|
Mapping from Space & GPS
|
Effect of coherence on dems derived from sar interferometry:
A case study of mayon volcano, philippines
Relationship between coherence and DEM accuracy
Here, by taking profile in DEMtopo, DEMInSAR and coherence images, we want to check the effect of coherence on the height difference. The result is illustrated in Figure 5. On the left side, i.e. high coherence value, two profile curves from DEMInSAR and DEMtopo are nearly coincident. On the other side, i.e. layover area with low coherence value, the profiles show the bigger differences. Then, the comparison was done more statistically. The distribution of coherence values in both data sets is illustrated in Table 3. Both data sets introduce low coherence value.
Table 3 - Statistic result of coherence value
| Data | Min | Max | Mean | Stdev |
| 1996 | 0.01 | 0.87 | 0.365 | 0.142 |
| 1997 | 0.01 | 0.85 | 0.255 | 0.138 |
Figure 5 - Illustration of profile taken at azimuth 248Comparison of DEMInSAR and DEMtopo (left); Coherence (right)
In Figure 6 (left), we can see that when coherence value increases, the standard deviation of height difference between two DEMs decreases quickly. This trend follows exponential function.
Relationship between coherence and NDVI
There are many factors that affect on phase coherence. Here, we consider in land cover. We have already known that when land cover changes with the time, the coherence has the trend of decreasing. The correlation analysis between 1996 coherence, 1997 coherence and NDVI (see Table 4) shows that it is also correct here. The graph shown the trend of coherence coefficient against NDVI can be seen in Figure 6 (right).
Table 4 - Correlation analysis result
| Data | 96 vs 97 | 96 vs NDVI | 97 vs NDVI |
| Correlation Coefficient | 0.799 | -0.903 | -0.797 |
As we knew, SAR ERS1&2 data with C band and VV polarization weakly interacts to vegetation. The results from this research introduce some interesting further researches in this point.
Figure 6 - Relationship between coherence and DEM accuracy (left)Relationship between NDVI and coherence (right)
4. Conclusion
In this paper, ERS1&2 tandem data was used for applying of SAR interferometry in tropical Asian region. The result shows very low coherence value. However, by using 1996 data, we can make DEM using SAR interferometry. In conclusion, DEM accuracy decreases following the exponential function when coherence value increases. In addition, this paper also concluded that the coherence coefficient decreases when NDVI increases.
From this study, the results can be used for the further researches either in application site or in technique site. Phase coherence is very important in SAR interferometry research. This kind of research should be conducted for Southeast Asian region.
5. References
-
Gens, R., Van Genderen, J.L., 1996. Review Articles SAR interferometry: issues, techniques, applications. Int. J. Remote Sensing, 17 (10), pp. 1803-1835.
- Kenyi, L., Raggam, H., 1996. Accuracy assessment of Interferometrically derived DTMs. Fringe 1996.
- Qian, Lin, Vesecky, F.J., Zebker, H.A., 1994. Comparison of elevation derived from InSAR data with DEM over large relief terrain. Int. J. Remote Sensing, 15, pp. 1775-1790.
- Madsen, S.N, Zebker, H.A., Martin, J.M., 1993. Topographic Mapping Using Radar Interferometry: Processing Techniques. IEEE Trans. GE and RS, 31 (1), pp. 246-256.
- Seymor, M. S., 1999. Refining Low-Quality Digital Elevation Models Using Synthetic Aperture Radar. Doctoral Thesis, University of British Columbia, Canada.
- Wehr, A., et al., 1996. The effects of different land covers on the accuracy of interferometric DEM. Fringe 1996.
- Zebker, H.A., Werner, C., Rosen, P.A., Hensley, S., 1994. Accuracy of Topographic Maps Derived from ERS-1 Interferometric Radar. IEEE Trans. GE and RS, 32 (4), pp. 823-836.
- Zebker, H.A., Villasenor, J., 1992. Decorrelation in Interferometric Radar Echoes. IEEE Trans. GE and RS, 30 (5), pp. 950-959.
- Final Report, 1999. Remote Sensing and GIS analysis for base data creation and change detection; the study on comprehensive disaster prevention around Mayon volcano in the Republic of The Philippines. (Source Asian Center for Research on Remote Sensing)
|
|
|
|
|
