Lucas Reymond

This project focuses on estimating Black Carbon (BC) concentrations in urban areas using Bayesian computations. Since BC is a harmful pollutant, precise monitoring is crucial. While stationary sensors provide accurate measurements, mobile sensors are more flexible but introduce noise into the data. The goal is to combine these two sources of data to improve BC estimation.

A Bayesian approach is used, assuming a Weibull prior distribution for BC concentrations. The project implements a Metropolis-Hastings Markov Chain Monte Carlo algorithm to generate samples from the posterior distribution of BC given noisy observations. The algorithm is tested with different observed values of BC concentration, and the empirical distributions from MCMC are compared with the true posterior densities.