Monitoring the impacts of drought on land use/cover: a developed object-based algorithm for NOAA AVHRR time series data
This project was implemented on Zayandeh-Rud basin which is one of the most important and strategic regions of Iran. The main goal of the research was to monitor and evaluate the temporal land use/cover changes affected by drought episodes over a 10-year period (1992-2003). Primarily, raw weather da...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Asian Network for Scientific Information
2011
|
| Online Access: | http://psasir.upm.edu.my/id/eprint/23159/1/Monitoring%20the%20impacts%20of%20drought%20on%20land%20usecover%20a%20developed%20object-based%20algorithm%20for%20NOAA%20AVHRR%20time%20series%20data.pdf http://psasir.upm.edu.my/id/eprint/23159/ http://scialert.net/abstract/?doi=jas.2011.3089.3103 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Universiti Putra Malaysia |
| Language: | English |
| Summary: | This project was implemented on Zayandeh-Rud basin which is one of the most important and strategic regions of Iran. The main goal of the research was to monitor and evaluate the temporal land use/cover changes affected by drought episodes over a 10-year period (1992-2003). Primarily, raw weather data preparation activities were carried out to produce Standard Precipitation Index (SPI) spatial point thematic layer as a base map, then, satellite image pre-processing works applied on the remotely sensed data to generate the time series land use/cover maps. As a novel idea in this study, it developed a new object-based classification algorithm for AVHRR (Advanced Very High Resolution Radiometer) data. The model works based on the seasonal values of Normalized-difference Vegetation Index (NDVT) in the study area. The algorithm was statistically compared with maximum likelihood supervised classification method. Results demonstrated an increase in overall accuracy from 74.34 to 90.07% and the Kappa index from 70.58 to 88.8%. Based on the statistical analysis results, drought had not statistically significant effect on the land use/cover changes. Therefore, there are more important factors other than natural weather drought in the study area and these could be summarized as a term of disorder on land management in the Zayandeh-Rud river basin. Results of the investigation provide a scientific tool enabling governmental land and water managers to monitor, make decision and manage the water crises due to drought upon the catchment. |
|---|