Abstract | Full Paper | PDF | Printer Friendly Format

Page 3 of 3
| Previous |


Surface Water Monitoring And Evaluation of Indravati Reservoir using the application of Principal Component Analysis using Satellite Remote Sensing Technology


Table: 1 Eigen Values Computed for the Covariance Matrix

Bands Eigen Values Percentage Difference Total Variance
1 701.54 57.2 401.74 1226.32
2 299.80 24.4 155.36
3 144.44 11.8 95.62
4 48.82 3.98 21.92
5 26.90 2.19 22.08
6 4.82 0.39 ----


Table: 2 Eigenvectors (akp) (Factors Scores) computed for the Covariance Matrix

Bands 1 2 3 4 5 6
1 0.34 -0.26 0.11 -0.52 -0.50 0.52
2 0.55 -0.23 0.15 -0.42 0.55 -0.39
3 0.54 0.43 0.55 0.42 -0.20 -0.02
4 0.22 -0.35 -0.31 0.19 -0.57 -0.61
5 0.39 -0.38 -0.37 0.54 0.29 0.45
6 0.32 0.65 -0.65 -0.22 -0.02 0.02

Percentage of total variance in the data explained by each component:

Results And Discussion
Using Principal component analysis the changes of Indravati reservoir studied in the following: -
  1. Change in the turbidity
  2. Loss of water levels resulting in the sinkage of the area along the periphery, and
  3. Loss of water levels resulting in discontinuities of the reservoir.
The reflection properties of different bands depict in the above changes can be seen in the figure No. 2 (Principal Component Analysis).

The turbidity levels or better reflectance in band -2 (Green) data and the compared evaluation showed the turbidity levels relatively higher the northern parts of the reservoir. This should be viewed in terms of not only characteristics of the watersheds contributing to the northern part of the reservoir, but also the depth levels of the reservoir. On the other hand the band -4 (NIR) data because of the use water absorption characteristics, the discontinuities of the reservoir or clearly seen in the band -4 in April data. The total reservoir area is well exhibited in band -4 of January data.

In order to bring out all these temporal characteristics to study the reservoir changes,the Principal Component Analysis was done.

It can be seen from the table (eigen2) the band -4 data of both the season contributed to large variability of PC -2 and PC -3 where in predominantly in the temporal changes described above were reflected.

Predominantly the changes in water land along the periphery another hand PC -3 component exhibited the changes both from turbidity levels and also the total loss of water resulting in discontinuing.

Hence, the False color composite (FCC) of PCA -3, PCA -2 and PCA -1 was made to brought the total reservoir area and the temporal changes in different color tones. In this way the surface water monitoring using Remote Sensing data analysis will be useful to understand the surface water bodies different characteristics area.

Surface Water Monitoring Using Multi-Temporal Satellite Data


References
  • John R. Jensen, “Introductory Digital Image Processing”- A Remote Sensing Perspective, Second edition.
  • Jensen. S.K and J.O.Domingue, 1988, Extraction topographic structure from digital elevation data for geographic system analysis, Photogrametry Engineering and Remote Sensing, vol.54, No.11, pp.1593-1600.
  • Manual of Procedure for preparation of wastelands digital data base using Remote Sensing & GIS techniques, prepared by NRSA, Dept. of Space, Govt. of India, Balanagar, Hyderabad-37.
  • Remote Sensing Application in Water Resources, The Indian Society for Technical Education, Lecture Notes, May 15-June4, 1985, Dept of Civil Engineering, Anna University, Madras-25.
  • Thomas M. Lillesand and Raiph W. Kiffer “Remote Sensing and Image Interpretation”, second edition, 1986.
  • Tanji K.K. and B. Yaron (Eels), Management of water use in Agriculture, Springer- Verlag, Berlin Heidelberg, New York.
Page 3 of 3
| Previous |