Modelagem do acúmulo de biomassa e de nitrogênio da cana-de-açúcar em distintos manejos hídricos e de adubação nitrogenada
Carregando...
Data
2021-02-25
Autores
Cosmo, Bruno Marcos Nunes
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
O Brasil destaca-se na produção agrícola devido à sua extensão territorial, diversidade edafoclimática e disponibilidade hídrica. Contudo, o setor encontra-se vulnerável aos fatores ambientais. Para os próximos anos, são descritas possíveis modificações climáticas que influenciaram a agricultura. Dentre as culturas possivelmente afetadas encontra-se a cana-de-açúcar, com potencial para incrementar sua produtividade. No Brasil a cultura produz cerca de 70 a 80 t ha-1 de colmos, porém, apresenta potencial teórico para quase 400 t ha-1, exigindo pesquisas para tal. Entretanto, tais pesquisas demandam tempo, mão-de-obra e outros recursos. Como meio de reduzir estas limitações, surgem os modelos de predição de safras. Neste sentido, a presente pesquisa teve como objetivo avaliar o uso do modelo APSIM-Sugarcane para simulações do acúmulo de biomassa e nitrogênio na cana-de-açúcar em diferentes condições de manejo. O estudo foi conduzido na Unidade de Pesquisa e Desenvolvimento “Hélio de Morais”, pertencente ao Instituto Agronômico de Campinas, Jaú-SP. Foram utilizados dados reais do experimento: “Disponibilidade hídrica e utilização do nitrogênio em cana-de-açúcar irrigada por gotejamento subsuperficial” – Dissertação. Para a calibração do modelo, foram gerados bancos de dados agrometeorológicos (2005-2018) para Jaú-SP, além da inserção de dados de solo, manejo e análises de desenvolvimento da cana-de-açúcar. Na validação foram empregados 8 parâmetros estatísticos: erro médio, erro médio absoluto, raiz quadrada do erro médio, coeficientes de correlação (r) e determinação (R²) e índices de eficiência de modelagem (E), concordância (d) e confiança (c), sendo a premissa para utilização do modelo a obtenção de “E” > 0,5 e “c” > 0,66. Após a validação, os modelos gerados foram corrigidos com a utilização de regressões lineares que correlacionaram os resultados de saída do APSIM-Sugarcane com os resultados reais. Os resultados simulados resultantes foram reavaliados após as correções. Com os modelos ajustados e corrigidos foram realizadas projeções de acúmulo de matéria seca e nitrogênio na parte aérea da cana-de-açúcar e da produtividade de colmos e açúcar para o Centro-Oeste Paulista considerando o ambiente irrigado e de sequeiro sem aplicação de nitrogênio (N) e com aplicação de 140 kg ha-1 de N. Estas projeções foram realizadas considerando cenários de mudanças climáticas que envolveram a elevação isolada na temperatura em 1,5 ou 3,5ºC, e combinada com a elevação ou redução de 20% na precipitação. Os resultados da validação após correção das simulações do APSIM foram superiores aos parâmetros de avaliação para acúmulo de matéria seca e nitrogênio em todos os tratamentos, e para produtividade de colmos e açúcar em ambiente irrigado, porém, as estimativas de produtividade de colmos e açúcar em ambiente de sequeiro não foram validadas. Os cenários de projeção de mudanças climáticas indicam potencial elevação no acúmulo de matéria seca e nitrogênio na parte aérea da cana-de-açúcar, bem como de produtividade, desde que as necessidades hídricas provenientes da elevação de temperatura sejam atendidas. O modelo APSIM-Sugarcane, após as correções pelas equações lineares possibilitou a realização de simulações assertivas, tornando-se uma ferramenta aplicável para projeções de manejo e planejamento agrícola.
Brazil stands out in agricultural production due to its territorial extension, edaphoclimatic diversity and water availability. However, the sector is vulnerable to environmental factors. For the next few years, possible climate changes that influenced agriculture are described. Among the crops possibly affected is sugar cane, with the potential to increase its productivity. In Brazil, the crop produces about 70 to 80 t ha-1 of stalks, however, it has theoretical potential for almost 400 t ha-1, requiring research to do so. However, such surveys take time, labor and other resources. As a means of reducing these limitations, crop prediction models appear. In this sense, the present research aimed to evaluate the use of the APSIM-Sugarcane model for simulations of the accumulation of biomass and nitrogen in sugarcane under different management conditions. The study was conducted at the Research and Development Unit “Hélio de Morais”, belonging to the Agronomic Institute of Campinas, Jaú-SP. Actual data from the experiment were used: “Water availability and nitrogen use in sugarcane irrigated by subsurface drip” - Dissertation. For the calibration of the model, agrometeorological databases (2005-2018) were generated for Jaú-SP, in addition to the insertion of data on soil, management and analysis of sugarcane development. In the validation, 8 statistical parameters were used: mean error, mean absolute error, square root of the mean error, correlation coefficients (r) and determination (R²) and modeling efficiency indexes (E), agreement (d) and confidence (c), and the premise for using the model is to obtain “E”> 0.5 and “c”> 0.66. After validation, the generated models were corrected using linear regressions that correlated the output results of APSIM-Sugarcane with the actual results. The resulting simulated results were reevaluated after the corrections. With the adjusted and corrected models, projections were made of accumulation of dry matter and nitrogen in the aerial part of the sugarcane and of the productivity of stalks and sugar for the Midwest of São Paulo considering the irrigated and dryland environment without nitrogen application (N) and application of 140 kg ha-1 of N. These projections were made considering climate change scenarios that involved an isolated rise in temperature of 1.5 or 3.5ºC, and combined with an increase or reduction of 20% in precipitation. The results of the validation after correction of the APSIM simulations were superior to the evaluation parameters for accumulation of dry matter and nitrogen in all treatments, and for stalk and sugar productivity in an irrigated environment, however, the stalk and sugar productivity estimates in rainfed environment have not been validated. The climate change projection scenarios indicate a potential increase in the accumulation of dry matter and nitrogen in the aerial part of the sugarcane, as well as in productivity, as long as the water needs arising from the temperature rise are met. The APSIM-Sugarcane model, after corrections by linear equations, made it possible to carry out assertive simulations, becoming an applicable tool for agricultural management and planning projections.
Brazil stands out in agricultural production due to its territorial extension, edaphoclimatic diversity and water availability. However, the sector is vulnerable to environmental factors. For the next few years, possible climate changes that influenced agriculture are described. Among the crops possibly affected is sugar cane, with the potential to increase its productivity. In Brazil, the crop produces about 70 to 80 t ha-1 of stalks, however, it has theoretical potential for almost 400 t ha-1, requiring research to do so. However, such surveys take time, labor and other resources. As a means of reducing these limitations, crop prediction models appear. In this sense, the present research aimed to evaluate the use of the APSIM-Sugarcane model for simulations of the accumulation of biomass and nitrogen in sugarcane under different management conditions. The study was conducted at the Research and Development Unit “Hélio de Morais”, belonging to the Agronomic Institute of Campinas, Jaú-SP. Actual data from the experiment were used: “Water availability and nitrogen use in sugarcane irrigated by subsurface drip” - Dissertation. For the calibration of the model, agrometeorological databases (2005-2018) were generated for Jaú-SP, in addition to the insertion of data on soil, management and analysis of sugarcane development. In the validation, 8 statistical parameters were used: mean error, mean absolute error, square root of the mean error, correlation coefficients (r) and determination (R²) and modeling efficiency indexes (E), agreement (d) and confidence (c), and the premise for using the model is to obtain “E”> 0.5 and “c”> 0.66. After validation, the generated models were corrected using linear regressions that correlated the output results of APSIM-Sugarcane with the actual results. The resulting simulated results were reevaluated after the corrections. With the adjusted and corrected models, projections were made of accumulation of dry matter and nitrogen in the aerial part of the sugarcane and of the productivity of stalks and sugar for the Midwest of São Paulo considering the irrigated and dryland environment without nitrogen application (N) and application of 140 kg ha-1 of N. These projections were made considering climate change scenarios that involved an isolated rise in temperature of 1.5 or 3.5ºC, and combined with an increase or reduction of 20% in precipitation. The results of the validation after correction of the APSIM simulations were superior to the evaluation parameters for accumulation of dry matter and nitrogen in all treatments, and for stalk and sugar productivity in an irrigated environment, however, the stalk and sugar productivity estimates in rainfed environment have not been validated. The climate change projection scenarios indicate a potential increase in the accumulation of dry matter and nitrogen in the aerial part of the sugarcane, as well as in productivity, as long as the water needs arising from the temperature rise are met. The APSIM-Sugarcane model, after corrections by linear equations, made it possible to carry out assertive simulations, becoming an applicable tool for agricultural management and planning projections.
Descrição
Palavras-chave
APSIM-Sugarcane, Mudanças climáticas, Modelo de produtividade, Climate changes, Productivity model