| VEGETATION DYNAMICS TREND USING SATELLITE TIME SERIES IMAGERY |
| Paper ID : 1249-SMPR |
| Authors: |
|
Zeynab Najafi1, Parviz Fatehi2, Ali Asghar Darvish Sefat *1 1Department of Forestry, Faculty of Natural Resources, University of Tehran 2Assistant Prof., Forestry and Forest Economics, Faculty of Natural Resources, University of Tehran, Karaj, Iran |
| Abstract: |
| In this study, the trend of vegetation dynamics in Kermanshah city assessed using NDVI MOD13Q1 product during 2000-2017. Based on time series imagery the pick of vegetation phenology stage (maximum mean NDVI) identified, and then the trend vegetation dynamic were investigated using Ordinary Least Square regression and Thiel Sen approaches. To generate a pixel-wise trend map, a pixel-based vegetation dynamics was implemented. A non-parametric Mann-Kendall statistics was used to evaluate the trend significance. The results showed, maximum mean NDVI has occurred in the first half or second half of April. Trend analysis using regression and Theil Sen methods indicated no-trend. The pixel-based trend assessment using regression showed that 50% of the study area faced a positive trend and reaming part faced a negative trend. But Theil Sen method was representative the no-trend for a large majority of area. The Mann-Kendall test indicated that only 20 percent of the area shows a statistically significant trend. |
| Keywords: |
| Mann-Kendall; NDVI; Regression; Thiel Sen; Time Series; Vegetation dynamic |
| Status : Conditional Accept (Poster) |
