METHODS FOR MONITORING LOCALITIES BASED ON REMOTE SENSING IMAGES. CASE STUDY: DUMBRAVITA, TIMIS COUNTY PUBLISHEDE. ANGHEL, M. NICOLA, D. VUCULESCU, M. V. HERBEI, F. SALA Banat University of Agricultural Sciences and Veterinary Medicine "King Michael I of Romania" from Timisoara email@example.com
The study aimed to analyze and characterize an urban area based on satellite imagery. UAT Dumbravita, Timis County, Romania, was studied under the aspect of the variation of NDBI and NDVI indices. It was considered a period of four years, 2017 - 2020, for the study, and as the period of the year the summer season was taken into account. Satellite scenes, Landsat 8, were used, taken in July - August during the study period. Based on spectral information and established formulas, NDBI and NDVI indices were calculated. Data sets of 21011 were analyzed for each index calculated and year of study. The series of values of the two indices studied (NDBI, NDVI) presented statistical distributions of histogram type - normal fit. The ANOVA test evaluated and confirmed the data safety and the presence of the variance in the data series (F>Fcrit, p<0.001). According to the Diversity profile, NDVI presented a higher variation in 2020 and a lower one in 2018. Intermediate values were recorded for 2017 and 2019. The variation of NDVI index values in relation to NDBI during the study period was described by 2nd order polynomial equations for 2017 and 2018 in statistical safety conditions (R2 = 0.729, p <0.001 , F = 28287 for 2017; R2 = 0.773, p <0.001 F = 35695 for 2018). In the conditions of 2019 and 2020, the NDVI variation in relation to NDBI was best described by linear equations, in conditions of statistical safety (R2 = 0.716, p <0.001, F = 53038 for 2019; R2 = 0.798, p < 0.001, F = 83229 for the year 2020). The general analysis over the study period, mean values of NDBI and NDVI indices, led to a spline model, which most appropriately described, and in statistical safety, the NDVI variation relative to NDBI.
NDBI, NDVI, monitoring, spline model, periurban area