Radar ve optik görüntüler kullanarak çeltik bitkisi yaprak alan indeks kestirim

Rice is one of the most important crops as it is the staple food for more than 3.5 billion people worldwide. As a result, a number of studies have been conducted in order to improve the production and productivity of rice. One of the focus areas of the studies is the estimation of rice variables by making use of satellite images as the variables can be used as indicators of the growth and productivity of the crop. Leaf Area Index (LAI), a dimensionless index defined as the leaf area per unit ground area, is one of the most crucial biophysical variables used for monitoring the growth and productivity of crops. However, undertaking field measurement of LAI over a large area is not feasible and, thus, the use of satellite data to rapidly and accurately compute LAI values has been considered as the most convenient way. The objective for this study was to evaluate the performance of rice LAI estimation using satellite-based radar and optical images compared with in-situ measurements. In-situ LAI measurements were collected with CI-110 from six paddy rice fields located in Ipsala region (Yenikarpus farm), which is one of the major rice producing areas in Turkey. Paddy rice crop has three growth stages: vegetative stage (germination to panicle initiation), reproductive stage (panicle initiation to flowering) and ripening stage (flowering to mature grain). The CI-110 based field measurement was carried out during the reproductive stage of the paddy rice crop. The satellite images used as data sources for estimating the LAI of the rice crop in the study area were captured by two Synthetic Aperture Radar (SAR) sensors named Sentinel-1 and TerraSAR-X and an optical sensor known as Landsat-8 during the reproductive stage of paddy rice crop. Both Sentinel-1 and TerraSAR-X operate in all-weather conditions. However, Sentinel-1 operates in C-band with 5.405 GHz frequency, 0-100 MHz bandwidth and VV and VH polarization, whereas TerraSAR-X operates in X-band with center frequency of 9.65 GHz, bandwidth of 300 MHz and VV and VH polarization. Once the SAR images were collected, all images were pre-processed to account for geometric distortions due to acquisition geometry of the imagery, and backscattering coefficients were calculated. The backscattering coefficient values were then used as inputs into the Water Cloud Model to determine LAI. The reliability of this technique of determining LAI was investigated by comparing the estimated LAI results with the values of the in-situ LAI measurements by taking into consideration the correlation coefficients (R) as performance evaluation criteria. With regards to the optical sensors (Landsat-8 images), the Normalized Difference Vegetation Index (NDVI), Modified Normalized Difference Water Index (NDWI), Green Normalized Difference Vegetation Index (GNDVI), Gitelson and Merzlyak Index (GMI), Simple Ratio Index (SR) and Zygielbaum Water Stress Index (ZWSI) were calculated for each pixel of the 6 paddy-rice fields. Here again, correlation coefficient was used to investigate the presence of relationships between the in-situ LAI values and the above mentioned indices generated from the Landsat-8 image. The results of the study showed that, in general, the indices generated from the Landsat-8 image were found to be highly correlated to the in-situ LAI than the values of the backscattering coefficients determined from the Sentinel-1 and TerraSAR-X images with various polarizations. In particular, the SR index was found to be the best index ( R = 0.91) related to the in-situ LAI values followed by the NDVI, NDWI and GNDVI indices. In conclusion, based on the results of this study, it could be stated that Landsat-8 images could be used as a reliable data source to acquire reliable LAI for paddy rice crop in the study area much better than SAR images. Here again, it should be noted that only the reproductive stage (not time series based analyses) of paddy rice crop was considered to meet the objective of this study.