Relationship between canopy brdf and physical parameters Of 3-d structure of vegetaion in northern wetlands in japan

3.3.1 Multi Angle Spectral Reflectance Measured at Single Bands:
At first the spectral reflectance in the Green band ( 530 -550nm ), the Red band ( 650 - 670nm ), and the NIR band ( 760 - 810nm ), were measured with multi angular. From the NIR and Green band point of view, we could not classifier the each spectral reflectance but could discriminate the difference by the measurement of the oblique forward 15°- 45°. Especially, it was very effective to discrimination of Sphagnum moss. In addition to that, it was also effective to the classification for the vegetation mixed with shrub. For concrete, when the shrub, Chamaedafune invaded to Carex vegetation, the reflectance of Carex vegetation became small. That is to say, we could extract the spectral reflectance of the vegetation mixed the shrub stratum, especially 10% in the Green band, 40% in the NIR band. The processing of the calculation of the band ratio for the spectral reflectance, NDVI and VI were effective to extract the typical three wetland vegetation type measured at ±15°rather than nadir angle. In addition to that, in Red/NIR, Green/NIR and NDVI, we could not find the effective differences except for the Sphagnum moss group, site C however effective to the vegetation mixed with Sphagnum and the shrub ( Fig.6 ).



3.3.2 Relationships between Band Ratio and LAI:
By selecting the effective and specific band ratio and vegetation indices with multi angles to discriminate each vegetation, it was possible to classifier the specific wetland vegetation by using the spectral data measured at 2m high above the canopy. It proved that even in the case that the difference of LAI was slight though the band ratio was actually quite different or the case that the difference of band ratio was similar though the LAI was quite different, the measurement at not only multi angles but also the nadir angle was very effective to the classification.

4. Discussion

4.1 Classification of Wetland Vegetation Concerning Canopy Productive Structure and Multi Angle Spectral Signature
It was obviously easy to extract such as Sphagnum from other vegetation without multi spectral measurement and was enough for only nadir angle. However, for the vegetation which had the similar kinds of canopy, such as both phleum and Carex were the first dominant species and the second dominant species were mixed with other herb species, multi angular measurement was evidently effective to discriminate from other similar kinds of vegetation. In addition to that, taking biomass into considerations, it is distinctly effective to compare and discriminate following type of vegetation. The vegetation with different spectral signature though the actual biomass was quite similar and the vegetation with quite different biomass though the actual reflectance was similar. It was verified that the multi spectral measurement was effective to the classification of the vegetation whose canopy has similar herb stratum. When we grasp wide range of wetland vegetation, we will be able to understand and estimate the spatial arrangement of the vegetation community by specifying the indefinite spectral signature in detail. As one of the basic technique of the remote sensing of vegetation, it appeals that it is important to clarify the physical relationship between the biological data including biomass and spectral signature with multi observation for verifying the effectiveness of remote sensing data. In the near future, it seems that the focus are making on the various spectral principles applied for the remote sensing data for the analysis of the research in order to clarify the spectral characteristics relating to the function of the ecosystem.

5. Conclusions
Following three statements are concluded in this study.

1. Each productive structure was specified by the relationship between LAD- SLA in each layer.
2. The characteristic of spectral signature measured at multi angle was clarified.
3. By clarifying the relationship between the effective band ratio, it became clearly possible to classify the specific types of wetland vegetation especially herb stratum, mixed shrub stratum, and the only moss vegetation and by selecting specific vegetation index and observation angle, it became remarkably possible to estimate the biomass in each vegetation type in detail.

6. References

1. Asrar, G., Kanemasu, E.T., Miller, G.P., and weiser, R.L., 1986. Light interception and leaf area estimates from measurements of grass canopy reflectance. IEEE Trans. Geosci.remote Sens.GE-24:76-82
2. Christensen, S., and Goudriaan, J.,1993. Deriving Light Interception and Biomass from Spectral Reflectance Ratio, Remote Sens.Environ.43:87-95.
3. Watson R. T., Zinyowera, C. Z., and Moss, R. H.,1996. 'Climate Change1995': Impacts, Adaptations and Mitigation of Climate Change: Scientific-Technical Analyses, Working Group?to the Second Assessment Report of the Intergovernmental Panel on Climate Change, Intergovernmental Panel on Climate Change, Cambridge University Press