GIS-Aided Small-Scale Quantitative Vegetation Analysis.
|
 |
|
Post a Comment
|
|
|
|
|
|
ABSTRACT:
Sessile plant individuals only interact with neighbours over restricted distances. Due to the dispersal ability and the death of individuals, the neighbour could be the same species or not. Neighbourhood composition and neighbourhood interaction would be affected by both biotic and abiotic factors, thus the plants appear as different spatial pattern, various neighbourhood effects, and unlike neighbourhood compositional and structural diversity. Pattern, neighbourhood effects and neighbourhood diversity are mutually influenced and then constraint the composition, dynamics and diversity of a community. Quantitative vegetation analysis focuses on the spatio-temporal pattern of populations or communities. The quantitative techniques mainly include plot methods and plotless methods, and the latter are based on neighbourhood/distance. Nearest-neighbour method is adopted in the dissertation, which can be divided into two categories: single nearest-neighbour method and multiple nearest-neighbour method. The former is simple and convenient, and the latter is accurate but complicated. We should choose one with the research objective.
Being a useful analytical tool for landscape and global ecology, GIS (Geographic Information Systems) also has some advantages for small-scale pattern analysis. In this paper, GIS-aided quantitative vegetation analyses are applied for a tropical ravine forest at Bawangling National Nature Reserve, Hainan Island. Sampling techniques, intra-/inter- specific pattern (including distribution, interspecific segregation and interspecific association and their mutual relationship), neighbourhood effects and neighbourhood-based diversity indices are discussed. The function of GIS in the above small-scale vegetation analysis includes: 1. creation, query and visualization of spatial data (e.g. tables, figures). 2. tabular operations (e.g. summary, calculate, simple statistics). 3. nearest neighbour analysis. 4. proximity mapping. 5. correlation coefficient tool. 6. measuring spatial attributes (e.g. distance, area). In the dissertation, we have proposed the concepts of "minimal neighbour number" and "heterospecific attraction", and a new neighbourhood-based diversity index. Besides, we have introduced Haberman's residual analysis, Goodman's loglinear method and Williams's multiple nearest neighbour method for vegetation pattern analysis. And then we classify neighbouhood-based diversity indices into two categories: compositional and structural. Moreover, a statistical procedure for Excel tables (EXCELLENT) is programmed with VC++, which largely improve the effectiveness of vegetation studies. Many pattern analytical methods provided here are also suitable for large-scale pattern analysis.
|
|
|
|
STATISTICS
|
|
Click on # to view
|
|
Citations
|
|
5
|
|
References
|
|
0
|
|
Comments
|
|
0
|
|
Quality
|
|
0/0.00
|
|
Interest
|
|
0/0.00
|
|
View(er)s
|
|
2/415
|
|
|
|
|
|
|
| Prev |
Next |
|