Geri Dön

Using satellite imagery toevaluate the impacts of oilpipeline on vegetation

Başlık çevirisi mevcut değil.

  1. Tez No: 757442
  2. Yazar: MUSTAFA SEYİS
  3. Danışmanlar: DR. M.T MARSHALL, DR. T.A GROEN
  4. Tez Türü: Yüksek Lisans
  5. Konular: Jeoloji Mühendisliği, Geological Engineering
  6. Anahtar Kelimeler: Belirtilmemiş.
  7. Yıl: 2021
  8. Dil: İngilizce
  9. Üniversite: University of Twente
  10. Enstitü: Yurtdışı Enstitü
  11. Ana Bilim Dalı: Belirtilmemiş.
  12. Bilim Dalı: Belirtilmemiş.
  13. Sayfa Sayısı: 46

Özet

Özet yok.

Özet (Çeviri)

Oil pipelines have various effects on vegetation both during and after construction. Oil pipeline construction activities and their impacts over time pose a threat to the sustainability of plant species and other biodiversity conservation, especially as it passes through protected or adjacent areas (Castell et al., 1994). These impacts from the pipeline vary according to different land cover types and elevations. This study aims to evaluate the magnitude of change and the duration of the recovery process in forest and grass fields, depending on elevation, using the LandTrendr algorithm and Landsat images after pipeline construction. For this study, a part of the Baku Tbilisi Ceyhan (BTC) pipeline remaining in Georgia was selected. After the construction activities covering the years 2003-2005, the change in the vegetation was studied using the Normalized Difference Vegetation Index (NDVI) with the help of remote sensing. The magnitude of the disturbance in the vegetation experienced in this period and the time required for the recovery of the vegetation afterwards were determined. First, preprocessing was performed for data preparation, which included cloud masking, harmonization, and measurement of surface reflectance to ensure continuity over time. In the second phase, LandTrendr segmentation parameters were established to detect changes in which time series algorithm was run, temporal interruptions, abrupt changes and temporal segmentation were performed. After determining the spatial and temporal changes through LandTrendr, the magnitude of the change in our sample areas, forests and grasslands, and the duration of vegetation recovery were evaluated using 500 samples (pixel values). We also created a regression model based on three variables to understand the impact of elevation, distance to pipeline, and different land cover classes on the magnitude of changes and recovery period. Finally, we evaluated our model using K-fold cross validation and interpreting the spatial temporal changes of land cover due to pipeline disturbance. According to the disturbance result, the changes in NDVI increase with altitude, and the regression result shows that forest and grass land cover, distance to pipeline, and elevation can explain 36.8% (RMSE=0.067) of the changes in NDVI. Also, NDVI variation in grass increases while forest decreases with altitude. The regression results in forest and grass were R2=0.374 (RMSE=0.062) and R2=0.403 (RMSE=0.066), respectively. For the duration of recovery period, the change in NDVI increases with elevation, and the regression result shows that forest and grass land covers, distance to pipeline, and elevation can explain 6.5% (RMSE=4.425) of the changes in NDVI. Also, the recovery period in forest is longer than in grass, and the regression results in forest and grass show R2 = 0.0344 (RMSE=4.07), R2 = 0.091 (RMSE=3.39), respectively. As a result, K-fold cross validation was run for the disturbance and recovery period, showing R2=0.415 (RMSE=0.057) for the disturbance and R2=0.286 (RMSE=4.080) for the recovery period. Therefore, this study shows that LandTrendr algorithm is suitable to estimate the magnitude of change and duration of recovery in vegetation greenness after pipeline disturbance.

Benzer Tezler

  1. Türkiye'deki dağlık havzalarda uygulanan swat modeli ile iklim değişikliğinin incelenmesi

    Evaluating climate change impacts using swat in a mountainous region of Turkey

    İSMAİL BİLAL PEKER

    Yüksek Lisans

    Türkçe

    Türkçe

    2020

    İnşaat MühendisliğiEskişehir Teknik Üniversitesi

    İnşaat Mühendisliği Ana Bilim Dalı

    DR. ÖĞR. ÜYESİ ALİ ARDA ŞORMAN

  2. Determination of the impacts of marine farms on marine ecosystems by using remote sensing: Ildırı Bay

    Balık çiftliklerinin denizel ekosisteme etkilerinin uzaktan algılama ile tespiti: Ildırı Körfezi

    FETHİ BENGİL

    Yüksek Lisans

    İngilizce

    İngilizce

    2011

    Deniz BilimleriDokuz Eylül Üniversitesi

    Deniz Bilimleri ve Teknolojisi Ana Bilim Dalı

    YRD. DOÇ. DR. KEMAL CAN BİZSEL

  3. Yüksek çözünürlüklü uydu verileri kullanılarak şehir alanlarının incelenmesi

    Investigation of city areas using high-resolution satellite data

    RAZİYE HALE TOPALOĞLU

    Doktora

    Türkçe

    Türkçe

    2022

    Jeodezi ve Fotogrametriİstanbul Teknik Üniversitesi

    Geomatik Mühendisliği Ana Bilim Dalı

    PROF. DR. ELİF SERTEL

  4. Road geometry extraction with fusion of low resolution satellite imagery and GPS trajectory using deep learning methods

    Düşük çözünürlüklü uydu görüntüleri ve GPS rotaları birleştirilmesi ve derin öğrenme yöntemleri kullanılarak yol geometrilerinin tespiti

    NECİP ENES GENGEÇ

    Doktora

    İngilizce

    İngilizce

    2024

    Jeodezi ve Fotogrametriİstanbul Teknik Üniversitesi

    Geomatik Mühendisliği Ana Bilim Dalı

    PROF. DR. ERGİN TARI

  5. Yüksek çözünürlüklü uydu görüntüleri ile İstanbul pendik ilçesinin alansal değişiminin incelenmesi

    Analysis of field changes in pendi̇k province of İstanbul with high resolution satellite images

    HALİL İHSAN ALTUNDAŞ

    Yüksek Lisans

    Türkçe

    Türkçe

    2018

    Jeodezi ve Fotogrametriİstanbul Teknik Üniversitesi

    İletişim Sistemleri Ana Bilim Dalı

    DOÇ. DR. ÇİĞDEM GÖKSEL