Mathematical Modeling For Predicting PM 2.5 And PM 10 Concentrations In Pokhara, Nepal

Abstract

Particulate matter, specifically PM 2.5 and PM 10 is a threat to the ecosystem; therefore, reducing their concentrations in our surroundings demands motivated steps. The goal of this work is to generate a mathematical formulation that would predict PM 2.5 and PM 10 concentrations in Pokhara. With the meteorological parametric values of Pokhara, the linear regression model (LRM) and the nonlinear regression model (NERM) have been generated. Error evaluation function analysis is applied to the developed models to find out the degree of their variation from the data that has been observed. Error assessment functions ARE, ERRSG, RMSE, MRPE, SEE, and EABS are used. Statistical tools such as R 2, K 2, and F-test are applied to examine the relevance of the generated models. When compared to NERM, the results suggested that LRM has a minor discrepancy from the observed data using the error evaluation function analysis. Additionally, the statistical analysis revealed that LRM correctly matched the observed data better than NERM, and as a result, LRM can be used to analyze the observed data. LRM is a good model for predicting the pollution of the location Pokhara, with reference to graphical comparisons as well. From this study, it is concluded that the LRM is a good choice for predicting the levels of PM 2.5 and PM 10 pollutants in study location.