Automatização da predição de variáveis micrometeorológicas utilizando a Teoria da Complexidade

• Autor(a) principal: Annunciação, Ana Claudia da Silva
• Data de Publicação: 2017
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da UFMT
• Texto Completo: http://ri.ufmt.br/handle/1/2943
• topic: Pantanal; Teoria da complexidade; Sistema dinâmico; CNPQ::CIENCIAS EXATAS E DA TERRA::FISICA; Pantanal; Complexity theory; Dynamic system

The Theory of Complexity offers an organizational analysis of open systems, resulting from the interaction of agents that adapt to the environmental contingencies. Because of this, its use in predicting information is quite propitious given the relational, integrated and systemic character of science, technology, economics, health, and especially climate phenomena, because in designing the future we are, therefore, entangled in complex Temporal relationships, which lead us to results that are between predicted and unforeseen. The objective of this study was to automate the processes, through a software with micrometeorological variables of balance of temperature, temperature and humidity of the air (with data of 15 in 15 minutes), measured above the canopy, in a flooded forest in the Mato Grosso Pantanal , Located in the Private Natural Heritage Reserve (RPPN SESC) - Barão de Melgaço - MT. The automation of these processes was through an algorithm in the Python programming language, using the TISEAN (Nonlinear Time Series Analysis) package, which calculates the data series, in predetermined periods in the input parameters. This automation provides great speed in data execution so that decisions are made on time with speed and security. The results showed that in the dry period the average of the coefficient of determination is higher than in the rainy season, attesting that the predicted time series are very close to the measured time series. However, there was a tendency in the approximations of the variables, with the balance of radiation closer to the forecast, then the temperature and finally the relative humidity of the air. In general, the influence of the past is preponderant to make predictions in the future, because there is a feedback effect, these dynamics of the phenomena allowed to make reliable forecasts in the short term of up to seven days.