WoS İndeksli Yayınlar Koleksiyonu / WoS Indexed Publications Collection

Permanent URI for this collectionhttps://hdl.handle.net/20.500.14517/18

Browse

Browsing WoS İndeksli Yayınlar Koleksiyonu / WoS Indexed Publications Collection by Author "Aksu, Oktay"

Filter results by typing the first few letters
Now showing 1 - 3 of 3
  • Thumbnail Image
    Article
    Citation Count: 9
    Considerations on the land management system approach in Turkey by the experiences of a case study
    (Taylor & Francis Ltd, 2019) Aksu, Oktay; Iban, Muzaffer Can; Geomatik Mühendisliği / Geomatics Engineering
    In Turkey, applications and activities for the use, acquisition and arrangements of land tenure rights are executed under the authorisation and responsibilities of a variety of public institutions in terms of concerned legislative framework. In this paper, a case study has been implemented for Istanbul not only to evaluate the quality and usability of existing data from different institutions but also to get conclusions and recommendations for future land management works. In the context of this sample work of preparing an inventory for land management and land use in 1:25 000 scale and related synthesis and pre-feasibility analyses; the inventory praxis for urban and rural areas have been undertaken separately. Layer overlay and synthesis of the data in the database are the essential parts of analysing and extracting the outputs supporting land management activities for the purpose of taking correct decisions. In that manner, an overview map of Istanbul's natural and settlement areas has been extracted.
  • Thumbnail Image
    Article
    Citation Count: 12
    A model for big spatial rural data infrastructure in Turkey: Sensor-driven and integrative approach
    (Elsevier Sci Ltd, 2020) Iban, Muzaffer Can; Aksu, Oktay; Geomatik Mühendisliği / Geomatics Engineering
    A Spatial Data Infrastructure (SDI) enables the effective spatial data flow between providers and users for their prospective land use analyses. The need for an SDI providing soil and land use inventories is crucial in order to optimize sustainable management of agricultural, meadow and forest lands. In an SDI where datasets are static, it is not possible to make quick decisions about land use. Therefore, SDIs must be enhanced with online data flow and the capabilities to store big volumes of data. This necessity brings the concepts of the Internet of Things (IoT) and Big Data (BD) into the discussion. Turkey needs to establish an SDI to monitor and manage its rural lands. Even though Turkish decision-makers and scientists have constructed a solid national SDI standardization, a conceptual model for rural areas has not been developed yet. In accordance with the international agreements, this model should adopt the INSPIRE Directive and Land Parcel Identification System (LPIS) standards. In order to manage rural lands in Turkey, there are several legislations which characterize the land use planning, land classification and restrictions. Especially, the Soil Protection and Land Use Law (SPLUL) enforces to use a lot and a variety of land use parameters that should be available in a big rural SDI. Moreover, this model should be enhanced with IoT, which enables to use of smart sensors to collect data for monitoring natural occurrences and other parameters that may help to classify lands. This study focuses on a conceptual model of a Turkish big rural SDI design that combines the sensor usage and attribute datasets for all sorts of rural lands. The article initially reviews Turkish rural reforms, current enterprises to a national SDI and sensor-driven agricultural monitoring. The suggested model integrates rural land use types, such as agricultural lands, meadowlands and forest lands. During the design process, available data sets and current legislation for Turkish rural lands are taken into consideration. This schema is associated with food security databases (organic and good farming practices), non-agricultural land use applications and local/European subsidies in order to monitor the agricultural parcels on which these practices are implemented. To provide a standard visualization of this conceptual schema, the Unified Modeling Language (UML) class diagrams are used and a supplementary data dictionary is prepared to make clear definitions of the attributes, data types and code lists used in the model. This conceptual model supports the LPIS, ISO 19156 International Standard (Geographic Information: Observations and Measurements) catalogue and INSPIRE data theme specifications due to the fact that Turkey is negotiating the accession to EU; however, it also provides a local understanding that enables to manage rural lands holistically for sustainable development goals. It suggests an expansion for the sensor variety of Turkish agricultural monitoring project (TARBIL) and it specifies a rural theme for Turkish National SDI (TUCBS).
  • Thumbnail Image
    Article
    Citation Count: 0
    SHAP-Driven Explainable Artificial Intelligence Framework for Wildfire Susceptibility Mapping Using MODIS Active Fire Pixels: An In-Depth Interpretation of Contributing Factors in Izmir, Türkiye
    (Mdpi, 2024) Iban, Muzaffer Can; Aksu, Oktay; Geomatik Mühendisliği / Geomatics Engineering
    Wildfire susceptibility maps play a crucial role in preemptively identifying regions at risk of future fires and informing decisions related to wildfire management, thereby aiding in mitigating the risks and potential damage posed by wildfires. This study employs eXplainable Artificial Intelligence (XAI) techniques, particularly SHapley Additive exPlanations (SHAP), to map wildfire susceptibility in Izmir Province, T & uuml;rkiye. Incorporating fifteen conditioning factors spanning topography, climate, anthropogenic influences, and vegetation characteristics, machine learning (ML) models (Random Forest, XGBoost, LightGBM) were used to predict wildfire-prone areas using freely available active fire pixel data (MODIS Active Fire Collection 6 MCD14ML product). The evaluation of the trained ML models showed that the Random Forest (RF) model outperformed XGBoost and LightGBM, achieving the highest test accuracy (95.6%). All of the classifiers demonstrated a strong predictive performance, but RF excelled in sensitivity, specificity, precision, and F-1 score, making it the preferred model for generating a wildfire susceptibility map and conducting a SHAP analysis. Unlike prevailing approaches focusing solely on global feature importance, this study fills a critical gap by employing a SHAP summary and dependence plots to comprehensively assess each factor's contribution, enhancing the explainability and reliability of the results. The analysis reveals clear associations between factors such as wind speed, temperature, NDVI, slope, and distance to villages with increased fire susceptibility, while rainfall and distance to streams exhibit nuanced effects. The spatial distribution of the wildfire susceptibility classes highlights critical areas, particularly in flat and coastal regions near settlements and agricultural lands, emphasizing the need for enhanced awareness and preventive measures. These insights inform targeted fire management strategies, highlighting the importance of tailored interventions like firebreaks and vegetation management. However, challenges remain, including ensuring the selected factors' adequacy across diverse regions, addressing potential biases from resampling spatially varied data, and refining the model for broader applicability.