiotic factors played limited roles in regulating diversity patterns, and it is possible that the observed patterns are also driven by some biotic and abiotic factors not considered here.

Project description:Wood density (WD) is not only an important parameter to estimate aboveground biomass but also an indicator of timber quality and plant adaptation strategies to stressful conditions (i.e., windthrow, pests, and pathogens). This study had three objectives: (1) to compare WD among seven subtropical tree species; (2) to determine how tree growth traits may influence possible differences in WD between the pioneer and shade-tolerant species; and (3) to examine whether or not WD differs by tree social status (dominant vs. suppressed trees) within species. To do this, 70 trees were destructively harvested. From each tree, disks at different stem heights were obtained and subjected to a method of stem analysis to measure whole tree level WD. The results showed that WD differed significantly among the seven species (p < .001). Their average WD was 0.537 g/cm3, ranging from 0.409 g/cm3 for Choerospondias axillaris to 0.691 g/cm3 for Cyclobalanopsis glauca. The average WD of the four pioneer species (0.497 ± 0.13 g/cm3) was significantly lower (p < .01) than that of the three shade-tolerant species (0.589 ± 0.12 g/cm3). The WD of the pioneers had a significant positive correlation with their stem diameter at breast height (DBH), tree height (H), and tree age, but WD had a significant negative correlation with relative growth rate (RGR). In contrast, the WD of the shade-tolerant tree species had no significant relationships with DBH, H, tree age, or RGR. The dominant trees of the pioneer species had a higher WD than the suppressed trees, whereas the shade-tolerant species had a lower WD for dominant trees than the suppressed trees. However, the differences in WD between dominant and suppressed trees were not significant. Taken together, the results suggest that classifying species into pioneer and shade-tolerant groups to examine the effects of tree growth traits and social status could improve our understanding of intra- and interspecific variation in WD among subtropical tree species.

Dataset Information

Classification of Tree Species in Overstorey Canopy of Subtropical Forest Using QuickBird Images.

Publications

Classification of Tree Species in Overstorey Canopy of Subtropical Forest Using QuickBird Images.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets