Mathematical Modeling of the Biochemical Oxygen Demand (BOD) from Water Table in Sugar Cane Plantation Fertirrigated with Vinasse in Brazil

Abstract

The uses of the vinasse as fertilizer through fertirrigation for sugar cane plantations constitute a potential high source pollutant of soils and underground waters. Thus, it is necessary to monitor areas fertirrigated with vinasse to evaluate the quality conditions of those waters in order to conduct inferences and or diagnoses of the environmental quality. The decomposition of the vinasse organic matter, represented by the biochemical oxygen demand (BOD) is the most important parameter in terms of pollution indicators. The BOD laboratory analysis is not a direct measurement and can last some days to determine it. The adjustment of mathematical models can generate equations to estimate BOD as a function of independent variables that can be determined directly in short time. The data set for this study came from an area of 12 hectares planted with sugar cane, fertirrigated with vinasse. Thirty observation wells were installed with depths of 3 m, distributed in parallel transects 100 m from one to another. Eleven water quality parameters from three collect along the time (three, six and nine months) were analyzed. Simple and multiple linear and non-linear models, the Chapman-Richards and Silva-Bailey models were used to estimate the BOD. The Adjusted Fit Index in percent (AFI%) obtained presented values of adjustments superior to 95%. The Silva-Bailey model adjusted with two independent variables COD (chemical demand of oxygen) and T (age of the plantation) showed the higher AFI% = 98.78%. In the experimental conditions, the results showed that BOD could be estimated mathematically through its relationship with other chemical variables of simplified analytical procedures. The chemical oxygen demand (COD) stood out among the selected variables.