CO2 gradients and variability in Mexico City from in situ measurements and simulations

Abstract

Emission inventories are the primary source of information regarding the spatial and temporal distribution of CO2 sources and sinks in urban environments. In most cities around the world, the impact of mitigation strategies is evaluated conventionally through an inventory methodology which -by nature- is subject to high uncertainties.  In this work, we report on the utility of deploying a small network of medium-cost, non-dispersive infrared (NDIR) CO2 microsensors in Mexico City, to directly quantify the gradients and variability of the molar fraction of atmospheric CO2. The measurements obtained were compared with simulations of a high-resolution 3D transport model (WRF-Chem), prescribed with city-scale, inventory-based CO2 emission maps. The focus of this presentation is on the description of the sensors construction, the characterization of each individual sensor’s performance evaluated against a reference instrument, and the ability of the network to represent the spatial and temporal variability of CO2 in a complex urban environment. Preliminary results of a multivariate calibration of the medium-cost microsensors using a Picarro G2401 as reference instrument, along with air temperature, relative humidity and pressure, results in RMSE ranging from 1 to 5 ppm of CO2. We discuss the potential that this network of sensors offer to evaluate whether the distribution of sources and sinks declared in an inventory can result in the variability of concentrations of CO2 measured in the atmosphere, and the possibility of being used as a tool to incorporate the contribution of out-of-city emissions or mobile sources emissions, currently not accounted for in the inventory.