Akademik Veri Yönetim Sistemi
UZALCBS 2022 Sempozyumu, Ankara, Türkiye, 17 - 19 Kasım 2022, ss.1-6
Today, the risks posed by climate change and other global factors (population growth, migrations, etc.) have made it a great necessity
to continue agricultural production in a sustainable manner. In this context, many countries and international organizations develop
agricultural policies to support agricultural production. For this purpose, various institutions directly or indirectly support
agricultural production in our country. In the planning of agricultural supports, the main basis is the data of agricultural production.
It is very important for decision makers to record these data correctly. In our country, information on agricultural activities is stored
on the Farmer Registration System (FRS). Although the FRS provides many advantages in the collection of agricultural data, there
are sometimes inconsistencies between the FRS records and land use, due to the fact that the producers may make incorrect or
incomplete declarations. Although these inconsistencies can be eliminated with on-site detection, these checks that need to be made by
technical personnel are costly and time-consuming.
Satellite images have been used for a long time in many ground observation studies, especially in monitoring changes in land cover
types. In recent years, the increase in access to medium/high resolution satellite images has made the use of these images widespread
in studies aimed at monitoring agricultural areas. The aim of this study is to monitor the production activities on the basis of parcels
between April 2022 and July 2022 in the agricultural parcels where sugar beet production is planned, and to determine the
inconsistencies between the producer's declaration and the land reality. In this context, 206 agricultural parcels located in the Polatlı
district of Ankara were selected as the test area. In order to monitor agricultural production activities, normalized difference vegetation
index (NDVI) maps were produced using Sentinel-2 images collected on different dates. By overlapping these produced maps with the
parcel areas, the parcels without agricultural production were determined automatically. Our results show that the proposed model
can also be applied in larger areas and will save cost and time by limiting on-site inspection studies.