Introdução de níveis tecnológicos no modelo Simanihot e sua aplicação na produtividade de mandioca em anos com influência do fenômeno ENOS

Abstract

The Simanihot model is a dynamic mathematical model that simulates the growth, development and productivity of cassava in the State of Rio Grande do Sul at the potential level and using water balance models. Even though it is available for use, this model needs to be constantly under study and improvement so that it can accurately describe the ecophysiological processes of the crop and better represent commercial crops, making this model an efficient tool for a wider range of cassava producers. Thus, the objectives of this thesis were (i) to improve the Simanihot cassava model so that it is considered the cultivation in technological levels under the growth, development and productivity of roots of the cultivar Vassourinha and (ii) to evaluate the influence of the ENSO phenomenon in the productivity of Vassourinha cassava cultivar in Rio Grande do Sul, at different technological levels using the Simanihot model. For calibration of the cultivar Vassourinha at high, medium and low technological level, experiments were carried out in Júlio de Castilhos in the crop year 2017/2018, and to estimate the parameters, the trial and error method was used, which minimizes the average square of error between the observed values and estimated values. The independent data for the evaluation of the model refer to the experiments carried out in Dilermando de Aguiar, Itaqui, Ibarama and Frederico Westphalen in the crop years 2017/2018 and 2018/2019. The performance of the Simanihot model was evaluated using the root statistics of the mean square of error (RQME), BIAS index, correlation coefficient (r), agreement index (dw) and efficiency of the model (EF). For the evaluation of the influence of the ENOS phenomenon on the productivity of the cassava cultivar Vassourinha, the Simanihot model was used, using the water balance model of Thornthwaite and Mather, which is coupled to the Simanihot. The model has been run for the last 58 years, for six planting dates within the period I recommend for the agroclimatic zoning of the crop, for 11 locations (Bagé, Cruz Alta, Encruzilhada do Sul, Passo Fundo, Pelotas, Porto Alegre, Santa Maria, Santana Livramento, São Luiz Gonzaga, Torres and Uruguaiana) and for the predominant soil mapping units in each location. The evaluations were carried out according to the average productivity curves found for the location and phase of the ENSO phenomenon. This work allowed the calibration of the Simanihot model to simulate the growth, development and productivity of tuberous roots and aerial part of cassava of the cultivar Vassourinha, at the high, medium and low technological levels. The occurrence of the ENOS phenomenon does not cause much difference in productivity (tuberous roots and aerial part) of the cultivar Vassourinha. The greatest interannual variability in simulated productivities occurs in neutral years. The delay in the planting season in 80 days determines a greater decrease in the yield of cassava tuberous roots, cultivar Vassourinha, when cultivated at a low technological level. The average dry mass productivity of shoots is higher in years with the occurrence of El Niño, for most of the studied locations.