Estimating net primary productivity of forest ecosystems over Turkey using remote sensing approach

Thesis Type: Doctorate

Institution Of The Thesis: Orta Doğu Teknik Üniversitesi, Faculty of Engineering, Department of Civil Engineering, Turkey

Approval Date: 2018




Understanding the fluctuations in carbon balance and global warming with respect to the global climate change and creating solutions, has become one of the most important topics in ecological studies especially during last decades. These changes, in particular for the terrestrial ecosystems, can be monitored using gross primary productivity (GPP), its derivative net primary productivity (NPP) (the subtraction of autotrophicrespirationfromGPP)andnetecosystemproductivity(NEP)(subtraction of both plant respiration and autotrophic use from GPP) as key components, which are directly affected from climate change. Despite their importance for the ecological researches, the difficulties in field measurements forced the scientists to find new methods such as statistical methods and process based modelling techniques to estimate these quantities. The aims of this study are (i) to evaluate a widely used global model’s outputs for Turkey’s forest ecosystems using field measurements, (ii) to improve the accuracy of a global model using local datasets and (iii) to create a new modelling approach to increase the accuracy of NPP estimation over forest ecosystems. Results show that the global model has a significant correlation with field data of Turkey (R2 = 0.34). However, it still lacks reflecting actual conditions over the area. Theusageoflocaldataslightlyimprovestheaccuracyofthemodel(R2 =0.35). Inthisstudy,anewmodellingapproachfortheoptimumtemperaturewasalsoimplemented. The results show that, the distribution of the optimum temperature values is more meaningful (the value of each pixel with respect to its neighbours). Moreover, the model accuracy was increased from 35% to 43% and from 39% to 43% for two different APAR (Absorbed Photosynthetically Active Radiation) estimation methods which are discussed in the thesis. The analysis showed that only 51% accuracy can be achieved using the field data. The potential reasons are discussed in the study.