Efeitos da instabilidade pluviométrica sobre a previsão da produção de lavouras de sequeiro em áreas sujeitas à desertificação (ASD) no semiárido do estado do Ceará: casos de Irauçuba e Tauá

Abstract

The degradation of natural resources with environmental impacts is not a recent subject. From earliest times, records the hist over time. This assault on nature causes severe damage to natural resources. Among others, the phenomenon of "desertification". In the Brazilian semiarid, this phenomenon is quite evident in some areas. Another peculiar feature semi-arid ecosystem is the rainfall instability that has a direct relationship in agricultural production. This study sought to examine how this binomial formed by rainfall instability and desertification influences the production of bean crops, manioc and corn grown on a "dry land" in the municipalities of Irauçuba and Tauá. These municipalities are situated in Areas Subject Desertification (ASD) in Ceará State semi arid region. The specific objectives of the research are: A- Draw models which capture behavior of annual rainfall in the municipalities of Irauçuba and Taua, compared to what happens in the state of Ceará in the period extending from 1974 to 2017; B - Evaluate the instability / stability of the variables as well as the evolution of the variables harvested area, yield per hectare, average price of beans, cassava and corn in each of the rainfall regimes identified in the survey in the municipalities of Irauçuba and Taua in the same period; C - Draw probabilistic models previews of variables that are defining the production of upland farmer in ASD of Ceará Stet semiarid region in the period under investigation; D - To evaluate the effect of rainfall on these forecasts. The research uses data collected by the FUNCEME and the IBGE, and covers the period from 1974 to 2017. The methodological procedures used are segmented according to the especificam objectives of the research. Thus, we use the coefficients of variation (CV) to measure the instability / stability of different types of weather that the search feature to municipalities selected in the study. CV will also be used as measures of instability / stability of productivity of the land, harvested areas and average prices of crops, evaluated within each of the rainfall regimes that will be identified in the survey. To make predictions of the endogenous variables the work used ARIMA models developed by Box and Jenkins (1976). To assess the impact of the rains on the estimates of endogenous variables the research uses regression model that evaluates the rains impacting the noises generated in the forecast or directly impacting forecasts. The results confirm rainfall instabilities in both municipalities. Show that there are statistical differences between the defined rainfall regimes for each municipality. The research confirms the central hypothesis that rainfall variations impact the forecasting of endogenous variables in the production of beans, cassava and maize in Irauçuba and Taua, especially on land productivity.