Investigation of The Ten-Year Change of Longos Forests with NDVI Time Series Analysis Using Google Earth Engine (GEE)

Creative Commons License

Kafes B., Yılmaz O. S., Balık Şanlı F.

2. International Baku Scientific Researches Conference, Baku, Azerbaijan, 28 - 30 April 2021, vol.1, pp.287-292

  • Publication Type: Conference Paper / Full Text
  • Volume: 1
  • City: Baku
  • Country: Azerbaijan
  • Page Numbers: pp.287-292
  • Yıldız Technical University Affiliated: Yes


Longos (flooded) forests refer to forests where the water level is at or near the surface of the soil and the area is periodically or rarely covered with shallow water. Submerged lakes in the flooded forests will cause changes in water levels. Increasing and decreasing water levels on the forest floor is the cause of the rich biodiversity. Igneada Longos Forest, which is one of the rare flooded forests in Turkey, was selected as the study area. Logos Forests, located in the Igneada town of Demirköy district of Kırklareli province, was certified as a National Park in 2007 with the decision of the Council of Ministers. In this study, an analysis of Turkey's Igneada Longos Forests' long-term changes in the last ten years was carried out using Landsat 7 (ETM +) and Landsat-8 (OLI) satellite images. Owing to the greenness of the tree leaves in the spring and summer seasons of the year, the satellite images were selected among the cloudless images gathered in the April and August months of all the monitored years. The Google Earth Engine (GEE) platform, which is available in the cloud environment, was developed by Google, and has been highly preferred in Remote Sensing (RS) studies in recent years, especially for mapping large areas, examining changes that occurred in the past, and making future predictions. GEE, a platform that works with JavaScript and Python coding languages, can easily and simultaneously access all archives of Landsat satellite images, and process petabyte-scale data. The detection of the change on the Igneada Flooded Forests was performed by time series analysis using the Normalized Difference Vegetation Index (NDVI) with the JavaScript coding language on the GEE platform. According to the time series analysis, it is detected that there has been no significant change in forest areas in the last ten years.