hide
Free keywords:
-
Abstract:
We introduce "NitroNet", a deep learning model for the prediction of tropospheric NO2 profiles from satellite column measurements. NitroNet is a neural network, which was trained on synthetic NO2 profiles from the regional chemistry and transport model WRF-Chem, operated on a European domain for the month of May 2019. This WRF-Chem simulation was constrained by in-situ and satellite measurements, which were used to optimize important simulation parameters (e.g. the boundary layer scheme). The NitroNet model receives vertical NO2 column densities (VCDs) from the TROPOMI satellite instrument and ancillary variables (meteorology, emissions, etc.) as input, from which it reproduces NO2 concentration profiles. Training of the neural network is conducted on a filtered dataset, meaning that NO2 profiles with strong disagreement (> 20 %) to colocated TROPOMI column measurements are discarded.
We present a first evaluation of NitroNet on a variety of geographical domains (Europe, US west coast, India, and China) and different seasons. For this purpose, we validate the NO2 profiles predicted by NitroNet against monthly-mean satellite, in-situ, and MAX-DOAS measurements. The training data were previously validated against the same datasets. During summertime, NitroNet shows small biases and strong correlations to all three datasets (bias = +6.7 % and R = 0.95 for TROPOMI NO2 VCDs, bias = −10.5 % and R = 0.75 for AirBase surface concentrations). In the comparison to TROPOMI satellite data, NitroNet even shows significantly lower errors and stronger correlation than a direct comparison with WRF-Chem numerical results. During wintertime considerable low biases arise, because the summertime training data is not fully representative of all atmospheric wintertime characteristics (e.g. longer NO2 lifetimes). Nonetheless, the wintertime performance of NitroNet is surprisingly good, and comparable to that of classic RCT models. NitroNet can demonstrably be used outside the geographic domain of the training data with only slight performance reductions. What makes NitroNet unique compared to similar existing deep learning models is the inclusion of synthetic model data, which has important benefits: Due to the lack of NO2 profile measurements, empirical models are limited to the prediction of surface concentrations learned from in-situ measurements. NitroNet, however, can predict full tropospheric NO2 profiles. Furthermore, in-situ measurements of NO2 are known to suffer from biases, often larger than +20 %, due to cross sensitivities to photooxidants, which empirical models inevitably reproduce.
