Detection of Iran's Vegetation Seasons Using NDVI

Introduction and Awareness of vegetation and its health is representative of climate conditions for each place. To study and monitor the vegetation on a global and regional scale access to timely field data are usually difficult and limited. Estimation of vegetation on an ordinary method that includes an overall of vegetation is time-consuming and does not provide precise information. Thus remote sensing is a very helpful way that is in priority as provide a large overview and is repetitive and time can be saved using this method. Application of vegetation indices that is widely used can be exploited to quantify the net annual products in planetary and other scales. The present study is motivated to find out a methodology to detect and classify vegetation and also to detect Iran’s vegetation seasons.
In this effort as the first step MODIS_Aqua 16 days of Normalized Difference Vegetation Index (NDVI) from 1381.4.13 to 1393.12.23 were downloaded from MODIS web site. And then nearly 10 billion codes were analyzed in matlab to calculate the mean 16 days of NDVI and then a Matrix of 23*7541502 was constructed. In the last step, the cluster analysis was applied to the data using Ward method.
The investigations indicated that according to the spatial values of NDVI five separate seasons are observable in the country as follow densely (February 26th to April 15th), very dense (April 15th to June 2nd), sparse (June 2nd to October 24th), transient (October 24th to November 25th) and very sparse (November 25th to February 26th) seasons.
Discussion and In general, vegetation is massive in the spring, sparse in summer and winter, and the autumn is the transition season.