Abstract

Introduction

Background

   GreenRevolution

   Gujarat, India

   SatelliteImagery

 

Part I-Vegetation Derivation

Methods

Results/Discussion

Conclusions

 

Part II-Land Cover Change

    Dams

    Irrigation

    Desertification

Conclusions

 

Final Thoughts

Acknowledgements

Works Cited 

List of Figures and Appendix

 

Figure 14

Implications of Soil & Saturation Effects in NDVI for Vegetation Change Analysis

The diagram above demonstrates the various ways that soil and saturation problems may affect vegetation change results. The triangular shape is representative of the scatter plot of the two methods for detecting vegetation abundance (i.e. Figure 7). The circle represents a single pixel changing over time. Here a pixel starts with very high percent vegetation cover (represented by the right-most circle). If vegetation cover decreases slightly then NDVI may not detect any change while SMA should detect the subtle decrease. If that pixel then decreases further to an abundance that allows for soil backgrounds to become significant, then further accuracy problems should arise using NDVI. If the background soil of that pixel is extremely dark then change B would be calculated as the vegetation change according to NDVI. If that same amount of change occurred but the soil background in that pixel was extremely bright then change A would be the calculated vegetation change according to NDVI. However, SMA should detect the same amount of vegetation change for that pixel regardless of the background color (Elmore et al., 2000).

Back to Index

For more information please contact Laurel_Firestone@yahoo.com