REMOTE SENSING MONITORING OF COMMERCIAL CROPS

                                                     !"#$%&'()&*& *)++,&$ -    .!/0/  10 2()31450"(#4.67   (%82  2  9.&()2( 1  9 4 -  :;9"<! :()(,.!/0/  10  REMOTE SENSING MONITORING OF COMMERCIAL CROPS Cassiane Jrayj de Melo Nelson Mario Victoria Bariani 1 TABLE OF CONTENTS Chapter 1  Chapter 2  Chapter 3  Chapter 4  2 CHAPTER 1 - FOLLOW-UP AND MONITORING OF THE SOYBEAN CROP USING NDVI IN URUGUAIANA-RS / GRANJA SÃO RAMÃO Juliano Mott Urquiza Cassiane Jrayj de Melo Nelson Mario Victoria Bariani SUMMARY Remote sensing is a tool that makes it possible to monitor the growth and development of multiple plant species, thus enhancing decision making for the management of different crops. This report aims to discuss the activities developed in the Integrated Interdisciplinary Laboratory (LABii) of Unipampa, Campus Itaqui - Rio Grande do Sul, during the supervised internship, in follow-up and monitoring of the development and management of soybean crop in the field and through remote sensing techniques. A 67-hectare commercial plantation was monitored. Normalized Difference Vegetation Indices (NDVI) composed of the spectral bands (B8A and B04) of the MSI sensor from the Sentinel 2 satellite were extracted during the crop development cycle, using the EOS/Land Viewer platform and comparing its development with the monitoring of variables observed in the field. Through this process, it was possible to see that the NDVI values are close to 0 when there is no presence of vegetation and increase according to the soybean development, reaching maximum values in the range of 0.9 until the reproductive stage 3 R5, and receding after the R6 stage, with the onset of senescence, indicating the end of the cycle and physiological maturity phase of the crop. The occurrence of the La Nina phenomenon and consequently the lack of regular rainfall in the state of Rio Grande do Sul, was associated with anomalies identified in the NDVI, as well as the delay in the establishment of the crop. Keywords: vegetation indices, Sentinel 2, Glycine max 4 3. OBJECTIVES 3.1 General Objective The present work aims to track in the field and remotely monitor a commercial crop. 3.2 Specific Goals 1To monitor the vegetative development cycle of the soy crop in the municipality of Uruguaiana, state of Rio Grande do Sul; 2To remotely monitor with Sentinel2A satellite images the vegetative development cycle of a commercial crop with the soybean crop in place; 3Evaluate color compositions in the visible range and vegetation indices obtained from Sentinel2A satellite images on the LandViewer platform; 4Compare the results from item 3 with the information from the field monitoring. 4. DEVELOPED ACTIVITIES 4.1 Internship Location The supervised internship was carried out remotely, due to the Sars-Cov-2 pandemic, in the Integrated Interdisciplinary Laboratory of the Federal University of the Pampa, located on the Itaqui-RS campus, and was coordinated by professors Nelson Mario Victoria 11 Bariani and Cassiane Jrayj de Melo and by the Education Technician Roberto Felice, with the participation of other professors such as Professor Michele da Silva Santos, supervisor of this internship. According to the text of the project entitled "Labii - Integrated Interdisciplinary Laboratory: Curricularly Integrated Service Provision Proposal - SIPPEE Registration: 05.075.20" : "The Integrated Interdisciplinary Laboratory arises as a response to the need to contribute to the analysis and search for solutions on some regional problems, whose complexity leads us to consider variables that correspond to multiple areas of knowledge. Some important challenges that appear in the region covered by Unipampa are linked to agricultural activity and its related industries and services (BARIANI2018). The observed trend is a need to know the agricultural, environmental, economic, and social variables with increasing precision. Planted agricultural area, degraded areas, quality of crops and natural environments, water and soil quality, effective impact of current or future agricultural managements, advantages, disadvantages and difficulties of using the new technological possibilities available, cost/benefit ratio, social changes underway due to financial/industrial pressure, historical perspective of the problems, are some examples of the multiple aspects and multiple variables that can be studied to understand the existing situations (BARIANI&VICTORIA, 2019a, b, c, d). The ultimate goal of all human activity in the region (and on the planet) is the global achievement of a pleasant quality of life for all, 12 where food, education, access to housing, decent and healthy work, and access to a well cared for environment allow the joy of living to flow. Although the great advances in the availability of food, transportation, housing, hygiene, communication, machinery and technology that have revolutionized the quality of life and work that is possible today are undeniable, the procedures used so far in anthropic activities still do not sufficiently meet these parameters of quality and sustainability. Notably in urban agglomerations and agricultural production is noticeable an important process of deterioration in the quality of the environment, presenting worrying trends (MEUS&VICTORIA, 2018). These trends need to be known in order to organize effective actions to improve the productive and social processes." For these reasons, the laboratory aims to train students in techniques for measuring environmental, agricultural, industrial or social variables, and as results for society studies, reports, articles or opinions about regions or processes based on field data and satellite sensors. In short, the laboratory conducts studies on basic sanitation, agriculture, food production, health, research on topics of social interest in the academic sphere provided by Unipampa. In the area of remote sensing, LABii has several publications on agricultural and environmental monitoring, some of which appear in the bibliography. 13 4.2 Study Area The study area is located in the Rural Property Granja São Ramão, in the municipality of Uruguaiana-RS (Figure 1), in the Western Frontier of the State of Rio Grande do Sul. According to Koppen's climate classification, the local climate is of type Cfa, subtropical without defined dry season with hot summers. Figure 1. Location of the study area in Uruguaiana - RS. The area chosen for tracking and monitoring the soybean crop on the property was 67 hectares, and in the 2020/2021 crop year irrigated rice was grown. 14 4.2 Soil Collection, Analysis and Preparation First, soil analysis was performed in the study area for fertilizer recommendation. Fifteen soil sub-samples were collected in a zig-zag pattern, using a cutting spade (Figure 2). After this, the samples were homogenized to form a single sample and taken to the laboratory of the University of Cruz Alta (UNICRUZ). With the results in hand, it was possible to determine the amount of fertilizer to be used to meet their needs and provide the best for the soy crop. In September 2021 the soil was prepared, incorporating the haystack of the rice crop into the soil and later passed the remaplan for its leveling (Figure 3). In November 2021 the ridges and furrows were made for soybean irrigation (Figure 4). A 1 hectare area was left without the furrows and ridges to get a real idea of the behavior of the soybean crop in this area (Figure 5). Figure 2. Collecting soil sample for analysis. 15 Figure 3: Soil preparation for soybean cultivation. Figure 4: Preparation of the ridges and furrows for soybean cultivation. 16 Figure 5: Area left without ridges and furrows 4.3 Sowing With the area ready, the sowing could be done in November, within the recommended window for its degree of maturity being 5 - 6 for our microregion 101, the choice of soybean seed was according to its maturity for the region being its degree of maturity: 6.4 and its growth habit indeterminate. The crop deployment area has 67 hectares of sown area for the 17 2021/2022 harvest with the soybean cultivar, Monsoy 6410 IPRO. Sowing started on November 26, 2021, the base fertilization used for the cultivar was 270 kg/ha (MAP): 11-52-00 (NP 0 -K252 O) on a 33 hectare area; (Bio Atyvo Fós): 300 kg/ha of 00-16-00 (NP205K2O) on a 34 hectare area (KCl): 58 % K2O hauled on a 30 hectare area at a dosage of 50 kg/ha. The sowing was done transversally to the furrows so as not to lose sowing lines and to gain in plant population, as we are sowing in lowlands, 18 seeds per meter were used, thus avoiding loss of seeds due to non-germination, the area's desiccation was done after the plants were in the V3 and V4 stages, with an area of 2 hectares not desiccated due to rice crops next to this area. 4.4 La Nina Phenomenon According to the BBC (2021) the climate in the year 2021 was a La Nina year, that is, it is an oceanic-atmospheric phenomenon in which the surface waters of the Equatorial Pacific Ocean cool down exceptionally. The NOAA (National Oceanic and Atmospheric Administration), classifies this year's La Nina event as moderate with SST anomalies (Sea Surface Temperature) reaching -1°C, usually lasting 5 to 12 months. Due to this phenomenon, the state of Rio Grande do Sul has been facing a great drought, which has been causing losses to agriculture. Data from the Secretariat of Agriculture, Livestock and Rural Development of the state indicate that the drought can generate a loss of 39% of crops. This year's La Nina is less intense than last year's, but because 18 it is the second year in a row, its effects are being magnified. In addition, the summer is more humid in the Northeast than in the Center-South region. In other words, it is normal to have more drought in the South than in the North. And there is still a third and final factor that also potentiates this picture of extremities in the country: the waters of the Atlantic Ocean are cold in the South and hot in the North, which means that the Western Frontier Region suffers a lot from drought, which is an extremely important factor for the establishment of soybean cultivation, affecting the decision making about the most appropriate management. 4.5 Sentinel Satellite Images and Image Processing The collection of satellite images was performed through the Land Viewer platform, which makes the scenes available for free at 10-day intervals. The satellite chosen for use was Sentinel 2 L2A, which used images without cloud cover from the sowing date (November 2021) until the end date of the stage (March 2022), aiming to correlate the soybean development stages with the NDVI generated. The Land Viewer platform also provides free of charge the calculation of the Sentinel 2 NDVI, using for this the bands 18 B8A and B04, through the formula NDVI: (B8A-B04) / (B8A+B04). For this, images of the areas under study were selected according to their fertilization and satellite images coinciding with the development stage of the crop and the NDVI was collected. Figure 6: Screen for search selections of the EOS/Land Viewer 19 platform with Sentinel 2 satellite scenes. Figure 7. Sentinel satellite image (A) and application of the NDVI index (B) of the study region on 11/24/2021. Uruguaiana-RS 20 to obtain a satisfactory production, We can also observe through Figure 10 that the shades of color more orange are those that have values of 0.2 to 0.4 corresponding to 14.32 hectares, which we can say that will not produce the expected due to lack of rainfall, since the color in light green tone represents values of 0.4 to 0.6 corresponds to 22.76 hectares of soybeans culture in a satisfactory development with a denser leaf area. In Figure 11 the area was sown on December 3, 2021 and fertilized with Bravya Bio Atyvo Fós, the values of sparse vegetation reach 82.15% and 16.43% of moderate vegetation, we can say that the time of sowing and the lack of moisture in the sowing damaged much in the establishment of culture, we can see through Figure 11 that its desiccation was done very late, due to nearby rice fields, thus affecting the emergence of soybean culture. The regions that are in shades of red with pixel values in the range of 0.2 are those that indicate low quantity of photosynthetically active vegetation, they represent the initial stages of crop development, these values are observed from sowing until the end of emergence. In Figure 12 we can see that even using potassium chloride in the haul was not possible to improve the NDVI indexes of the soybean crop, as values close to 0.2 to 0.4 of sparse vegetation were maintained in an area of 28.28 hectares, and of moderate vegetation there were 6.08 hectares, proving that the lack of rain in this specific area, even with the correct fertilization was very detrimental to its establishment in comparison with Figure 10. 27 The NDVI values found in (Figure 13) correspond to 6 months that the area under study was monitored, from its soil preparation until the establishment of culture, we can see that in the months of September the NDVI values are in the range of 0.2 if maintained until the month of November when the emergence of the soybean crop begins, already in the month of January we have the peak of the culture with indices of 0.4 to 0.5 already in stage of development R1 when it most needs rainfall and due to lack of it is that in February these indices fall to 0.2 again, thus having the loss of its development due to lack of water and thus the flower abortion and loss of leaf area, reflecting the low rates of NDVI. 28 6. CONCLUDING REMARKS The internship carried out at LABII allowed us to follow up on a tool that is still little explored by most of the producers. The study proposal allowed the verification of a new tool for soybean crop management and crop monitoring, even at a distance and for free. With this tool you can observe and study several possibilities, helping even in decision making, comparisons with past crops, observations of the dams, and get an idea of how much water was used in other years. Based on the vegetation indices it was possible to observe through the values, the behavior of the soybean crop, especially in certain areas. In which, were previously used for rice culture. Thus, areas were mapped that in the other crop suffered late decisions to be adjusted as necessary so that this does not happen again in the coming years. In the development of the soy crop it was possible to observe that the area of the Bravya Bio Atyvo Fós fertilizer was the most damaged area, as shown in the vegetation indexes, but not in relation to fertilization, but in relation to the time of sowing and delays in decision making, but throughout the sown area the main factor was the lack of rain in the periods that the soy crop most needs for its establishment. 29 7. REFERENCES ALMEIDA, T. S.; FONTANA, D. C.; MARTORANO, L. G.; BERGAMASCHI, H. Vegetation indices for soybean crop under different water conditions and soil management system. In: Simpósio Brasileiro de Sensoriamento Remoto, 12. (SBSR), 2005, Goiânia. Annals... São José dos Campos: INPE, 2005. p. 17-24. CD-ROM, Online. ISBN 85-17-00018-8. Available at: <http://urlib.net/ltid.inpe.br/sbsr/2004/11.18.17.02>. Accessed on: 23 Feb. 2022 BARIANI, C. J. M. V.; BARIANI, N. M. V. ; MARQUES NETO, G. C. M. . REMOTE SENSING APPLIED TO MONITORING THE PRINCIPAL PHENOLOGICAL STAGES OF IRRIGATED RICE IN SOUTH BRAZIL. In: IV Inovagri International Meeting, 2017, Fortaleza. Proceedings of the IV Inovagri International Meeting - 2017. Fortaleza: INOVAGRI/ESALQUSP/ABID/UFRB/INCTEI/INCTSal/INSTITUTO FUTURE, 2017. BARIANI, C. J. M. V.; BARIANI, N. M. V. ; NETO, G. C. M. . SPECTRAL CHARACTERISTICS OF IRRIGATED RICE CULTIVARS DURING THE DEVELOPMENT CYCLE IN SOUTH BRAZIL. In: IV Inovagri International Meeting, 2017, Fortaleza. Annals of the IV Inovagri International Meeting - 2017. Fortaleza: INOVAGRI/ESALQUSP/ABID/UFRB/INCTEI/INCTSal/INSTITUTO FUTURE, 2017. BARIANI, C. J. M. V.; BARIANI, N. M. V. ; NETO, G. C. M. . MONITORING OF SPATIAL VARIABILITY DURING THE PERIOD OF DEVELOPMENT OF IRRIGATED RICE CROP. In: IV Inovagri International Meeting, 2017, Fortaleza. Proceedings of the IV Inovagri International Meeting - 2017. Fortaleza: INOVAGRI/ESALQUSP/ABID/UFRB/INCTEI/INCTSal/INSTITUTO FUTURE, 2017. BARIANI, C.J.M.V.; CARESANI, J. R. F. ; VICTORIA, N.M. . MULTITEMPORAL ANALYSIS OF RICE USING SURFACE NDVI. In: XI Brazilian Congress of Agroinformatics, 2017, 30 Campinas. XI SBIAgro, 2017. v. 11. BARIANI, C.J.M.V.; FELICE, R. D. ; VICTORIA, N. M. . Remote sensing of phenological stages of irrigated agricultural crops using NDVI. In: Anais.... 8° Salão Internacional de Ensino, Pesquisa e ExtensãoSIEPE, Uruguaiana-RS. v. 8. 2016. BARIANI, C.J.M.V.; VICTORIA, N.M. . SPECTROTEMPORAL CHARACTERIZATION OF CULTIVARS: IRRIGATED RICE. 1st. ed. Latvia, European Union: New Academic Editions, 2019. 53p . BARIANI, C.J.M.V.; VICTORIA, N.M. . SPECTRAL BEHAVIOR OF DIFFERENT IRRIGATED RICE CULTIVARS DURING THEIR DEVELOPMENT CYCLE. REMOTE SENSING STUDIES APPLIED TO COMMERCIAL CROPS. 1°ed.Latvia, European Union: New Academic Editions, 2018, v. , p. 45-54. BARIANI, C.J.M.V.; VICTORIA, N.M. . MAPPING THE MAIN PHENOLOGICAL STAGES OF THE IRGA 409 CULTIVAR USING REMOTE SENSING TECHNIQUES. STUDIES IN REMOTE SENSING APPLIED TO COMMERCIAL CROPS. 1°ed.Latvia, European Union: New Academic Editions, 2018, v. , p. 72-78. BARIANI, C.J.M.V.; VICTORIA, N.M. . SPECTROTEMPORAL IDENTIFICATION OF IRRIGATED RICE CULTIVARS USING SATELLITE IMAGES. REMOTE SENSING STUDIES APPLIED TO COMMERCIAL CROPS. 1°ed.Latvia, European Union: New Academic Editions, 2018, v. , p. 79-86. BARIANI, C.J.M.V.; VICTORIA, N.M. . MAPPING AND CALIBRATION OF NDVI FOR THE ESTIMATION OF PHENOLOGY OF THE CULTIVAR IRGA 424 RI. REMOTE SENSING STUDIES APPLIED TO COMMERCIAL CROPS. 1°ed.Latvia, European Union: New Academic Editions, 2018, v. , p. 64-71. BARIANI, C.J.M.V.; VICTORIA, N.M. . MONITORING OF PHENOLOGICAL STAGES OF SOYBEAN BY REMOTE SENSING. STUDIES IN REMOTE SENSING APPLIED TO 31 COMMERCIAL CROPS. 1°ed.Latvia, European Union: New Academic Editions, 2018, v. , p. 14-23. BARIANI, C.J.M.V.; VICTORIA, N.M. . REMOTE SENSING TECHNIQUES FOR MONITORING THE MAIN PHENOLOGICAL STAGES OF RICE. REMOTE SENSING STUDIES APPLIED TO COMMERCIAL CROPS. 1°ed.Latvia, European Union: New Academic Editions, 2018, v. , p. 56-62. BARIANI, C.J.M.V.; VICTORIA, N.M. . SPATIAL VARIABILITY DURING THE DEVELOPMENT CYCLE OF IRRIGATED RICE. STUDIES IN REMOTE SENSING APPLIED TO COMMERCIAL CROPS. 1°ed.Latvia, European Union: New Academic Editions, 2018, v. , p. 37-44. BARIANI, C.J.M.V.; VICTORIA, N.M. . SENSITIVITY OF NDVI FOR REMOTE SENSING MONITORING OF SOYBEAN STAGES UNDER IRRIGATION IN SOUTHERN BRAZIL. In: XI Brazilian Congress of Agroinformatics, 2017, Campinas. XI SBIAgro, 2017. v. 11. BARIANI, C.J.M.V.; VICTORIA, N.M. . REMOTE SENSING FOR THE STUDY OF VEGETATION DYNAMICS IN IRRIGATED AREAS. REMOTE SENSING STUDIES APPLIED TO COMMERCIAL CROPS. 1°ed.Latvia, European Union: New Academic Editions, 2018, v. , p. 24-36. BARIANI, C.J.M.V.; VICTORIA, N.M. . RESOLUTION OF NDVI FOR PHENOLOGY MONITORING OF IRRIGATED RICE IN SOUTHWESTERN BRAZIL. In: XI Brazilian Congress of Agroinformatics, 2017, Campinas. XI SBIAgro, 2017. v. 11. BARIANI, C.J.M.V.; VICTORIA, N.M. ; MARQUES NETO, G. C. ; CARESANI, J. R. F. . SPECTRO-TEMPORAL CHARACTERIZATION OF IRRIGATED RICE CULTIVARS USING LANDSAT IMAGES. In: X CONGRESSO BRASILEIRO DE ARROZ IRRIGADO, 2017, GRAMADO. ELECTRONIC PROCEEDINGS X CBAI, 2017. 32 BARIANI, C.J.M.V.; VICTORIA, N.M. ; MARQUES NETO, G. C. ; ESTEVO, L. F. ; GELAIN, N. . NDVI ASSOCIATED WITH THE MAPPING OF THE MAIN PHENOLOGICAL STAGES OF THE CULTIVAR IRGA 409 IN ITAQUI, RS. In: X CONGRESSO BRASILEIRO DE ARROZ IRRIGADO, 2017, GRAMADO. ANNAIS ELETRÔNICOS X CBAI, 2017. BARIANI, C.J.M.V.; VICTORIA, N.M. ; MARQUES NETO, G. C. . CALIBRATION OF NDVI FOR MAPPING THE PHENOLOGY OF THE IRGA 424 RI CULTIVAR IN ITAQUI, RS. In: X CONGRESSO BRASILEIRO DE ARROZ IRRIGADO, 2017, GRAMADO. ANNAIS ELETRÔNICOS X CBAI, 2017. CONAB. Acompanhamento da safra brasileira de grãos. v.8, n.10, p.1-110, 2021. Available at: http://www.conab.gov.br. Accessed 02/22/2022. CONAB - COMPANHIA NACIONAL DE ABASTECIMENTO. Acompanhamento da Safra Brasileira de Grãos, Brasília, DF, v. 9, harvest 2021/22, n. 5 fifth survey, February. 2022. CRUSIOL, L. G. T. et al. NDVI of development stages of soybean BRS 284 under field conditions. In: Embrapa Soja-Article in conference proceedings (ALICE). In: JORNADA ACADÊMICA DA EMBRAPA SOJA, 8., 2013, Londrina. Expanded abstracts... Londrina: Embrapa Soja, 2013. p. 8791.(Embrapa Soja. Documentos, 339)., 2013. DRUSCH, M. et al. Sentinel-2: ESA's optical high-resolution mission for GMES operational services. Remote Sensing of Environment, v. 120, p. 25-36, 2012. EMBRAPA. Soja em números - Safra 2020/21. Londrina: Brazilian Agricultural Research Corporation, 2022. Available at: https://www.embrapa.br/soja/cultivos/soja1/dados-economicos>. Accessed on March 09, 2022. ESQUERDO, J. C. D. M.; ANTUNES, J. F. G.; BALDWIN, D. G.; EMERY, W. J.; ZULLO JÚNIOR, J. An automatic system for AVHRR land surface product generation. International Journal of 33 Remote Sensing, v.27, p.3925-3942, 2006 FERNANDES, J. L. Monitoramento da cultura de cana-de-açúcar no estado de São Paulo por meio de imagens spot Vegetation e dados meteorológicos. 2009. 114 f. Thesis (Master in Agricultural Engineering) School of Agricultural Engineering - UNICAMP. Campinas, São Paulo, 2009. FRAMPTON, W. J.; DASH, J.; WATMOUGH, G.; MILTOM, E.J. Evaluating the capabilities of Sentinel-2 for quantitative estimation of biophysical variables in vegetation. ISPRS Journal of Photogrammetry and Remote Sensing, v. 82, p. p. 83-92, 2013. HAGOLLE, O. et al. SPOT-4 (Take 5): simulation of Sentinel-2 time series on 45 large sites. Remote Sensing, v. 7, n. 9, p. 12242-12264, 2015. JENSEN, J. R. Introductory digital image processing: a remote sensing perspective. 2nd ed. Upper Saddle River: Prentice-Hall, 1996. LILLESAND, T. M.; KIEFER, R. W. Remote sensing and image interpretation. 3.ed. New York: John Wiley, 750p. 1994. MORAES, R. S. ; BARIANI, C.J.M.V. ; BREDOW, E. C. ; VICTORIA, N. M. ; RIBEIRO, L. S. . SPATIAL VARIATION ANALYSIS OF NDVI IN COMMERCIAL IRRIGATED RICE FARMS. In: SALÃO INTERNACIONAL DE ENSINO, PESQUISA E EXTENSÃO, 2018, SANTANA DO LIVRAMENTO. Proceedings of the International Salon of Teaching, Research and Extension, 2018. NOVO, E. L. M. Sensoriamento remoto: princípios e aplicações. Edgar Blucher, São Paulo, 1989. SEGL, K. et al. S2eteS: An end-to-end modeling tool for the simulation of Sentinel-2 image products. IEEE Transactions on Geoscience and Remote Sensing, v. 53, n. 10, p. 5560-5571, 2015. VICTORIA, N.M. ; BARIANI, C.J.M.V. ; KICH, G. . PHOTOINTERPRETATION OF SATELLITE IMAGES IN RICE CROPS: A TREND IN PRECISION AGRICULTURE. 1st. ed. Latvia, European Union: New Academic Editions, 2019a. 52p . 34 VICTORIA, N.M. ; BARIANI, C.J.M.V. ; KICH, G. . INTERPRETATION OF SATELLITE IMAGES DURING THE DEVELOPMENT CYCLE OF IRRIGATED RICE IN COMMERCIAL CROPS. In: X CONGRESSO BRASILEIRO DE ARROZ IRRIGADO, 2017, GRAMADO. ELECTRONIC PROCEEDINGS X CBAI, 2017. VICTORIA, N.M. ; BARIANI, C.J.M.V. ; SILVA, G. K. . REMOTE SENSING APPLIED TO RICE CROP MANAGEMENT SYSTEMS: IN SEARCH OF SUSTAINABLE PRODUCTIVITY. 1st. ed. Latvia, European Union: New Academic Editions, 2019. 53p . VICTORIA, N.M. ; BARIANI, C.J.M.V. ; SILVA, G. K. . REMOTE SENSING APPLIED TO RICE CROP MANAGEMENT SYSTEMS: IN SEARCH OF SUSTAINABLE PRODUCTIVITY. 1st. ed. Latvia, European Union: New Academic Editions, 2019b. 53p . VICTORIA, N.M. ; BARIANI, C.J.M.V. . INTERPRETATION OF LANDSAT IMAGES DURING THE RICE DEVELOPMENT CYCLE. REMOTE SENSING STUDIES APPLIED TO COMMERCIAL CROPS. 1°ed.Latvia, European Union: New Academic Editions, 2018, v. , p. 7-13. 35 CHAPTER 2 - MONITORING THE PHENOLOGICAL STAGES OF SOYBEAN CULTIVATION USING NDVI IN ALEGRETE-RS Leandro Noetzold Ratts Cassiane Jrayj de Melo Nelson Mario Victoria Bariani SUMMARY The report presents and discusses the activities developed in the Integrated Interdisciplinary Laboratory of Unipampa, Campus ItaquiRio Grande do Sul, during the supervised internship, in monitoring the phenological stages of the soybean crop by NDVI. A commercial plantation of 137.90 hectares with 5 cultivars of different maturity groups was monitored throughout the crop cycle. Normalized Difference Vegetation Indices (NDVI) composed of the spectral bands (B8A and B04) of the MSI sensor from the Sentinel 2 satellite were extracted for all phenological stages of the crop, using the EOS/Land Viewer platform and compared with "in loco" monitoring of the dates of occurrence of the crop development stages. From the results, it was possible to identify that values that start close to 0 when there is no presence of vegetation and increase according to the development of soybeans reaching maximum values in the range of 0.9 until the R5 stage and after the R6 stage these values decline with the onset of senescence indicating the end of the cycle and physiological maturity phase of the crop. Keywords: vegetation indices, Sentinel 2, Glycine max. 36 the crop behavior during its development. The monitored area has 137.90 hectares and is being cultivated in the 2020/2021 crop with 5 different Soybean cultivars, being them the 55I57 RSF IPRO on 21.56 hectares, DM 5958 RSF IPRO on 32.16 hectares, 59I60 RSF IPRO on 38.09 hectares, 64I6I RSF IPRO on 20.31 hectares and DM 66I68 RSF IPRO on 21.07 hectares. The sowing was started on October 28, 2020, the base fertilization used for all cultivars was the fertilizer BASA 04.28.08 10 Ca;0.5 Mg;7.2S - 250 kg/ha and FIDAGRAN 16Ca;1.5Mg;10S;0.1B - 50 kg/ha - Fixed Rate, in the V3 stage was used 160 kg/ha of 00.00.60. 3.2 Sentinel Satellite Images and Image Processing The collection of satellite images was carried out through the Land Viewer platform, which makes the scenes available for free at 10-day intervals. The satellite chosen for use was the Sentinel 2 L2A, which used images without cloud cover from the sowing date (October 2020) until the end date of the soybean cycle (April 2021), aiming to correlate the stages of corn development with the NDVI generated. The Land Viewer platform also provides free of charge the calculation of the Sentinel 2 NDVI, using for this the bands 18 B8A and B04, through the formula NDVI: (B8A-B04) / (B8A+B04). For this, fifteen (15) satellite images coinciding with the stages of development followed were selected and the NDVI collected. Figure 2 - Screen for search selections of the EOS/Land Viewer 43 platform with scenes from the Sentinel 2 satellite. Figure 3 - Image captured by the Sentinel satellite (A) and application of the NDVI index (B) of the study region on the date of 20/04/2021. Alegrete-RS. 2020 44 3.3 Field data collection The field data survey was conducted during technical visits accompanied by the Agronomist responsible for the property, Lucas Bastos. Photographs were collected during periodic technical visits to the property, to monitor the development of the crop, for comparison with satellite images. Figure 4 - Photo report of the different development stages of the Soybean crop and approximate dates of occurrence. 45 4. RESULTS AND DISCUSSIONS The NDVI generated by the EOS/Land Viewer platform allowed the mapping of NDVI during the phenological stages of the soybean crop. By looking at the sequence of images in order by dates from sowing from October 28, 2020 to the end of the soybean cycle in late March/mid April, we can observe the phenological storm of the soybean crop. Figure 5 - NDVI of Soybean crop on different image capture dates in rural property of Alegrete-RS in the 2020/2021 crop 46 47 Figure 6 - NDVI values along the development stages of the Soybean cultivars followed. The NDVI values presented in figures 5 and 6 vary from 0.20 to 0.90. The values close to 1 represent denser vegetation, it is possible to observe that with the advancement of plant development the NDVI index. The red-shaded regions with pixel values in the range 0.2 are those that indicate low amounts of photosynthetically active vegetation, 48 representing the initial stages of crop development; in the monitored area, these values are observed from planting until the end of emergence. From the V2 stage (11/17/2020) to the VN stage (12/29/2020) an evolution in the NDVI index from 0.3 to 0.6 is observed because there is more leaf area Crusiol et.al (2013). The reproductive period begins in late January, in the R1 stage we find NDVI of 0.60 - 0.70 the maximum value is found in the R3 stage until the R5 stage in the range of 0.90. After the maximum NDVI found in the R5 stage, from the R6 stage on, the senescence of the crop begins. Figure 6 shows that this period occurs on different dates, according to the maturity group of the cultivars analyzed, from the beginning of the R6 stage until the R7 stage, the NDVI values decrease from 0.9 to 0.4. These results are in agreement with those obtained by Bariani; Felice; Victoria (2016) who in their study found that NDVI values between 0.1-0.2 represent the period of beginning of vegetative growth, while rapid growth has values between 0.2-0.7. For the intermediate period, represented by the end of vegetative growth, flowering and beginning of grains, NDVI values of 0.7 to 0.9 were obtained. And between the onset of senescence to physiological maturity values between 0.9-0.4 were observed. 49 4. CONCLUDING REMARKS The internship carried out at LABII allowed us to follow up on a tool that is still little explored by most of the producers. The study proposal allowed the verification of a new tool for the management of the corn crop and the monitoring of the crop, even at a distance and for free. It can be observed that the tool is useful for the study of the dynamics of vegetation and the identification of phenological stages for soybean cultivation, where it can be seen that the values that start close to 0 when there is no vegetation increase according to the soybean development until the R6 stage, and then these values decrease with the onset of senescence, indicating the end of the cycle and the physiological maturity stage. 50 5. REFERENCES ALMEIDA, T. S.; FONTANA, D. C.; MARTORANO, L. G.; BERGAMASCHI, H. Vegetation indices for soybean crop under different water conditions and soil management system. In: Simpósio Brasileiro de Sensoriamento Remoto, 12. (SBSR), 2005, Goiânia. Annals... São José dos Campos: INPE, 2005. p. 17-24. CD-ROM, Online. ISBN 85-17-00018-8. Available at: <http://urlib.net/ltid.inpe.br/sbsr/2004/11.18.17.02>. Accessed April 25, 2021. BARIANI, C.J.M.V.; FELICE, R. D. ; VICTORIA, N. M. . Remote sensing of stages phenological of irrigated agricultural crops using NDVI. In: Anais.... 8° Salão Internacional de Ensino, Pesquisa e ExtensãoSIEPE, Uruguaiana-RS. v. 8. 2016 CRUSIOL, L. G. T. et al. NDVI of development stages of soybean BRS 284 under field conditions. In: Embrapa Soja-Article in conference proceedings (ALICE). In: JORNADA ACADÊMICA DA EMBRAPA SOJA, 8., 2013, Londrina. Expanded abstracts... Londrina: Embrapa Soja, 2013. p. 8791.(Embrapa Soja. Documentos, 339)., 2013. DRUSCH, M. et al. Sentinel-2: ESA's optical high-resolution mission for GMES operational services. Remote Sensing of Environment, v. 120, p. 25-36, 2012. EMBRAPA. Soja em números - Safra 2019/20. Londrina: Empresa Brasileira de Pesquisa Agropecuária, 2020. Available at: https://www.embrapa.br/soja/cultivos/soja1/dados-economicos>. Accessed on October 02, 2020. ESQUERDO, J. C. D. M.; ANTUNES, J. F. G.; BALDWIN, D. G.; EMERY, W. J.; ZULLO JÚNIOR, J. An automatic system for AVHRR land surface product generation. International Journal of Remote Sensing, v.27, p.3925-3942, 2006 FERNANDES, J. L. Monitoramento da cultura de cana-de-açúcar no estado de São Paulo por meio de imagens spot Vegetation e dados 51 meteorológicos. 2009. 114 f. Thesis (Master in Agricultural Engineering) School of Agricultural Engineering - UNICAMP. Campinas, São Paulo, 2009. FRAMPTON, W. J.; DASH, J.; WATMOUGH, G.; MILTOM, E.J. Evaluating the capabilities of Sentinel-2 for quantitative estimation of biophysical variables in vegetation. ISPRS Journal of Photogrammetry and Remote Sensing, v. 82, p. p. 8392, 2013. HAGOLLE, O. et al. SPOT-4 (Take 5): simulation of Sentinel-2 time series on 45 large sites. Remote Sensing, v. 7, n. 9, p. 12242-12264, 2015. JENSEN, J. R. Introductory digital image processing: a remote sensing perspective. 2nd ed. Upper Saddle River: Prentice-Hall, 1996. JUNGES, A. H.; FONTANA, D. C. Evaluation of the development of cereal crops winter in Rio Grande do Sul through temporal profiles of the vegetation index by normalized difference. Ciência Rural, v.39, p.1 15, 2009. LILLESAND, T. M.; KIEFER, R. W. Remote sensing and image interpretation. 3.ed. New York: John Wiley, 750p. 1994. NOVO, E. L. M. Remote Sensing: principles and applications. Edgar Blucher, São Paulo, Brazil. Paulo, 1989. SEGL, K. et al. S2eteS: An end-to-end modeling tool for the simulation of Sentinel-2 image products. IEEE Transactions on Geoscience and Remote Sensing, v. 53, n. 10, p. 5560-5571, 2015. SHIMABUKURO, Y. E. Índice de Vegetação e Modelo Linear de Mistura Espectral no Monitoramento da região do Pantanal. Pesquisa Agropecuária Brasileira, (33): 1729-1737, 1998. 52 Figure 1. RGB image of the region under study 2.3 NDVI There are numerous vegetation indices extracted from satellite images. One of the most widely used and known indices is NDVI, normalized difference vegetation index (ROUSE et al., 1974). The normalization is done by the following equation: Where pnir is the reflectance value in the near infrared range; and pred is the reflectance value in the visible red range. According to Jensen (2011), NDVI is important because: i) Seasonal and interannual changes in vegetation development and activity can be monitored; ii) The ratio reduces many forms of multiplicative noise (differences in solar illumination, cloud shadows, some atmospheric attenuation, some topographic variations) present in 59 multiple bands of images from multiple dates. NDVI can be used to monitor droughts, monitor and predict crop production, help predict dangerous fire zones, and map desert enhancement. NDVI is a standardized vegetation index that allows you to generate an image demonstrating relative biomass. Chlorophyll absorption in the red band and large vegetation reflectance in the NIR are used to calculate the NDVI. Figure 2: NDVI image of the region under study 2.4 Agriculture (SWIR1, Red8, Blue) This combination of bands is useful for monitoring agricultural crops. In the image (figure 3), strong green represents vigorous, healthy vegetation while non-agricultural vegetation such as trees appear in dark green. Coniferous forests are dark green while deciduous forest is strong green. Poor vegetation and areas without 60 vegetation are brown and mauve (figure 3). Figure 3: Image of agricultural crop monitoring. 2.5 Healthy Vegetation (Red8, SWIR1, Blue) Healthy vegetation appears in red, brown, orange, and yellow. Soils can appear brown and green. Urban features are cyan, gray, and white. Areas of deep blue represent recently mowed areas and reddish areas show new vegetation growth, probably sporadic grass areas. Clear, deep water is very black and if the water has sediment it becomes clearer. For vegetation studies, the addition of the Mid-IR band increases sensitivity to detect various stages of flora growth or stress. 61 Figure 4: Image representing healthy vegetation 2.6 NDWI NDWI makes use of reflected near-infrared radiation and visible green light to eliminate the presence of vegetation and soil. It is suggested that NDWI be used by researchers for estimations of water bodies to gauge their degree of turbulence. Figure 5. NDWI image that eliminates the presence of vegetation and soil and check its turbulence degree 62 2.7 Vegetation Analysis (SWIR1, Red8, Red) This combination provides by using a lot of color information and contrast. Healthy, light green vegetation and soils are mauve. This combination of bands is useful for vegetation studies and is used for firewood and pest infestation management. Figure 6: Analysis image of healthy vegetation or pest infestation 3. MATERIAL AND METHODS This research used the Land Viewer platform and its potential as a study tool, with the purpose of collecting images and monitoring agricultural plots that presented: i) anomalies; ii) climate damage; and iii) productivity, in a rural property in the municipality of Barra do Quaraí, Western Frontier of the State of Rio Grande do Sul. The EOS Land Viewer platform is a real time image processing and analysis software, available at <https://eos.com/pt/products/landviewer/>. The service was created and is used for observing satellite images in order to search and obtain 63 important and valuable information from the data found, so as to work with everyday agribusiness situations. It is worth mentioning that one of the characteristics of this platform are the medium and high resolution images, the latter at a cost, and the possibility of viewing and obtaining information from different vegetation indices and compositions of different bands and sensors, which helps a lot at the time of analysis. Another attractive feature is the opportunity to save time and cut costs. After all, Land Viewer instantly focuses on specific areas of interest to facilitate the search for results, all in a shorter time. Therefore, this new technology combines practicality and faster results, facilitating processes that previously took more time and were costly. In this study it can be observed that vegetation indices are used for comparison and correlation with productivity and anomalies found in the analyzed plots. Considering that the NDVI is the most used vegetation index for monitoring agricultural areas, it was used through an automated processing within the platform and its information was exported in an Excel spreadsheet and the subsequent analysis and interpretation of the received data. The general step-by-step description of this process is described below. The first stage of interaction with the software begins by researching the study city, in this case the city of Barra do Quaraí (Figure 7). 64 Figure 7. First stage with Land Viewer software, searching for the study city. In the second step, satellite images of the region under study for the 18/19 and 19/20 harvests were sought (Figure 8). Figure 8. Second stage with the Land Viewer software, surveying the study area, in the crops under analysis. In the next step a combination of channels is selected and used to obtain various surface features, such as vegetation indices, coverage types, agricultural areas, etc., as already presented in the theoretical framework. Then you get data on the spatial distribution of the index values in percentages or absolute units that are later exported to an .xlsx 65 table. To quantify the analysis of the NDVI time curves, extracted from the Land Viewer .xlsx table, the USGS methodology described in Hargrove et al., (2010) was applied, where the minima and maxima on each side of the graph are found and the 20% and 80% of the rapid development and senescence stages are calculated. The intervals of the curves are analyzed as described by FAO, shown in Figure 10. Figure 9. Methodology for analyzing the behavior of the NDVI curves. Source: Adapted from USGS. The nomenclature used for each section of the curves is used by FAO and was adapted from Allen and Pereira (2009), illustrated in Figure 10. 66 Figure 10. FAO nomenclature for each section of the NDVI curves. Source: Adapted from Allen and Pereira (2009). 4. RESULTS AND DISCUSSION Satellite images in RGB format, as well as NDVIs, in the 2018/2019 and 2019/2020 harvests are presented below in figures 11 and 12. In the 2018/2019 harvest, the property under study had a decrease in productivity due to the flooding of the Quaraí River, it can be seen in the images of figure 11 that the agricultural plot under study was sown in October 2018 and was harvested in April 2019. Also in the same year, due to high rainfall in the month of January, the Quaraí River overflowed, leaving its normal bed, and its waters advanced up to the highlighted plot. Figure 11 shows that the images, of the highlighted field, collected in the Land Viewer platform were 67 affected by the flooding of the Quaraí River, thus generating significant losses in rice farming as it was already in its reproductive stage. It is noteworthy that in the subsequent months after the river returned to its normal course the affected area continued its development, with one part having been harvested in April and the other part, which was affected by the flood, was left intact because the area was insured against climatic events. 68 5 FINAL CONSIDERATIONS With the conclusion of this work, it is necessary to revisit the objectives initially proposed. The general objective was to "Evaluate the potential of the Land Viewer tool for monitoring agricultural areas during the production process". This objective was achieved, since the two agricultural areas evaluated in the different crops (2018/2019 and 2019/2020) could be visualized through the satellite images, and the NDVI values were extracted pixel by pixel and analyzed in a time curve of the average NDVI values throughout the cycle of the rice crop. Next, the specific objectives are resumed, as well as the main results and comments related to each of them, along with the suggestions and recommendations identified. - 1st Specific objective: "To monitor in spatial and temporal scale agricultural plots of a rural property located in Barra do Quaraí, in the harvests 2018/2019 and 2019/2020". The monitoring and spatial and temporal analysis was done through the preparation of space-time maps that show the development of the crop throughout its cycle, in the two harvests, as shown in Figures 11 and 12. It can be observed that the space-time maps developed were assembled in order to present the images in RGB composition, as well as with the NDVI vegetation index, for each date analyzed. With this it was possible to observe the development of the crop throughout the cycle, as well as the increase of green coloration, indicating the increase of NDVI values and consequent increase of plant biomass, and the decrease of NDVI 75 values, indicated by the yellowish and orange coloration on the maps (Figures 11 and Figure 12), in the final stages of senescence and harvest. The map and space-time of the 2018/2019 harvest also allowed us to verify the advance of the flood in the analyzed plot, as well as the damage and effects caused by it (Figure 11). Thus, the results obtained through temporal analysis indicated the potential of the tool to analyze the development of the rice crop, as well as the areas affected by the flooding of the Quaraí River. - 2nd specific objective: "To evaluate the dynamics of flooding within the property''". This objective was achieved by means of the timespace maps shown in Figures Figures 11 and Figure 12, as well as by the NDVI values that remained lower throughout the crop cycle, as can be seen in the graphs in Figure 13 and Table 1. - 3rd specific objective: "To evaluate the difference in productivity of the areas affected by the floods and other unaffected areas, associating vegetation indices with the productivity achieved in each plot". This objective was achieved, since the crops showed a productivity difference of 4,310 kg/ha, and the area affected by the flood climatic event, 2018/2019 crop, obtained a productivity of 5,747 kg/ha while in the crop that did not suffer extreme climatic events, 2019/2020, productivities around 10,057 kg/ha were reached. Thus, there is evidence that the Land Viewer platform is a tool with high potential for monitoring the temporal space of satellite images, as well as NDVI curves, in a fast and efficient way. 76 6 REFERENCES ALLEN, R. G. et al. FAO Irrigation and drainage paper No. 56. In: FAO Food and Agriculture Organization of the United Nations. Rome: [s.n.], 1998. p. 2640. ALLEN, R. G.; PEREIRA, L. S. Estimating crop coefficients from fraction of ground cover and height. Irrig. Sci., 28, 2009. 17-34. BARIANI, C.J.M.V.; FELICE, R. D. ; VICTORIA, N. M. . Remote sensing of phenological stages of irrigated agricultural crops using NDVI. In: Anais.... 8° Salão Internacional de Ensino, Pesquisa e ExtensãoSIEPE, Uruguaiana-RS. v. 8. 2016. CAP BRAZIL, CONFEDERATION OF AGRICULTURE AND CATTLE RANCHING OF BRAZIL (CNA). Agribusiness GDP reaches 26.6% participation in Brazilian GDP in 2020. 2020. Available at: <https://www. cnabrasil .org.br/boletins/pib-do-agronegocio-alcancaparticipacao-de-26-6-no-pib-brasileiro-em-2020> Accessed on: 14 Jul. 2021. DENG, J.; WANG, K.; HONG, Y. Q. J. Spatio-temporal dynamics and evolution of land use change and landscape pattern in response to rapid urbanization. Landscape and Urban Planning, p.187-198, 2009. FERNANDES, J. L. Monitoramento da cultura de cana-de-açúcar no estado de São Paulo por meio de imagens spot Vegetation e dados 77 meteorológicos. 2009. 114 f. Thesis (Master in Agricultural Engineering) School of Agricultural Engineering - UNICAMP. Campinas, São Paulo, 2009. HARGROVE, W. W. et al. Toward a National Early Warning System for Forest Disturbances Using Remotely Sensed Land Surface Phenology. USGS, 2010. Available at: <https://www.geobabble.org/~hnw/first/ncdc/slideshow.html>. Accessed June 13, 2020. JENSEN, J. R. Remote sensing of the environment: a perspective on terrestrial resources. [S.l.]: Parenthesis, v. 2, 2011. 598 p. Authorized translation. LIU, T.; YANG, X. Mapping vegetation in an urban area with stratified classification and multiple end-member spectral mixture analysis. Remote Sensing of Environment, p. 251-264, 2013. LILLESAND, T. M.; KIEFER, R. W. Remote sensing and image interpretation. 3.ed. New York: John Wiley, p. 750, 1994. ROUSE, J. W. et al. Monitoring Vegetation Systems in the Great Plains with ERTS. Third Earth Resources Technology Satellite-1 Symposium. Greenbelt: NASA. 1974. 78 CHAPTER 4 - MONITORING THE PHENOLOGICAL STAGES OF IRRIGATED CORN CROP USING NDVI IN ITAQUI-RS Daniel Pinto Malgarim Cassiane Jrayj de Melo Nelson Mario Victoria Bariani SUMMARY Remote sensing is a tool capable of monitoring the growth and development of plant species, assisting in decision making for the management of different crops, in a continuous and updated way. The present report presents and discusses the activities developed in the Integrated Interdisciplinary Laboratory of Unipampa, Campus ItaquiRio Grande do Sul, during supervised internship, in monitoring the development stages of the corn crop irrigated by central pivot with NDVI index. A field with cultivar BG 7318 (109 ha-1) was monitored from sowing to flowering (August to November) in the 2020/2021 crop. Nine (9) normalized difference vegetation indices (NDVI) composed of the spectral bands (B8A and B04) of the MSI sensor of the Sentinel 2 satellite were extracted at 10-day intervals, made available for free on the EOS/Land Viewer platform and compared with "in loco" monitoring of the dates of occurrence of the crop development stages. From the results, it was possible to identify the NDVI variations during the monitored period, with indices ranging from 0.1 (sowing) to 1 (flowering). The use of nitrogen fertilization at the V4 stage promoted an increase in NDVI from 0.2 to 0.7, 30 days after application. The activities proposed in the internship provided learning about sensing techniques not yet experienced and that the use of free images can help producers and technicians quickly and accurately about what happens in the field in real time and thus adjust the decision making accurately and efficiently. Keywords: vegetation indices, Zea maysL., digital agriculture, Sentinel 2. 79 1 INTRODUCTION Corn is grown in almost all regions of the world, due to hybrid technology with higher yielding capacity, the adoption of transgenic cultivars, improvements in the ability to more efficiently withstand water deficiency, and improvements in nutrient use efficiency. According to Santi (2007) the adaptations and skills for high production are due to the advances in genetic improvement, qualification, application efficiency, and nutrient uptake, the quality of soil use and management, irrigation, and improvement in the management of agricultural resources, and the adoption of precision agriculture (PA) tools. Among PA tools, remote sensing stands out, which consists of acquiring information from an object without the need for direct contact with it, and in an agricultural environment the characteristics of vegetation make it possible to understand how their agronomic characteristics are. The use of sensors makes it possible to establish relationships between the spectral responses of the crop and the development parameters, with the possibility of using the images to detect changes in vegetation cover over time (ENCINA, 2018). One of the main vegetation indices used is the Normalized Difference Vegetation Index (NDVI), proposed by Rouse et al. (1973), which can be employed for identifying the spatial variability of plant biomass production through canopy reflectance (BREDEMEIER et al., 2013). The NDVI index is based on the contrast between the 80 maximum absorption in the red band by the chlorophyll present in the plant and the maximum reflectance in the near infrared, due to the cellular structure of the leaf (JENSEN, 2011). According to Ponzoni & Shimabukuro (2007), the NDVI value is normalized to the range of -1 to +1, which represent an indirect measure of the density of photosynthetically active leaf phytomass per unit area. And therefore, successfully used for monitoring changes in vegetation (JUNGES & FONTANA, 2009). Therefore, NDVI presents correlation with physiological and biophysical characteristics of vegetation, such as leaf area, phytomass, anomalies, evapotranspiration, productivity and water condition (CRUSIOL et al., 2012) and when one has several images throughout the development of plants, it is possible to map and identify the temporal evolution of the phenological stages of the crop of interest (BARIANI; FELICE; VICTORIA, 2016). Thus, with the possibility of using NDVI to differentiate between crop phenological stages, this report aims to describe the activities developed during the mandatory internship in the Integrated Interdisciplinary Laboratory of Unipampa-Itaqui on the identification of phenological stages and possible anomalies during the cycle of corn crop irrigated by central pivot in a study area in the city of Itaqui-RS. 2 THEORETICAL FRAMEWORK 81 2.1 Remote Sensing Developments in remote sensing (RS) have simplified data acquisition for professionals in need of geographic information (NAGARAJU, 2016). SR allows obtaining information from the objects that make up the Earth's surface, making it possible to measure and monitor biophysical characteristics of the surface, as well as anthropic activities on the ground surface (JENSEN, 2011), being important for more accurate interpretation of vegetation (ANTUNES & ROSS, 2018). The application of SR techniques enables real-time evaluation of the mapped area, and acts as a tool for analyzing landscape modification on a temporal scale (OLIVEIRA et al., 2013), due to its synoptic, multispectral, and revisit satellite characteristics (SANTOS et al., 2019). The SR has numerous satellites with sensor image capture, which represent the electromagnetic energy that interacted with the atmosphere and the earth's surface (PARANHOS FILHO et al., 2008). Currently, images are made available in open source as from Landsat and Sentinel satellites (MA et al., 2020) and allows the application of the time series of images in land cover classification ( ZANG et al., 2019). SR images are composed of the field of view of the sensor (spatial resolution), wavelength of the bands (spectral resolution), and the numerical values of the measurement of the target's radiance (radiometric resolution), in addition to the date of image capture (temporal resolution) (MENESES, 2012). 82 The response to the four elements cited is given by the incident energy which interacts with the pigments of the leaves, water, intercellular spaces and soil, being part absorbed, part reflected, measuring the portion of the incident energy that is reflected, which is called reflectance (JENSEN, 2011). Also, it can be explained by the pixels coming from the sensor images, which are digital numbers, which related to the electromagnetic radiation of the targets that reaches the sensor, are directly proportional to their reflectance (TEIXEIRA et al., 2017). Thus, monitoring agricultural crops via SR over the growing seasons identifies changes in crop management (MANABE; MELO; ROCHA, 2018), stresses and productivity (HOMMA et al., 2017), in addition to the type of cultivar/variety, physiological aspects of the crops and their phenological stages (BARIANI; BARIANI; NETO, 2017). 2.2 Sentinel 2 Satellites The Sentinel-2 satellite is part of the European Space Agency (ESA) mission developed within the framework of the European Union Copernicus program (SEGL et al., 2015) and composes a set of two satellites simultaneously in orbit (Sentinel-2A since June 2015 and Sentinel-2B since 2016), with the sensor attached to them called MSI (MultiSpectral Instrument) ) (BROLO et al., 2020). The MSI sensor is a passive optical sensor with 13 spectral bands in a 20.6° orbital field of view, high and medium spatial 83 resolution (10, 20 and 60 m) and 12-bit radiometric resolution (PEREIRA et al., 2020). The mentioned aspects ensure that Sentinel 2 achieves the necessary data for agricultural monitoring (VAN DER MEER et al., 2014). Also the sensor is the most advanced of its kind and the first optical earth observation mission of its class, as it includes three bands in the "red edge" that provides key information on vegetation status (TAQUIA, 2015), and the optical design of the MSI telescope allows for a 290 km field of view (FOV) (ESA, 2015). Sentinel-2 images have 10 m spatial resolution (visible and mid-infrared bands) and 5 (five) days temporal resolution (the two satellites together) and are freely available from ESA (European Space Agency) (PEREIRA-SANDOVAL et al., 2019 ). Sentinel 2 data has possibility of use in land, environmental, water, forest, vegetation, terrestrial carbon, natural resources and crop monitoring (ESA, 2015) and according to Frampton et al. (2013) Sentinel 2 offers has better spatial and temporal resolution compared to Landsat and SPOT images, by bands taken around the red edge, and therefore useful for checking the state of vegetation. 2.3 Vegetation Index - NDVI The variations of phenological cycles, physiology and morphology of plants, provide information about the climatic, edaphic, geological and physiological characteristics of a region 84 Figure 5 - Image captured by the Sentinel satellite (A) and application of the NDVI index (B) of the study region on the date of 04/11/2020. Itaqui-RS. 2020. 3.4 Field data collection The field data survey was carried out together with the person in charge of the property, Agronomist Engineer Pablo Chaves Rodrigues, being available the photographic survey of the personal collection of the person in charge and collected files "in loco", of the corn crop development in the studied area and the due approximate dates of occurrence (Figure 6), in order to correlate with the NDVI data (ROUSE et al., 1973), obtained later. The development stages followed the phenological scale of Ritchie et al., (2003) for corn culture. 91 Figure 6 - Photo report of the different development stages of the corn crop and approximate dates of occurrence. 92 4 DISCUSSION OF THE DATA FOUND The NDVI generated by the EOS/Land Viewer platform allowed the mapping and behavior of the NDVI during the phenological stages of the corn crop (Figure 7). The NDVI presented a temporal, from 0.1 at sowing to 1 at flowering. According to Martinko et al. (2000), time series analysis of satellite sensors, allows weekly comparisons of crop conditions. Figure 7 - NDVI of the corn crop cultivated under central pivot on different image capture dates in a rural property in the municipality of Itaqui-RS in the 2020/2021 crop. 93 In the period from germination to V4 (0.1 to 0.3) the cultivated areas show low NDVI responses, the NDVI rise point was at V5 (0.4) and indicates the beginning of accelerated plant development. The regions in shades of red with pixel values from 0.1 to 0.3 are those that indicate low quantity of photosynthetically active vegetation, representing the initial stages of crop development in the monitored area, from sowing to the V4 stage. The corn was sown in spacing of 55 cm between rows, thus the values found represent, for the most part, the residues of straw from the previous crop, since the bare soil would present NDVI close to -1. From the V5 (09/20/2020) to V10 (10/20/2020) stage the NDVI obtained a more effective response for reflectance. From this stage occurs the rapid formation of leaves due to intra-specific competition and the NDVI values responded in the range of 0.3 to 0.8. According to Magalhães & Durães (2006), from V5/V6 the growing point and the stalk are above ground level, thus the stalk begins a period of accelerated elongation. The data found corroborate with Bariani; Felice; Victoria (2016), who studying the behavior of NDVI at different 94 developmental stages of irrigated corn found NDVI between 0.1 to 0.2 for early development, while rapid growth has values between 0.2 to 0.7. When the plants reached 12 leaves until flowering (10/30/2020 to 11/04/2020) the NDVI peaked between 0.8 and 1, that is, this stage configures with the maximum reflectance of the crop. For Leivas et al. (2013) the NDVI index shows high values (close to 1) in areas where the vegetation is more vigorous and lower values in areas with dry vegetation or little vegetation. At flowering it was possible to verify NDVI polygons with values of 1 and others with values of 0.8. This detail, according to the person in charge of the property, are places of greater fertility within the area, already verified in previous fertility maps, but that it was not possible to access them. Another aspect observed was that prior to the application of nitrogen (09/10/2020) the NDVI values were at values close to 0.2, after 30 days of the fertilizer application the area presented NDVI of 0.7. This is due to N increasing leaf protein and consequent increase in leaf area of the crop, thus the plants reflected more energy and consequent increase in the index. Saiz-Fernández et al. (2015) state that the V4 stage of corn is one that deserves attention for the adequate supply of N and is of fundamental importance for defining its productive potential. The V4 stage is when the differentiation of the corn stalk occurs and therefore the increase in NDVI found infers that the crop responded to fertilization and provided a higher relative yield than where N is not 95 used. Vian et al. (2018) correlated the critical NDVI thresholds for each vegetative development stage of the corn crop when targeting different yield classes (Figure 8). Figure 8 - Critical limits of the normalized difference vegetation index (NDVI) for defining classes of productive potential of the corn crop in different phenological stages of development. The data found and correlated with Figure 5 show that the studied crop is expected to have a low/medium yield, since the average NDVI at the V8 stage was 0.6. Drum et al. (2017) verified a relationship between NDVI and stages with corn grain yield, with a coefficient of determination of 0.70 for the V9 stage. Thus, the health and nutrition of the crop by the NDVI found at flowering are within normality for the stages studied. 96 5 CONCLUSIONS The internship at LABII allowed me to follow an area of study that was not studied in depth during my undergraduate studies. The study proposal allowed for the verification of a new tool for the management of the corn crop and the monitoring of the crop, even at a distance and for free. The NDVI is a useful tool for studying the dynamics of vegetation and identifying phenological stages for corn crops irrigated by central pivot, where the sowing values correspond to values close to 0 (zero) and with an increase in the photosynthetic area there is an increase in values until close to 1 (one), which characterizes a healthy crop in full vegetative development, caused by the flowering of the crop. In addition, it was observed that nitrogen fertilization has a relationship of increase in the NDVI where 30 days after application in the case studied, the tillage went from an NDVI index of 0.2 to 0.7. 97 6. REFERENCES ANTUNES, R. L. S; ROSS, J. L. S. Interpretation of landscape physiognomies and their physiology from remote sensing in southern Brazil. Geoambiente On-line. Special Edition, Jataí - GO, n. 30, 2018. BARIANI, C. J. M. V.; BARIANI, N. M. V.; NETO, G. C. M. Spectral characteristics of irrigated rice cultivars during the development cycle in southern Brazil. In: Anais... IV Inovagri International Meeting, Fortaleza-CE, 2017. Available At http: //www. inovagri.org/library/iv-inovagrimeeting/res25608080Accessed on:15 nov. 2020. BARIANI, C.J.M.V.; FELICE, R. D.; VICTORIA, N. M. . Remote sensing of phenological stages of irrigated agricultural crops using NDVI. In: Anais.... 8° Salão Internacional de Ensino, Pesquisa e ExtensãoSIEPE, Uruguaiana-RS. v. 8. 2016. BORATTO, I. M. P.; GOMIDE, R. L. Aplicacao dos índices de vegetação NDVI, SAVI e IAF na caracterização da cobertura vegetativa da região Norte de Minas Gerais. In: Anais... 57th Brazilian Symposium on Remote Sensing. Foz do Iguaçu - PR. 2013. BREDEMEIER, C.; VARIANT, C.; ALMEIDA, D.; ROSA, A. T. Estimation of yield potential in wheat using active optical sensor for variable rate nitrogen fertilization. Rural Science, v.43, n.7, p. 11471154, 2013. BROLO, R. R.; REIS, A. A. dos.; ANTUNES, J. F. G.; LAMPARELLI, E. A. C. Use of time series of remote sensing images for the characterization of crop-livestock integration systems (CLI). In: XXVII Congresso virtual de iniciação cientifica da Unicamp. 2020. Available at: https: //www.prp.unicamp .br/inscricaocongresso/resumos/2020P16396A33745O67.pdf Accessed on:10nov. 2020. CRUSIOL, L. G. T.; CARVALHO, J. F. C.; SIBALDELLI, R. N. R.; PROCÓPIO, S. O.; NEUMAIER, N.; NEPOMUCENO, A. L.; MARCELINOGUIMARAES, F. C.; FARIAS, J. R. B. Influence of 98 the time of measurements on NDVI values. In: Anais.... VI Brazilian soybean congress. Cuiabá, 2012. p. 82. Available at www.cbsoja.com.br/anais/trabalhos/127 s323.pdfAccessed on: 20 Nov. 2020. DRUM, M. A.; VIAN, A. L.; BREDEMEIER, C.; TRENTIN, C.; SILVA, J. A.; GIORDANO, C. P. S; SANTOS, F. L. NDVI as a tool to adjust nitrogen fertilization in corn crop. In: Anais 62° Reunião Técnica Annual Corn Research Meeting and 45th Annual Technical Meeting of the Sorghum crop. p. 50. Available at: http ://www. abms. org.br/previous events/rtams 2017/annai s papers/1404.pd f Accessed on: 18 nov. 2020. DUTRA, V. A. B.; TAVARES, P. A.; RIBEIRO, H. M. C. Analysis of cloud condition on Sentinel-2 MSI and Landsat-8 OLI images of a public supply lake in Belém-Pará-Brazil. Rev. Ambient. Water, v.14, n.4, e2375, 2019. ENCINA, C. C. C.; MARQUES, M. R.; DIODATO, M. A.; MOTTA, J. S.; GODOI, R. de. F.; OLIVEIRA, J, R, S, de.; GAMARRA, R. M.; DALMAS, F. B.; PARANHOS FILHO, A. C. Analysis of the Vegetation Structure of Fragments of the Cerrado Biome Through Remote Sensing. Anuário do Instituto de Geociências -UFRJ. v. 41,n. 2, p. 585-597, 2018. EUROPEAN SPACE AGENCY (2015). Sentinel-2 User Handbook Document. Technical Report, issue 1 revision 2. 64 p. Available at: https://sentinel.esa.int/documents/247904/685211/Sentinel-2 User HandbookAccessed October 22, 2020. FRAMPTON, W. J.; DASH, J.; WATMOUGH, G.; MILTOM, E .J. Evaluating the capabilities of Sentinel-2 for quantitative estimation of biophysical variables in vegetation. ISPRS Journal of Photogrammetry and Remote Sensing, v. 82, p. p. 83-92, 2013. GENÚ, A. M.; DEMATTÊ, J. A. M.; FIORIO, P.R. Spectral analysis of soils from the Mogi-Guaçú (SP) Region. Semina: Agrarian Sciences, Londrina, v. 31, p.1235-1244, 2010. 99 HOMMA, K.; MAKI, M.; HIROOKA, Y. Development of rice simulation model for remote-sensing (SIMRIW-RS). Journal of Agricultural Meteorology, v. 73, p. p. 9-15, 2017. IBRAHIM, Y.Z., BALZTER, H., KADUK, J., TUCKER, C.J. Land Degradation Assessment Using Residual Trend Analysis of GIMMS NDVI3g, Soil Moisture and Rainfall in Sub-Saharan West Africa from 1982 to 2012. Remote Sensing. v.7, p. 5471-5494. 2015. JENSEN, J. R. Remote Sensing of the Environment. 2. ed. São José dos Campos, Bracket, 2011. JENSEN, J. R. Remote sensing of the environment: A perspective on terrestrial resources. São José dos Campos: Parentese. 2009. JUNGES, A. H.; FONTANA, D. C. Evaluation of the development of winter cereal crops in Rio Grande do Sul through temporal profiles of the normalized difference vegetation index. Ciência Rural, v.39, p.1 15, 2009. LEIVAS, J. F.; ANDRADE, R. G.; TEIXEIRA, A. H. de. C.; VICTORIA, D. de. C. TORRESAN, F. E.; BOLFE, E. L. Dinâmica espaço temporal do NDVI em mesorregiões produtoras de milho segunda safra. Embrapa Monitoramento por Satélite (Documentos, 103), 2013. 20 p. MA, L.; SCHMITT, M.; ZHU, X. Uncertainty Analysis of ObjectBased LandCover Classification Using Sentinel-2 Time-Series Data. Remote Sens., v. 12, n. 22, p.3798, 2020. MAGALHÃES, P. C.; DURÃES, F. O. M. Fisiologia da produção de milho. Sete Lagoas - MG: Embrapa Milho e Sorgo, 2006 (Technical Circular). Available at: https: //ainfo. cnptia.embrapa. br/di gital/bitstream/CNPMS/19620/1/Circ 7 6. pdf Accessed on:20 nov. 2020. MANABE, V.; MELO, M.; ROCHA, J. Framework for Mapping Integrated CropLivestock Systems in Mato Grosso, Brazil. Remote Sensing, v. 10, n. 9, p. 1322, 2018. 100