CHARACTERIZATION OF VEGETATION AND PREDICTION OF BIOMASS PRODUCTION IN SILAGE CORN CROP BASED ON SATELLITE IMAGES

. The study used methods based on remote sensing to analyze the dynamics of a silage corn crop and predict biomass production. The corn crop, the Micado hybrid, was organized within the Didactic and Experimental Station of the "King Mihai I" University of Life Sciences in Timisoara, under the specific conditions of the 2021-2022 agricultural year. The soil in the crop plot was of chernozem type, and the crop was in a non-irrigated system. To obtain the satellite data, the Sentinel 2 system was used. To characterize the plot and the crop of silage corn, 14 sets of images were taken, in the interval March - July 2022. From the analysis of the images, the spectral data were obtained, and based on the consecrated formulas MSAVI, NDMI, NDVI and NBR indexes were calculated. The ANOVA test confirmed the reliability of the data and the presence of variance in the data set (F>Fcrit, p<0.001, for Alpha=0.001). The level of correlations identified between the calculated indices was between r=0.967 (NBR with NDVI) and r=0.996 (NDVI with MSAVI). The variation of the NBR index, in relation to the other indices, was described by linear equations, in terms of statistical safety (R 2 =0.965, p<0.001, in relation to MSAVI; R 2 =0.989, p<0.001 in relation to NDMI; R 2 =0.934, p<0.001 in relation to NDVI). The variation of indices in relation to time (t, days) during the study period was described by polynomial equations of the 3rd degree, in terms of statistical safety (R 2 =0.964 for MSAVI, R 2 =0.934 for NDMI, R 2 =0.941 for NDVI, and R 2 =0.961 for NBR). Regression analysis facilitated obtaining some equations, as models for predicting biomass production, under conditions of statistical safety. Based on the RMSEP parameter, it was found that the most reliable level of production prediction was obtained in the case of MSAVI and NDMI indices (RMSEP=0.05923).


INTRODUCTION
Techniques based on satellite images (different satellite systems), aerial (utility aircraft, UAV drones), or terrestrial (machines and agricultural machinery with high-resolution cameras, portable devices, various devices) facilitate the study of the areas considered based on the spectral information from the captured images (Błaszczyket al., 2022).
The accuracy of the information is closely correlated with the technique used (cameras, digital capturers, spectral bands, etc.), influencing factors and working methods (Zheng et al., 2018;Sánchez et al., 2020), and some studies have evaluated different image acquisition conditions in order to increase accuracy or find correction factors Rasti et al., 2022).
Agricultural practice benefits from the advantages of techniques based on remote sensing for evaluating the spatial and temporal variability of land and agricultural crops Duan et al., 2022), for land evidence, the classification and study of crops (Moran, 2010;Govedarica et al., 2015).
Based on remote sensing, agricultural crops were evaluated in relation to soil conditions and different soil improvement works (Yuzugullu et al., 2020;Bertici et al., 2022;El Behairy et al., 2022;San Bautista et al., 2022).
The influence of climatic conditions on agricultural crops have been studied and evaluated based on remote sensing and a series of estimation and prediction models have been found (Huang et al., 2019;Beyene et al., 2022).
The response of agricultural crops to fertilization was also evaluated by techniques based on remote sensing and different evolutionary changes were highlighted at the level of soil and vegetation indices under the influence of different categories of fertilizing resources (Blaes et al., 2016;Argento et al., 2021;Bertici et al, 2022).
The evaluation of the production of agricultural crops is one of the important aspects that has been studied a lot based on image analysis and techniques based on remote sensing (Karlson et al., 2020;Ali et al., 2022;Ma et al., 2022).
The purpose of this study was to predict the production of silage corn, under non-irrigated crop conditions and specific climatic conditions for the year 2022, through the analysis of satellite images and the calculation of specific representative indices in relation to the soil and the corn crop.

MATERIAL AND METHODS
The study used techniques based on remote sensing to analyze the dynamic evolution of a silage corn crop and the estimation of biomass production under the specific conditions of the 2021-2022 agricultural year.
The study took place within the Didactic and Experimental Station of ULS "King Michael I" from Timisoara, Timis County, Romania.
The corn crop was placed on a cambic chernoziom type soil, of medium fertility, in a non-irrigated cultivation system.
The biological material was represented by the corn hybrid Micado, a hybrid suitable for silage culture. The satellite images were taken in the Sentinel 2 system (Planet Team). (1) The silage corn crop was harvested at the beginning of August (August 4) being strongly affected by the climatic conditions specific to the year 2022.
In relation to the purpose of the study, the dynamics of the indexes calculated during the study period were evaluated in relation to time (t). Also, the level of correlation between the calculated indices was evaluated, as well as the interdependence relationship between certain indices. Biomass production was estimated (regression analysis) in relation to various calculated indices.
As statistical safety parameters, in relation to each analysis made, the correlation coefficient, the regression coefficient, the p parameter, and the RMSEP parameter were used in relation (5).
The PAST software and the statistical calculation module in EXCEL were used to analyze and process the obtained data, regarding the values of the calculated indices, the period of time and the production of biomass in silage corn culture (Hammer et al, 2001), Wolfram Alpha (2020).

RESULTS AND DISCUSSIONS
The retrieval of satellite images in order to evaluate the plot and the corn crop, the Micado hybrid, were taken between March 22, 2022 and July 23, 2022, in the Sentinel 2 satellite system (Planet Team).
The time interval during the study period and the values of the MSAVI, NDMI, NDVI and NBR indices calculated on the basis of satellite images are presented in table 1. To evaluate the general safety of the data, and the presence of the variant in the data set, the ANOVA test was used, and the values are presented in table 2.
Under the conditions of the agricultural year 2021 -2022, respectively the climatic conditions of the study interval, the satellite images accurately captured the state of vegetation of the corn crop, and very strong correlations were identified between the values of the calculated indices, in conditions of statistical safety (p< 0.001). Thus, the level of correlations was between r=0.967 (NBR with NDVI) and r=0.996 (NDVI with MSAVI).  Alpha=0.001 The NBR index is important in relation to the burning capacity, and in relation to this, it can be useful to express biomass production.
The variation of the calculated index values was analyzed in relation to time (t, days) during the study period. The variation of the MSAVI index was described by equation (9), under conditions of R2=0.964, p<0.001, F=89.663. The variation of the NDMI index in relation to time (t, days) was described by equation (10) under conditions of R 2 =0.934, p<0.001, F=47.18. The variation of the NDVI index in relation to time (t, days) was described by equation (11) under conditions of R 2 =0.941, p<0.001, F=53.92. The variation of the NBR index in relation to time (t, days) during the study period was described by equation (12) where: BPSM -Biomass Production in silage maize; x, y -calculated indices; a, b, c, d, e, f -coefficients of the equation (13)    The accuracy of predicting the production of agricultural crops based on remote sensing is variable, in relation to the images used, the method of analysis, calculated indices, methods and algorithms used, etc. (Muruganantham et al., 2022;Pham et al., 2022).
In general, high levels of statistical accuracy were communicated in estimating production based on remote sensing techniques, due to the fact that the vegetation reflects well through the status of the plants, the